Geology and Paleontology of the Lee Creek Mine. North Carolina, I CLAYTON E. RAY EDITOR NUMBER 53 SMITHSONIAN CONTBIBUTIONS TO PALEOBIOLOGY •x SERIES PUBLICATIONS OF THE SMITHSONIAN INSTITUTION Emphasis upon publication as a means of "diffusing knowledge" was expressed by the first Secretary of the Smithsonian. In his formal plan for the Institution, Joseph Henry outlined a program that included the following statement: "It is proposed to publish a series of reports, giving an account of the new discoveries in science, and of the changes made from year to year in all branches of knowledge." This theme of basic research has been adhered to through the years by thousands of titles issued in series publications under the Smithsonian imprint, commencing with Smithsoniar) Cor}tributior)s to Knowledge in 1848 and continuing with the following active series: Smithsonian Contributions to Anthropology Smithsonian Contributions to Astrophysics Smithsonian Contributions to Botany Smithsonian Contributions to the Earth Sciences Smithsonian Contributions to the Marine Sciences Smithsonian Contributions to Paleobiology Smithsonian Contributions to Zoology Smithsonian Studies in Air and Space Smithsonian Studies in History and Technology In these series, the Institution publishes small papers and full-scale monographs that report the research and collections of its various museums and bureaux or of professional colleagues in the world of science and scholarship. The publications are distributed by mailing lists to libraries, universities, and similar institutions throughout the world. Papers or monographs submitted for series publication are received by the Smithsonian Institution Press, subject to its own review for format and style, only through departments of the various Smithsonian museums or bureaux, where the manuscripts are given substantive review. Press requirements for manuscript and art preparation are outlined on the inside back cover. S. Dillon Ripley Secretary Smithsonian Institution SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY • NUMBER 53 Geology and Paleontology of the Lee Creek Mine, North Carolina, I Clayton E. Ray EDITOR SMITHSONIAN INSTITUTION PRESS City of Washington 1983 ABSTRACT Ray, Clayton E., editor. Geology and Paleontology of the Lee Creek Mine, North Carolina, I. Smithsonian Contributions to Paleobiology, number 53, 529 pages, frontispiece, 95 figures, 101 plates, 8 tables, 1983.—This volume of papers on the geology and paleontology of the Lee Creek Mine is the first of three to be dedicated to the late Remington Kellogg, who initiated Smithsonian studies of the mine. It includes the first 14 papers, as well as a biography of Remington Kellogg by Frank C. Whitmore, Jr., and a prologue by Clayton E. Ray. This study places the Lee Creek Mine in the larger context of the history of Neogene geology and paleontology of the middle Atlantic Coastal Plain. Jack H. McLellan outlines the development and operation of Texasgulfs phosphate mine and manufacturing plant at Lee Creek, particu- larly as they relate to geological and paleontological studies. Thomas G. Gibson describes the regional patterns of Miocene-Pleistocene deposition in the Salisbury and Albemarle embay- ments of the central Atlantic Coastal Plain. On the basis of cluster analysis of 16 samples, including 149 taxa of ostracodes from fossiiiferous beds above the Pungo River Formation, Joseph E. Hazel determines that the Yorktown Formation at the Lee Creek Mine is early Pliocene in age and the Croatan Formation spans the Plio-Pleistocene boundary. Among the ostracodes, 2 genera, 31 species, and one subspecies, are diagnosed as new. Walter H. Wheeler, Raymond B. Daniels, and Erling E. Gamble survey the post-Yorktown development in the region of the Neuse-Tar-Pamlico rivers. Primarily on the basis of auger holes, they begin with the Aurora paleoscarp marking the top of the Yorktown Formation, on which the organic-rich Small sequence (Croatan or James City Formation) was deposited, followed unconformably by the Pamlico morphostratigraphic unit; the inner edge of the Pamlico msu is associated with the Minnesott Ridge. H. Allen Curran and Patricia L. Parker divide the "Upper Shell" unit at the mine into three bivalve assemblage zones, probably formed through mass mortality in a series of local catastrophic events. Edward S. Belt, Robert W. Frey, and John S. Welch interpret Pleistocene deposition at the mine on the basis of biogenic and physical sedimentary structures, enabling them to recognize five major unconformities and four depositional sequences, indica- tive of a progradational shoreline under tectonically stable conditions. Their fourth depositional cycle includes a freshwater peat member thought to be of Sangamon interglacial age, on the basis of Donald R. Whitehead's pollen analysis. This analysis reveals high percentages of sedge and grass pollens, an absence of boreal indicators, tree pollen frequencies similar to those of interglacial deposits to the north and south, and general similarity of the fossil pollen spectrum to modern pollen assemblages of eastern North Carolina. Francis M. Hueber identifies the gymnospermous genera Pinus,Juniperus, and Taxodium, and tentatively the angiospermous genus Gleditsia, among the quartz-permineralized woods from the lower part of the Yorktown Formation at the mine; he also discusses the resin-like specimens, which are of unknown biological source and for which the stratigraphic source (Yorktown Formation, above the source of the woods) is known for only one specimen. William H. Abbott and John J. Ernissee report one silicoflagellate and two diatom assemblages (equivalent to Blow's zones N9 and Nil) in a diatomaceous clay of the Pungo River Formation from two cores in Beaufort County; one new species of diatom is described. On the basis of 30 species of planktonic Foraminifera and a few radiometric dates, Thomas G. Gibson assigns ages from latest Oligocene through early Pleis- tocene to 10 stratigraphic units in the central Atlantic Coastal Plain; he describes 37 species and subspecies of benthic Foraminifera, of which 10 species and 2 subspecies are new. Scott W. Snyder, Lucy L. Mauger, and W.H. Akers assign an age of late-early to early-late Pliocene for a 15-meter section of the Yorktown Formation at the mine, based on 29 taxa of planktonic Foraminifera. Druid Wilson describes as a new genus and species of barnacle a puzzling fossil from inside the shell of the bivalve Mercenaria from the Croatan Formation. Porter M. Kier reports one species of echinoid from the Pungo River Formation, three from the Yorktown Formation, of which one is new, and two from the Croatan Formation. John E. Fitch and Robert J. Lavenberg record 45 taxa of teleost otoliths from the Yorktown Formation, repre- senting 27 genera, of which 22 are new to the Pliocene of North America, and 6 are first fossil records. OFFICIAL PUBLICATION DATE is handstamped in a limited number of initial copies and is recorded in the Institution's annual report, Smithsonian Year. SERIES COVER DESIGN: The trilobite Phacops rana Green. Library of Congress Cataloging in Publication Data Main entry under title: Geology and Paleontology of the Lee Creek Mine, North Carolina. (Smithsonian contributions to paleobiology ; no. 53- ) Includes bibliographies. 1. Geology, Stratigraphic—Tertiary—Addresses, essays, lectures. 2. Geology—Stratigraphic Pleistocene—Addresses, essays, lectures. 3. Geology—North Carolina—Addresses, essays, lectures. 4. Paleontology—North Carolina—Addresses, essays, lectures. I. Ray, Clayton Edward. IL Series: Smithsonian contributions to paleobiology ; no. 53, etc. Q.E701.S56 no. 53, etc. 560s [551.7'8'09756] 82-600265 [QE691] Contents Page PROLOGUE, by Clayton E. Ray 1 REMINGTON KELLOGG, 1892-1969, by Frank C. Whitmore, Jr 15 PHOSPHATE MINING AT THE LEE CREEK MINE, by Jack H. McLellan 25 STRATIGRAPHY OF MIOCENE THROUGH LOW^ER PLEISTOCENE STRATA OF THE UNITED STATES CENTRAL ATLANTIC COASTAL PLAIN, by Thomas G. Gibson 35 AGE AND CORRELATION OF THE YORKTOWN (PLIOCENE) AND CROATAN (PLIOCENE AND PLEISTOCENE) FORMATIONS AT THE LEE CREEK MINE, by Joseph E. Hazel 81 THE POST-YORKTOVS^N STRATIGRAPHY AND GEOMORPHOLOGY OF THE NEUSE-PAMLICO AREA, EASTERN NORTH CAROLINA, by Walter H. Wheeler, Raymond B. Daniels, and Erling E. Gamble 201 OBSERVATIONS ON THE PALEOECOLOGY AND FORMATION OF THE "UPPER SHELL" UNIT, LEE CREEK MINE, by H. Allen Curran and Patricia L. Parker 219 PLEISTOCENE COASTAL MARINE AND ESTUARINE SEQUENCES, LEE CREEK MINE, by Edward S. Belt, Robert W. Frey, and John S. Welch ... 229 POLLEN ANALYSIS OF THE PEAT MEMBER FROM THE LEE CREEK MINE, by Donald R. Whitehead 265 FOSSIL WOODS AND RESIN-LIKE SUBSTANCES FROM THE LEE CREEK MINE, by Francis M. Hueber 269 BIOSTRATIGRAPHY AND PALEOECOLOGY OF A DIATOMACEOUS CLAY UNIT IN THE MIOCENE PUNGO RIVER FORMATION OF BEAUFORT COUNTY, NORTH CAROLINA, by William H. Abbott and John J. Ernissee 287 KEY FORAMINIFERA FROM UPPER OLIGOCENE TO LOWER PLEISTOCENE STRATA OF THE CENTRAL ATLANTIC COASTAL PLAIN, by Thomas G. Gibson 355 PLANKTONIC FORAMINIFERA AND BIOSTRATIGRAPHY OF THE YORKTOWN FORMATION, LEE CREEK MINE, by Scott W. Snyder, Lucy L. Mauger, and W.H. Akers 455 THE LEE CREEK ENIGMA, Mclellama aenigma, A NEW TAXON IN FOSSIL CIRRHIPEDIA, by Druid Wilson 483 UPPER CENOZOIC ECHINOIDS FROM THE LEE CREEK MINE, by Porter M. Kier 499 TELEOST FISH OTOLITHS FROM LEE CREEK MINE, AURORA, NORTH CAR- OLINA (YORKTOWN FORMATION: PLIOCENE), by John E. Fitch and Robert J. Lavenberg 509 111 Dedicated to Remington Kellogg 1892-1969 FRONTISPIECE.—Annotated false color composite image by NASA satellite (LANDSAT-3, image E- 30116-15030), from altitude of approximately 915 km (568 mi), of part of southeastern Virginia and eastern North Carolina, 29 June 1978. Scale 1:1,000,000. Courtesy of United Slates Department of the Interior, U.S. Geological Survey. Geology and Paleontology of the Lee Creek Mine, North Carolina, I Prologue Clayton E. Ray What's past is prologue; what to come, In yours and my discharge. -SHAKESPEARE The Tempest Act II, Scene 1 John Brickell, M.D., lived and practiced med- icine some 250 years ago in Edenton, North Car- olina, on the outer Coastal Plain less than 50 miles (80.5 km) north of the Lee Creek (phos- phate) Mine. With apologies for any license taken with his intent, with allowances for the accumu- lated knowledge and altered perspective of our age of specialization, and the concomitant reduc- tion in our scope to one aspect of the natural history of one place in North Carolina, the follow- ing excerpt from his preface to The Natural History of North-Carolina (Brickell, 1737:iv-vi) seems ad- mirably apropos to introduce the "Geology and Paleontology of the Lee Creek Mine, North Car- olina." The Writings of many Learned Men may be seen on this Head, who after having searched all the Records of Antiquity, shew much Erudiction, but nothing of certainty, concerning the Antient Affairs of America. I know the Memory of a Deluge is preserved amongst these people, but whether it is to be understood of the universal Flood, or the Inundation of some particular Provinces, I leave it to others to discourse Clayton E. Ray, Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. upon, for I am willing to lay aside all manner of Conjectures of this Nature, having enough of Truth to treat of. But waveing these Discourses, we here present the World with a Natural-History o/North-Carolina, it being a compendious Col- lection, of most things yet known in that part of the World; wherein I have laid down every thing with Impartiality and Truth, in the most plain and easie Terms, which indeed is the Duty of every Writer, and preferable to a more eloquent Stile, accompanied with many Falsities. I have therefore endeavour'd m the following Sheets to give as faithful and exact Account o/Carolina, as discoveries yet made will Authorize .... But not to amuse the Reader any longer with Encomiums on Carolina, / refer them to my Description of that Country, and it's Inhabitants, which they will find in the following Natural History, in which I have been very exact; and for Methods sake, have ranged each Species o/Animals, Vegetables, etc. under distinct and proper Heads. A Collection of the Natural Curiosities of this spacious part of the World, will, I hope, not only give Satisfaction and Pleasure to each Reader, but likewise Profit, to all that are inclined to live in those Parts. If these my Endeavours meet with this good success, I am thoroughly satisfied, having nothing more at Heart than to be in any Degree serviseable to the Publick; this being the principal Motive that induced me to undertake any Work of this Nature, (the Task being not only Laborious but Difficult) and not out of any Praise I expected from It. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY To conclude. Whatever Defects may be found in this Undertaking, we hope in time they will be supplied by the Labours and Industry of such as shall come after . . . and that their laudable Attempts may meet with just Encouragement, shall be my constant Wish and Desire. [Italics in original.] Although not of primary concern here, and not detracting materially from his eloquent preface, Brickell's apparently wholesale plagiarism must be acknowledged whenever his book is men- tioned. The subject has been reviewed recently by Simpson and Simpson (1981). The project of which the present publication is the culmination may fairly be said to have had its inception early in 1967 when Remington Kel- logg received "a small collection of vertebrate fossils from the Lee Creek Mine" from Jack H. McLellan (letter of 2 March 1967). With Kel- logg's encouragement, specimens continued to trickle into the National Museum of Natural History, where he began work on the fossil whales and enlisted others into looking at materials per- taining to their specialties. From the beginning through 1970, I had accumulated only a handful of seal bones (my favorite fossils) from the mine. The locality appeared to me to fit the all-too- familiar pattern for the Coastal Plain, that of yielding pinniped remains too sporadically to justify the expense of collecting trips. However, Jack McLellan visited the Museum on 10 Decem- ber 1970, handed me a monachine seal temporal bone, viewed our meager collection, and assured me that it could be augmented readily through more vigorous pursuit on our part at the mine. With his encouragement and upon cessation of other duties, I visited the mine for the first time in August 1971. Frank Whitmore, Robert Purdy (Department of Paleobiology), and I joined McLellan for two days of collecting, the results of which were very satisfactory. For my part, the collection included enough pinniped fragments to persuade me that intensive effort well might yield collections unprecedented in variety, quan- tity, and novelty for the Atlantic Coastal Plain. On the same trip and equally important in ret- rospect was our visit with Peter J. Harmatuk of Bridgeton, North Carolina, a widely known, avid, and able fossil collector, who at this writing has contributed more vertebrate fossils to the Na- tional Collections than any other person. The pattern for subsequent work was set and continues to the present. I have now made more than 50 visits to eastern North Carolina, typically of a few days each, in company with small groups from Washington. These groups generally consist of colleagues from the Smithsonian Institution and the U.S. Geological Survey, local students, and volunteers; but they have also included col- leagues from as far away as England, Romania, and Japan. On a typical trip we rendezvous at the mine with faculty and students from inter- ested schools, Texasgulf employees, amateurs from the area, and always with Peter Harmatuk. A part of every trip is devoted to other related activities, including prospecting other localities, examining institutional and private collections, assisting with the Aurora Fossil Museum, and participating in "fossil fairs." The continuing ad- dition of specimens to the National Collections from non-Smithsonian sources (resulting from the contacts established on these trips) has been re- sponsible, far more than our own collecting, for turning the initial trickle into a torrent; one product of this is that the vertebrate fauna of the Yorktown Formation is now one of the world's largest. As this prospect quickened my interest, I began to educate myself in the geology and paleontology of the region and of the mine. I soon discovered a remarkable reservoir of knowledge and contin- uing interest on the part of colleagues at the Smithsonian and U.S. Geological Survey, includ- ing Thomas Gibson, who had published the basic paper on the geology of the mine in 1967, Blake Blackwelder, Joseph Hazel, Porter Kier, Lauck Ward, Alexander Wetmore, Frank Whitmore, and Druid Wilson, all of whose expertise ex- ceeded, and whose interest antedated, my own. The combination of that resource, the ease with which other specialists were recruited across a broad range of relevant topics, and the burgeon- ing collections, culminated on 23 March 1972 with my proposing this publication project. During much the same period, I had been casting about independently for a suitable means NUMBER 53 of honoring Remington Kellogg in published form. It had seemed to me that such a tribute was long overdue from the institution that he had .served since 1928 as curator and administrator, and as the giant of his era in marine mammalogy. A typical festschrift, consisting of papers united primarily by the relationship between the con- tributors and the honoree, seemed less attractive than something on a unified topic. The fact that Kellogg had chosen at retirement in 1962 to devote his energies to curation, research, and publication on fossil marine mammals of the Neogene of the Atlantic Coastal Plain, that he had been directly responsible for initiating Smith- sonian research on the Lee Creek Mine, and that marine mammals are among the most conspicu- ous components of its fossil assemblage, all com- bined to suggest that this publication would con- stitute an appropriate and substantial tribute to his paleontological career in general and to his seminal role for Lee Creek in particular. Thus, this publication is dedicated respectfully to the memory of Remington Kellogg, and these vol- umes therefore begin with his biography and will conclude with a list and index of his publications. Between these "end pieces" an attempt has been made to be as comprehensive as possible with regard to the geology and paleontology of Lee Creek Mine. For vertebrates it was feasible to be essentially exhaustive, and the third volume will be made up exclusively of those contribu- tions, with the exception of a concluding chapter, comprising what is today popularly called an "overview," by Gibson and Whitmore, plus the appendices devoted to Kellogg. The first volume begins with a chapter about the mine and the mining itself, intended primarily to place our studies in context by revealing the opportunities created by the existence of the mine and at the same time the limitations imposed on collecting by the exigencies of mining. This is followed by chapters on the geology, concentrating on the regional setting, age, correlation, stratigraphy, paleoecology, and genesis of the deposits. Ob- viously, we have emphasized the aspects of geol- ogy most intimately related to paleontology and neglected or excluded many other potentially interesting aspects. These chapters are followed by three paleobotanical contributions, limited in number and scope by the availability of materi- als. The balance of the volume is devoted to invertebrates other than mollusks, plus a chapter on fish otoliths. The second volume is devoted exclusively to the mollusks, reflecting their abun- dance and importance. We have emphasized groups of special biostratigraphic value or special prominence or novelty at the mine, but have of necessity been governed also by the availability of appropriate specialists. The invertebrate fauna is so rich that, for practical purposes, the possi- bilities are unlimited. Obvious gaps in our cov- erage include the lack of comprehensive chapters on bryozoans and barnacles. These remain for future studies. Although it is hoped that the whole publication will be found greater than the sum of its parts, each chapter is largely self-contained to the extent that its contents will be intelligible without ref- erence to the whole, so that special interests can be satisfied through author's separates. This ob- jective inevitably has resulted in some repetition, especially in the citation of literature. Harking back to Brickell's expression of the writer's duty to lay down ""every thing with Impar- tiality and Truth, in the most plain and easie Terms," this goal often may be approached best through pictures, and I have accordingly urged contribu- tors to illustrate their topics generously. By this means I hope that these volumes will have been made more useful not only to specialists, but also to the host of serious, dedicated amateurs and students who are starved for reliable information, but who may not have command of the jargon that too often obscures the intrinsic interest of our subject. Having outlined what we intended to achieve through publication of these volumes and how we have attempted to do it, there remains the question of why. Why North Carolina? Why the Lee Creek Mine? Why the Yorktown Formation? Perhaps a thumbnail sketch of the history of development of our knowledge of the Neogene of the middle Atlantic Coastal Plain will aid in answering these questions. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Historical Perspective The middle Atlantic Coastal Plain is especially significant to the early history of the sciences of North American geology and paleontology. East- ern North Carolina is particularly so, as it was the locale of the first efforts at permanent settle- ment by the English. The words and disciplines of geology and paleontology had yet to be de- fined, and the origin and significance of fossils would be debated for many decades (Ewan and Ewan, 1970:309-312; White, 1953a: 137-138). However, explorers and colonists from the begin- ning had practical incentive to notice such mat- ters because they were pertinent to their survival. Thomas Harlot (or Harriot), whom Sir Walter Raleigh sent to the Roanoke Island colony in 1585, recognized the essential distinction between the Coastal Plain and Piedmont, and probably also the nature of the fossil shell beds of the Coastal Plain (Hariot, 1590; White, 1952b: 120; 1953a: 136). The shell beds were of great impor- tance as a source of lime for mortar (Bailey, 1938: 2). John Smith also recognized the distinction between the Piedmont and Coastal Plain, as evi- denced by his map of 1612 (White, 1953b: 125, 131), and William Strachey, first secretary of the colony at Jamestown, clearly characterized the fall line and demonstrated a surprisingly modern concept of the dynamics of the Coastal Plain (in Major, 1849:32): All the low land of South and North Virginia is conjec- tured to have bene naturally gayned out of the sea; for the sea, through his impetuous and vast revolution (who knowes not), savinge upon every coast, in some places wyns, and in other places looseth; and we find within the shoares of our rivers, whole bancks of oysters and scallopps, which lye unopened and thick together, as if there had bene their naturali bedd before the sea left them; likewise, the fashion of the earth is in smale rising mounts, which may well be supposed that the violence of the wynd hath cawsed, by dryving the light sand togither .... He went on to comment upon the thin top soil and the lack of indurated rock in the subsurface, which he attributed to "want of tyme."^ ' Although Strachey's writings were not published until much later, they were widely circulated in literary circles of The first explicit notice by Europeans of ver- tebrate fossils of the Coastal Plain Neogene was the entry of 3 August 1636 in Winthrop's journal (inHosmer, 1908:185-186): Samuel Maverick, who had been in Virginia near twelve months, now returned .... It is very strange, what was related by him and many others, that, above sixty miles [97 km] up James River, they dig nowhere but they find the ground full of oyster shells, and fishes' bones, etc.; yea, he affirmed that he saw the bone of a whale taken out of the earth (where they digged for a. well) eighteen feet [5.5 m] deep. Simpson (1942:134; 1943:27) was curiously reluc- tant to accept these as bona fide fossils; but in fact from slightly above Hampton Roads to Rich- mond (well over 60 miles [97 km] up the James River) the Neogene strata are superlatively fos- siiiferous, most conspicuously in remains of whales (Baum and Wheeler, 1977) and mollusks (Blackwelder and Ward, 1976; Gardner, 1948). Thus, there is every reason to believe, and no reason to doubt, that Maverick saw fossils, most likely of Miocene, or, if the distance was exagger- ated, at latest, early Pliocene (Yorktown), age. Two scientist-clergymen lived on the Virginia Coastal Plain and made perceptive observations on its geology and fossils in the latter part of the seventeeth century, John Clayton from 1683 to 1686, and John Banister from 1678 to 1692. Clayton later lived in England and Ireland until 1725, and published rather extensively; Banister was shot (accidentally?) while exploring along the Roanoke River in 1692, and his enormous influ- ence upon natural history in general and that of Virginia in particular has until recently not been widely appreciated. The analyzed and annotated works of each are now readily available in book form: Berkeley and Berkeley (1965) for Clayton, and Ewan and Ewan (1970) for Banister. In 1693 Clayton (Berkeley and Berkeley, 1965:57-59) commented at length on the extensive shell beds, speculated as to their derivation from living mol- lusks below sea level versus inorganic origin London in the 1620s and probably were available to Shake- speare. They are thought to have provided at least part of the inspiration for "The Tempest," quoted at the beginning of this prologue (Kermode, 1958:xxv-xxxiv). NUMBER 53 within the rock, and stated: Often, in the looser Banks of Shells and Earth, are found perfect Teeth petrefied, some whereof I have seen, could not be less than two or three Inches long, and above an Inch broad: Tho' they were not Maxilary Teeth, the part that one might suppose grew out of the Jaw, was polish'd and black, almost as Jett; the part which had been fasten'd in the Jaw and Gums, was brown, and not so shiningly polish'd, or smooth; if they were, as they seemed to be, really Teeth, I suppose, they must have been of Fishes [sharks?]. The Back-bone of a Whale, and as I remember, they told me of some of the Ribs, were digg'd out of the side of a Hill, several Yards deep in the Ground, about four Miles distant from James-Town, and the River. Mr. Banister, a Gentle- man pretty curious in those things, shew'd me [in 1686; Ewan and Ewan, 1970:xx, 58] likewise the Joynt of a Whale's Back-bone, and several Teeth, some whereof he said, were found in Hills beyond the Falls of James River .... At least some of the teeth undoubtedly repre- sented Pleistocene mammals from west of the Coastal Plain, as various authors have supposed (Ewan and Ewan, 1970:331), but the whale ver- tebra and probably some of the (shark?) teeth must have been local, especially in view of Ban- ister's own writings, for example (in Ewan and Ewan, 1970:332): 20 or 30 miles [32 or 48 km] up ye freshes of James River I found great variety of petrified oysters, scallops, bones &c. among them these strange stones, which I am not so good an ichthyologist to assign to what fish or fishes they might belong, if they were ever reall teeth. & higher up yet within 12 miles [19 km] of ye falls, about 1/2 mile [0.8 km] from the River in a gully on ye side of a hill, near a small creek 40 or 50 foot [12 or 15 m] perpendicular above ye flowing of ye tyde, I met with another of these teeth, like ye first but smaller, its armature of ye colour of cinammon, the rest liver- colour'd, & with it ye like variety of ye sea shells &c. I am informed to[o] that divers of ye high banks downwards (tho out of ye tydes reach as it now flows) are compos'd almost wholly of them, & that they are found there also in many places remote from ye river. He went on to debate whether the fossils indicate former presence of the sea or direct formation in the rock, and he made drawings of a shark tooth, fossil invertebrates, and a sting ray spine (Ewan and Ewan, 1970, figs. 65, 67, 68). There is no doubt that Banister aided and influenced Martin Lister, and his drawings, notes, and specimens may well have provided the basis for Lister's (1685-1692; in Ewan and Ewan, 1970:315) widely heralded first published illustrations and descriptions of American fossils (in Ward and Blackwelder, 1975:3; Wilson, in prep.; Ewan and Ewan, 1970:312 etsqq.). In his widely known work, Mark Catesby (1731:second vii) made the following comments on Neogene fossils of the Coastal Plain: There is no Part of the Globe where the Signs of a Deluge more evidently appears than in many Parts of the Northern Continent of America, which, though I could illustrate in many instances, let this one suffice. Mr. Woodward, at his Plantation in Virginia, above an Hundred Miles [161 km] from the Sea, towards the Sources of Rappahannock River, in digging a Well about seventy Feet [21 m] deep, to find a Spring, discovered at that Depth a Bed of the Glossopetrae [shark teeth], one of which was sent me. All Parts of Virginia, at the Distance of Sixty Miles [96.6 km], or more, abound in Fossil Shells of various Kinds, which in Stratums lie imbedded a great Depth in the Earth, in the Banks of Rivers and other Places, among which are frequently found the Verlibras, and other Bones of Sea Animals. Lewis Evans (in Gipson, 1939; White, 1952a) was a very perceptive mapmaker of the middle 1700s who understood the nature of fossils, and delineated physiographic features, including the Coastal Plain and the fall line. One of the great unknowables of American geology is the impact that the works of Johann David Schopf would have had, had they been widely available in English to his contemporaries and immediate successors. His American travels, published in 1788 in German and not translated until 1911, remained rare until reprinted recently (Morrison, 1968), and his American geology, pub- lished in 1787, was not published in translation until 1972 (Spieker, 1972). Although now widely appreciated and acknowledged as an accurate observer and clear thinker in a time of uncertainty in geology, his works had no known impact in America for well over a century after their pub- lication, which is to say not until long after the progress of the science had passed them by. Schopf was in America from 4 June 1777 to 29 March 1784, during most of which time he was closely limited to the vicinity of New York and Philadelphia as surgeon to German troops in service to the British. On 22 July 1783, however, he left New York on his generally southward SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY travels that took him through the Coastal Plain of Virginia and North Carolina. He went out of his way to visit Yorktown (Morrison, 1968 (2):82- 85), because of its significance as a "remarkable theatre of a decisive military event, as well as by the wish to examine the great shell-banks there, which are an object of curiosity to every stranger"; he discussed in enthusiastic terms a shell bed exposed in a mill race halfway between Williamsburg and Yorktown, noting as well "large bone-fragments, presumably of whales." Continuing southward, he noted shell banks also on the Tar River (Morrison, 1968(2): 125), where he had already mentioned that proboscidean re- mains had been found (Morrison, 1968(1):269). In his geological treatise (in Spieker, 1972:48-49) Schopf made more generalized statements about the distribution of the shell bed and mentioned "sharks' teeth, whale and other bones" as well. Intimations of things to come in the new re- public, even without benefit of Schopf's insights, are provided by three sets of observations pub- lished before 1800. Their significance lies not in their originality but in their suggesting fairly widespread geological sophistication and the de- velopment of a society in which homegrown in- vestigations of rather narrow esoteric topics could find outlet in American journals. The first of these, published in 1785, by the American Acad- emy of Arts and Sciences in Boston, consists of a detailed description of the geologic section in the vicinity of Yorktown, Virginia. This publication is based upon observations made by a revolution- ary general, Benjamin Lincoln, during the last weeks of the war, prior to the British surrender at Yorktown on 19 October 1781. In June 1786, the Reverend Samuel West with Dr. William Baylies and others visited Gay Head, Martha's Vineyard, the northernmost emergent outlier of the Coastal Plain. Both West (1793) and Baylies (1793) published accounts of the visit. West had been "appointed by the Academy to be a committee, to examine the mineral produc- tion of Gay Head." According to West (1793:148), "the inhabitants presented us with a petrified bone, said to be one of the vertebrae of the whale, which they told us they found in the cliff: It is very heavy, owing, I apprehend, to a metallick impregnation. They also brought us two shell fish, which were petrified: These were taken out of the cliff" Baylies (1793:155) added: "The bones of whales, sharks' teeth, and petrified shellfish, are frequently picked up, scattered up and down the cliff, at a considerable distance above the surface of the water." The third example is that of Benjamin Henry Latrobe, a prominent architect and engineer who directed construction of the U.S. Capitol and the White House, and designed many other build- ings, as well as municipal water systems and canals (Lintner and Stapleton, 1979). He made practical use of geology in his profession, as re- flected by the section through Richmond, Vir- ginia, in his journal for 4 May 1798 in connection with construction of the penitentiary of his design (Lintner and Stapleton, 1979, fig. 1). On the same pages he recorded a detailed log to a depth of 71 feet [21.6 m] of the well at the penitentiary, and in 1799 published commentary in the Transactions of the American Philosophical Society on the fossil teeth and bones, which accompany this memoir,* and which with many hundred more, were dug out of a well at Richmond, from the depth of 71 feet .... * The teeth appear to be those of a shark. They are highly enamelled and extremely sharp: their roots are perfectly sound and entire, and the minute and almost transparent jags of many of them are as perfect as the rest. They are found in every well, dug in or near Richmond, to a sufficient depth; and, as I am informed, in every deep well for many miles below the city. The stratum in which they lie consists of highly sulphurated blue clay, abounding in pyrites, and which has the appearance of having been mud. They were first discovered in the beds of rivulets, which had worn their channels to the depth of this stratum; and obtained the name of Indian Dart-points, in the same manner, as the immense oysterbeds, which have been quitted by the ocean, are vulgarly called Indian oyster-banks. The bones were dug from the same stratum. Among them are two out of six bones, which formed A paw of some animal unknown to me. Many very sound vertebrae of fish, and a remarkably perfect thigh bone of a large bird have been in my possession. The paper is accompanied by well-executed drawings (reproduced by Lintner and Stapleton, NUMBER 53 1979, fig. 2) of four shark teeth, the two recovered bones, and an outline of a third bone. The bones are indeed those of the paw, or forelimb, of a small porpoise, well known in the Miocene de- posits of the Coastal Plain, including Richmond (Ray, 1976:10). The two recovered are the hu- merus and ulna of a mature individual, and the bone outlined is the radius. The reference to the fossil bird bone may be the first for the Coastal Plain. It is interesting to note that on 16 April 1818, Latrobe advised on the proposed construction of a canal in North Carolina without going there, based on extrapolation from his knowledge of geology from New York to the Roanoke River (Lintner and Stapleton, 1979:112). It is equally interesting, and perhaps not entirely coincidental, that in April of the following year, Latrobe's fellow surveyor and engineer, William Smith (also the founder of stratigraphic paleontology), seri- ously considered an offer to come to North Car- olina as an advisory engineer; he declined, how- ever, and by June found himself instead in debtor's prison (Eyles, 1969:157). Might the sub- sequent history of geology in the Coastal Plain of North Carolina have been significantly altered had he decided otherwise? Probably the ap- proaches to both Latrobe and Smith stemmed from the Board of Internal Improvements of the state, which concerned itself with surveys of rivers, and for railroads, turnpikes, canals, and swamp drainage (Merrill, 1920:363). Samuel Latham Mitchill (1818), in what may be regarded as one of the last major publications of the classical period, reviewed fossil records in North America. Among many others he noted several of interest for the Neogene of the middle Atlantic Coastal Plain, including some occur- rences of fossil wood and the following of verte- brates: I remember, that petrified bones, apparently of a whale, were brought from the shore of Chesapeake Bay, near the place where the river Patuxent enters it, to the City of Washington, by Mr. O'Neale. (Mitchill, 1818:394) Shark's teeth, or glosso-petrae, are often raised on digging wells, further down the [Potomac] river, as at Diggas's point, for example. (Mitchill, 1818:396) Mr. Chevallie brought me, from Richmond, entire tri- angular teeth, apparently of sharks, and pieces of bones, probably of whales, dug from the depth of between sixty and one hundred feet [18 and 30 m], in the city of Richmond .... in the neighborhood of Williamsburgh, in 1802, a considerable portion of a whale's skeleton was discovered. It was about four or five feet [1.2 or 1.5 m] under ground; two miles [3.2 km] distant from the shore of James' river, and fifty [80.5 km] from the Atlantic ocean. Among other parts were fragments of the ribs, and all the vertebrae regularly arranged, and very little impaired as to its figure. (Mitchill, 1818:397) At a place called Fishing creek, 150 miles [241 km] from the sea coast, and almost four [6.4 km] from Tarborough, in digging some little depth, they found a part of the skeleton of a whale, with sea shells in abundance .... The skeleton of another whale, together with a petrified portion of a shark's jaw with teeth, has been found at a place called Williamstown, more than 100 miles [161 km] from the sea coast. About a year ago, the skeleton of a huge animal was found on the bank of the Meherrin river, near Murfrees- borough. It was dug out of a hill, distant sixty miles [97 km] from the ocean. Capt. Neville and Dr. Fowler, who visited the spot, gathered the scattered vertebrae which the negroes had thrown out, and laid them in a row thirty-six feet [11 m] in length. If to this the head and tail be added, the creature must have been perhaps fifty feet [80.5 m] or more in length. The former of these gentlemen enriched my collection with two of the teeth and a joint of the back bone that he brought away. The teeth weigh sixteen ounces [0.45 kg] each. They are covered with an ash-coloured enamel, except at the roots where they were fastened in the jaws. Their figure is triangular, the sides towards the apex meas- uring six inches [15.24 cm] each, and the base four inches and a half [10.16 cm] across. The joint of the back is not cartilaginous, but actually bony. It is in some degree petri- fied, and weighs twelve pounds and a half [5.7 kg]. It, in all likelihood, belonged to a shark or a sea-serpent. (Mitchill, 1818:400-401) Although the distances from the sea are exag- gerated, the records of large whales from the vicinity of Tarboro and Williamstown probably apply to mysticetes preserved in the Yorktown Formation; however, the "petrified portion of a shark's jaw with teeth" is more suggestive of an archaeocete, which could only have come from the Eocene Castle Hayne Formation. Similarly, if the large triangular teeth from near Murfrees- boro were indeed from the same animal as the skeleton, they could scarcely represent any animal 8 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Other than a large archaeocete. However, it seems unlikely that the Castle Hayne Formation would have been penetrated in that area, at that time, but likely that the skeleton was that of a mysticete and the teeth those of Carcharodon associated in the same Neogene strata. These early investigations and reports were an essential prelude to the subsequent development of geology. For example, the creation by Benja- min Silliman of the American Journal of Science, which was the first American periodical of broad scope devoted primarily to geology, could scarcely have come into being earlier than 1818, because the ground rules of the science were only then being laid. Without the preceding primitive ef- forts as a substrate, there would have been neither authors nor audience for such a journal. Beginning in the 1820s and continuing apace through the next two decades, American science underwent rapid expansion and developing professionalism, characterized in geology by the first official state geological surveys and in pa- leontology by the development of increasingly standardized procedures, including adoption of Linnaean systematics. It is neither feasible nor necessary to attempt to chronicle the burgeoning developments from this time onward, for the history and literature have been thoroughly cov- ered in standard sources such as Darton (1896), Gregory et al. (1973, and volumes cited therein, by Camp et al.). Hay (1902), Hazen and Hazen (1980), Merrill (1906, 1920, 1924), Nickles (1923, 1924), and Schneer (1979). For the individual states the literature for Maryland is covered by Clark (1897), Mathews (1897), and Shattuck (1904); for Virginia, Clark and Miller (1912), and Roberts (1942); for North Carolina, Laney and Wood (1909), Clark, et al. (1912), and Stuckey (1965), and Riggs and O'Connor (1975). The first official state geological survey of North Carolina was conducted by Denison Olmsted and Elisha Mitchell, 1824-1827, and may be regarded with some justification as the first for any state (Back, 1959; Merrill, 1920:363). Following closely were the surveys of Julius Ti- moleon Ducatel for Maryland, 1833-1842, and of William Barton Rogers for Virginia, 1835-1841 (Aldrich and Leviton, 1982). These surveys, all including some work on the Coastal Plain, were followed by others in the nineteenth century. In North Carolina surveys were made by Ebenezer Emmons in the 1850s (Johnson, 1982) and Wash- ington Caruthers Kerr from the Civil War to 1885, and were supplemented by other work, conducted in part by the same geologists but also by others in increasing numbers. Serving as the capstone for nineteenth century efforts and as the foundation for all subsequent work on middle Atlantic Coastal Plain geology are the unifying, comprehensive publications by William Bullock Clark and his coworkers: for Maryland, Clark, Shattuck, and Dall, 1904: for Virginia, Clark and Miller, 1912; for North Carolina, Clark et al., 1912. By 1830, Timothy Conrad, Samuel G. Morton, and a few others had begun the work that would result in monumental publications (e.g., Conrad, 1830, 1842, and in Dall, 1893; Morton, 1829, 1834) in systematic and stratigraphic paleontol- ogy, based almost entirely on invertebrates, al- though Morton also published on vertebrates, mostly of Cretaceous age. Richard Harlan, char- acterized as America's first professional vertebrate paleontologist (Simpson, 1942:161), began work in the 1820s, and in 1842 he published the first formal description of a fossil cetacean from the Neogene of the Coastal Plain, Delphinus calvertensis (later transferred to Lophocetus). Work on the fabulously rich invertebrate faunas by numerous subsequent researchers, among whom Julia Gard- ner (1948) may be mentioned as a leading prac- titioner, continues as reflected in the present vol- umes. For the vertebrates, Harlan's small begin- ning was followed by the extensive work, primar- ily on cetaceans, of Joseph Leidy, Edward Drinker Cope, Frederick William True, and above all Remington Kellogg. Perusal of their many publications on fossil vertebrates of the Chesapeake Series (Hay, 1902; Gregory et al., 1973; Knapp, in prep.) reveals very little on the Yorktown Formation and relatively little on North Carolina. The reasons are readily appar- ent; in spite of the occasional notice of large whale skeletons since early colonial times and the NUMBER 53 superabundance of invertebrates, natural expo- sures have produced an unreliable crop of verte- brate material. Of that, very fittle of adequate quality reached the hands of researchers, as com- pared, for example, to the abundance of good specimens from the Calvert Formation of Mary- land. Beds of Calvert age are unknown in outcrop in North Carolina. All of this changed dramatically and suddenly with the opening of the Lee Creek Mine, which provided the first exposure in North Carolina of deposits in part equivalent to the Calvert For- mation (the Pungo River Formation) and contin- uously renewed access to the Yorktown Forma- tion. With respect to vertebrates, publication of the present volumes will transform the Yorktown Formation from virtual terra incognita to one of the richest known deposits. Conclusion Much is made these days of a priori research design, of deciding first on a significant problem to be pursued, then going forth to select an appropriate vehicle to carry the scholar to his goal. For this reason it is argued that museums should not be cluttered up with collections unless there is a specific proximate purpose in mind. Although there is some justification for this re- action to traditional methods ("stamp collecting" to some), it is all too characteristic of our culture to lurch from one extreme to another, to lose interest in and even discontinue an activity be- cause it is not new. In fact, however, the great work of discovering and deciphering the record of life on earth has barely begun. If there is a loss of confidence in museum science, if it collapses, it will not be through indifference or hostility from without (the National Museum of Natural His- tory had 5,464,229 visitors in 1979), but by im- plosion, when museums are no longer populated by museum scientists. A colleague recently stated that a good museum scientist should have a "sub- clinical obsession" with collections, seemingly more appropriate than a fear of pursuing our profession too vigorously. There will be no lack of external forces to set practical limits to growth of collections, not the least of which is availability in the case of vertebrate fossils. A strong element of self-deception creeps in if we deny the often dominant opportunistic factor in our research design. After his more than 40 years of productive research on marine mammals of the Atlantic Coastal Plain that had yielded virtually nothing from the Yorktown Formation or from North Carolina, Remington Kellogg understood the im- portance of capitalizing on the opportunity pre- sented. Before the Lee Creek Mine existed, there was no possible means to learn about the Pungo River Formation except through limited access by drill- ing, and there was no prospect of significant extension of knowledge of vertebrates of the York- town Formation. Although to my mind, the tra- ditional goal of increasing and diffusing knowl- edge of earth history is adequate justification for the study of the geology and paleontology of any place, it may be pointed out also that only through multiplication of richly documented points in space and time will we be able to perceive general patterns of distribution, evolu- tion, and correlation. It is hoped, therefore, that these volumes will demonstrate the utility of being ready to exploit opportunities as they arise. There can be no doubt that we could have done more and better. There remains a great need for better stratigraphic control, more comprehensive taxonomic coverage, and better quality materials, especially of the vertebrates, but these needs can be satisfied only by future work, most especially by more extensive and leisurely access to sections in place, perhaps through setting aside a research reserve. That, however, remains for another time and possibly for other hands. I can only refer again to John Brickell's felicitous concluding words quoted at the beginning of this prologue, ''Hhat their laudable Attempts may meet with just En- couragement shall be my constant Wish and Desire." Aeknowledgments Among the scores of people whose efforts have furthered this project, mention can be made here only of those whose contributions have been the most comprehensive and sustained. Most of these 10 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY same individuals and many others are thanked in appropriate chapters for specific contributions. First and foremost, it is my duty and pleasure to acknowledge the patience and forbearance of each contributing author, many of whom com- pleted their manuscripts literally years ago, and, through unfortunate delays have been forced to revise and update their chapters repeatedly. My grandiose planning and naive expectations have cost them dearly. If the neophyte traveler should never visit a place for the first time, then neither should the novice attempt to edit a collection of research papers for the first time. Texasgulf Inc. has been cooperative and hos- pitable from the beginning, and many employees have taken active interest in our work. Above all. Jack H. McLellan must be singled out, not only as the catalyst who set the entire process in mo- tion, but as a constant advisor and stimulator, an avid and perceptive collector, and a scholarly contributor. The breadth and depth of his knowl- edge and his mastery of concise explication have enabled him to move as comfortably in museum sciences as in engineering and mining circles, while his personal friendship and unpretentious capability have made the years of this project an experience to savor. Other officials of Texasgulf Inc. who have been uniformly helpful in promoting our work include John Althouse, H.V.W. Donohoo, David C. Ed- miston, Jr., Gino P. Giusti, Jack Hird, Earl M. Mason, David McDonald, Steve Mollison, James R. Paden, Wilton W. Smith, Scott Stidham, and Thomas J. Wright. Although the two lists are by no means mutually exclusive, those employees, past and present, whose assistance has been pri- marily with specimens, include William D. Ben- nett, John Boyd, Ralph Chamness, Nat Cowell, Raymond Douglas, Delbert R. Jones, Clyde Swindell, and Webster Walker. Colleagues and students from several universi- ties have contributed important material; Scott W. Snyder, Jean Lowry, and Stanley R. Riggs, of East Carolina University, have been especially helpful. Edgar A. Womble, Jr., first president of the North Carolina Fossil Club, Vince and Judy Schneider, and other members of the club, have contributed numerous specimens. Among Smith- sonian staff members, Franklin L. Pearce, now retired, and Gladwyn B. Sullivan, not only have collected fossils at the Lee Creek Mine, but have prepared virtually all of the vertebrate specimens that have come into the museum. Dolores Larkie has handled the myriad details of paperwork and typing. Jack F. Marquardt, Carolyn Sue Hahn, and their associates in the Smithsonian Institu- tion Libraries, have always been cordial, resource- ful, and fast in ferreting out even the most obscure literature. At the risk of slighting the numerous colleagues from the Smithsonian Institution and the U.S. Geological Survey who have been helpful in so many ways, I would be remiss if I failed to make special mention of Thomas G. Gibson for his extensive help with stratigraphy, including analysis of many samples of matrix associated with significant macrofossils, Druid Wilson for sharing his unique knowledge of Coastal Plain stratigraphy and paleontology, and Frank C. Whitmore, Jr., for being there cheerfully and without fail whenever help of whatever kind was needed. Reviewers of manuscripts are acknowl- edged in each chapter. Most of the financial support for Smithsonian field work was provided by the Smithsonian Research Foundation, the Walcott Fund, and the Kellogg Fund. I have saved until last my expression of thanks to Peter J. Harmatuk, for he is truly in a class by himself At the end of 1975 he retired early, at financial loss, from a successful career because it interfered with his paleontological field work. Yet he is the antithesis of the monomaniac misfit hiding an inadequate personality among fossils, for he is a leader in his community and pursues other, nonpaleontological interests with similar vigor. I would regard his acquaintance as one of life's rare pleasures had he never collected a fossil. However, that is far from the reality. In the course of hundreds of collecting days at the Lee Creek Mine, to say nothing of numerous other localities in eastern North and South Carolina, he has collected with unflagging enthusiasm more fossils of more kinds for science than anyone who has NUMBER 53 11 ever worked the middle Atlantic Coastal Plain. In many cases he has been the first to bring a locality to the attention of paleontologists, to influence other collectors toward a scientific ori- entation, to recognize an unusual stratigraphic occurrence, or to discover specimens unprece- dented in kind, quantity, or quality. More than once his tenacious curiosity has forced me to pay attention at length to something of interest pre- viously brushed aside. His rare combination of self-effacing humility and constitutional inability to accept glib answers based on faulty reasoning from vested authority has made our association a source of continuing satisfaction and education for me. If one ever needed a reminder that pa- leontology traditionally has been and remains largely a field science, the enjoyment and ad- vancement of which is open to Everyman to the extent of his ability, effort, and interest, Pete Harmatuk provides irrefutable proof With little of the externally conferred advantages of educa- tion, opportunity, and funding, his contributions demonstrate that there is no substitute for innate intellect and good character. Literature Cited Aldrich, Michele L., and Alan E. Leviton 1982. William Barton Rogers and The Virginia Geolog- ical Survey, 1835-1842. /« James X. Corgan, edi- tor, The Geological Sciences in the Antebellum South, pages 83-104, 5 figures, 1 table. University, Ala- bama: University of Alabama Press. Back, William 1959. Emergence of Geology as a Public Function, 1800- 1879. Journal of the Washington Academy of Sciences, 49(7): 205-209, 1 figure. Bailey, Worth 1938. Lime Preparation at Jamestown in the Seven- teenth Century. William and Mary College, Quarterly Historical Magazine, 18(1): 1-12, 2 drawings, 1 fig- ure. Baum, Gerald R., and Walter H. Wheeler 1977. Cetaceans from the St. Marys and Yorktown For- mations, Surry County, Virginia, yourwa/ of Paleon- tology, 51(3):492-504, 2 plates, 1 figure. Baylies, William 1793. Description of Gay Head. Memoirs of the American Academy of Arts and Sciences, 2(1): 150-155. Berkeley, Edmund, and Dorothy Smith Berkeley, editors 1965. The Reverend John Clayton: A Parson with a Scientific Mind, His Scientific Writings and Other Related Papers. Virginia Historical Society Documents, 6:lxiv -I- 170 pages, 1 map, 4 plates. Charlottes- ville: The University Press of Virginia. Blackwelder, Blake, W., and Lauck W. Ward 1976. Stratigraphy of the Chesapeake Group of Maryland and Virginia. (Guidebook 7b: Northeast-Southeast Sec- tions Joint Meeting 1976). 55 pages. Arlington, Virginia: Geological Society of America. Brickell, John 1737. The Natural History of North-Carolina, x -I- [iv] -I- xiv -I- 417 -I- [1] pages, frontispiece (folding map), 4 plates. Dublin, Ireland: James Carson. [Reprinted 1969 with a new introduction by Carol Urness. New York and London: Johnson Reprint Corpo- ration.] Catesby, Mark 1731. The Natural History of Carolina, Florida and the Bahama Islands: Containing the Figures of Birds, Beasts, Fishes, Serpents, Insects, and Plants: Particularly the Forest- Trees, Shrubs, and Other Plants, Not Hitherto Described, or Very Incorrectly Figured by Authors. Together with Their Descriptions in English and French. To Which, Are Added Observations on the Air, Soil, and Waters: With Remarks upon Agriculture, Grain, Pulse, Roots, &c. To the whole. Is Prefixed a New and Correct Map of the Countries Treated of. Volume 1, xii -I- xliv 4- 100 pages, map, 100 plates. London: Published by the author. Clark, William Bullock 1897. Historical Sketch, Embracing an Account of the Progress of Investigations Concerning the Physical Features and Natural Resources of Maryland. Maryland Geological Survey, 1(2):43-138, 4 plates. Clark, William Bullock, and Benjamin LeRoy Miller 1912. Physiography and Geology of the Coastal Plain Province of Virginia. Virginia Geological Survey Bul- letin, 4: 274 pages, 19 plates, 1 figure. Clark, William Bullock, Benjamin L. Miller, L.W. Stephen- son, B.L. Johnson, and Horatio N. Parker 1912. The Coastal Plain of North Carolina. North Caro- lina Geological and Economic Survey, 3:552 pages, 42 12 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY plates, 21 figures. Clark, William Bullock, George Burbank Shattuck, and William Healey Dall 1904. The Miocene Deposits of Maryland. Maryland Geo- logical Survey, 2:xxi-clvi, 9 plates. Conrad, Timothy A. 1830. On the Geology and Organic Remains of a Part of the Peninsula of Maryland.yo«r«a/o/^^ FIGURE 5.—Geologic column for Bal- timore Gas and Electric Company Well (BG&E) at nuclear power plant site at Calvert Cliffs, southeast of St. Leonards, Calvert County, Maryland. -^( light gray muddy silt, indurated missing interval olive-green muddy fine sand, indurated Bed 20 Bed IP (upper) olive-gr uddy fine -59 dark olive muddy fine missing interval _S7 "56 indurated -5J light gray muddy fine light gray mud 52 dark olive-gree ed 18 O ed 17 3 -9C alive-brown muddy fine sand, not tied and burrowed Bed 13 Bed 12? olive-brown muddy fine sand 3live-browr. .;ilt and fine sa s Bed 11 i- Bed 10 5 olive-brown muddy fine sand, nottled and burrowed muddy diatomaceous sand olive-green sandy diatomite olive-green diatomite, lens of nearly pure diatomite, and darker muddy diatomite silty diatomit uddy fine sand strongly mottled, more muddy diatomite sandy diatomite light olive-gray muddy dusky brown, ollve-greer reworked Eoce muddy glauconitic sand, oysters olive-Brown glauconitic sand, poorly sorted quartz pebbles, broken shells Popes Creek . , ., Sand *° 1 Members'' 42 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Depth (m) (ft) 0 0- Hammond Well 61 200- :^.;-.-;.v.:.&: •;:■•. y;.y.-^;:-;•;•;•£> ■ ■:'::^:■':^o■: ■ci'.-'-.fr'-:':'-:'.''^ gravelly sand gravelly very coarse sand post-Yorktown (?) ■.'•'. •".'•■.'• •'.■ v. •. • ••'o.'. .•■o.:;.-_- 244 800-—:.—A;—.;;vT^:- Jb■:^-■Z-■I.- Z-Z-3D.^^ gray silty clay, few fossils fine sand gray silty clay, few fossils Calvert Formation 305 lOOOH—— —_"—^T^ brownish gray silty cl ay f6w fossils glauconitic sand brown clay Eocene(?) FIGURE 6.—Geologic column for Larry G. Hammond Well, Dorchester County, Maryland, showing inner-shelf facies in Calvert and Choptank formations and delta front to topset delta conditions in St. Marys and Yorktown formations. and varies from relatively pure diatomite (as much as 65 percent of the sediment (Glaser, 1971:41)) to sandy and muddy diatomite (Figure 5). The member is 60 to 70 feet (18 to 21 m) thick in outcrop (Shattuck, 1904, pi. 5) and thickens down-dip to about 100 feet (30 m) in the BG&E Well (Figure 5). West of Chesapeake Bay, the top of the Fairhaven is marked by a distinct discon- tinuity indicated by an undulating surface that is overlain by a thin oyster bed of the Plum Point Marl Member (Dryden, 1936). Calcareous fossils are scarce in the diatomaceous beds in outcrop NUMBER 53 43 (Gibson et al., 1980), but Foraminifera are present in the uppermost, more sandy and less diatoma- ceous part, and Mollusca are found in the basal sand. Foraminifera are found throughout the dia- tomite beds in the BG&E core, although they are lacking in the diatomaceous intervals in many other subsurface sections. The lower beds of quartz sand in the Fairhaven Member are characterized by olive-brown, olive- green, to light greenish white fine to medium sand, muddy in places, that may contain coarse sand to fine gravel pebbles of quartz, phosphate, and glauconite. A thin layer of quartz and phos- phate pebbles, bone fragments, and phosphatized mollusk shells is often found at the base. Cylin- drical burrows are common in the sands. These sands crop out in Maryland west of Chesapeake Bay from near Fairhaven southward to Popes Creek, a tributary of the Potomac River, and extend into Virginia where they are muddier and thinner, such as those in the Oak Grove core (Reinhardt, Newell, and Mixon, 1980). The equivalents of zones 1 and 2 can be traced in the subsurface of the Eastern Shore of Maryland, where basal sands containing Pecten humphreysii are found in Well 220 (Figures 4: loc. 12; and 7), and into Delaware, where Valia, Khalifa, and Cameron (1977) described the basal unit as peb- bly glauconitic and quartzitic sands containing typical Calvert Foraminifera. Poag (1978) re- ported conspicuous glauconite in the lower few meters of the probable Calvert Formation equiv- alent in core holes in the Norfolk Canyon area east of the Delmarva Peninsula, extending the distribution even beyond the edge of the present- day shelf Because of the wide distribution of these distinctive sands (which represent the basal transgressive unit of the rapidly subsiding Salis- bury embayment) and because of their lithologic distinction from the remainder of the Fairhaven Member, I herein classify them as the Popes Creek Sand Member of the Calvert Formation (Figure 9). The type section is the outcrop 100 yards (91 m) southeast of the mouth of Popes Creek, Charles County, Maryland, on the north bank of the Potomac River (Figure 8). Glauconite is rare (Glaser, 1971:46) in the western outcrop exposures, but increases to significant proportions eastward. Valia, Khalifa, and Cameron (1977) showed that beds, which I consider a part of this member, consist of 20 to 50 feet (6 to 15 m) of pebbly glauconitic sand in the subsurface of Del- aware. The thickest part of the outcropping Fairhaven (zone 3 of Shattuck, 1904) is retained as the Fairhaven Member. This member comprises the sandy and muddy diatomaceous strata above the Popes Creek Sand Member and below the shelly sand of the Plum Point Marl Member. The Plum Point Marl Member consists of beds of olive-green to olive-brown silty and clayey fine sand that generally contain scattered to highly abundant molluscan shells, a considerable num- ber of marine mammal bones, and other verte- brate remains. Zones 4 to 15 of Shattuck (1904) compose this member, which reaches a thickness of 90 to 100 feet (27 to 30 m) in outcrop along the Calvert Cliffs. AREAL RELATIONSHIPS.—Very little change is found in the Calvert Formation from the Calvert Cliffs area (Figure 5) eastward to Well 220 near Bucktown (Figures 4: loc. 12; and 7) on the Eastern Shore of Maryland. Well 220 is 20 miles (32 km) obliquely down-dip from the BG&E Well, and the similar thickness and lithology is somewhat surprising. The Calvert in Well 220 is only slightly thicker than at the outcrop (~220 ft, 67 m), and the facies are similar to those in the BG&E core. The Popes Creek Sand Member is slightly thicker in Well 220 (25 ft, 7.6 m) and is more glauconitic in the lower part. The Fairhaven Member is almost identical in the two wells as Well 220 has 110 feet (34 m) of green diatoma- ceous clay .The Plum Point Marl Member also is similar; W^l 220 has 85 feet (26 m) of sand and sandy clays generally containing scattered shells, but abundant shells in some places. About 30 miles (48 km) to the east of Well 220, the Ohio Oil Company's Larry G. Hammond Well (Figure 4: loc. 13) penetrated 500 feet (152 m) of Calvert (Anderson, 1948). Here the entire Calvert consists of brownish gray to gray silty 44 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Well 220 Depth (m) (ft) 36 120- 52 170- to ^ ^ green sandy ^ ■ — —I clay green sand abundant shells green sandy clay bed 17(?) Choptank Formation green clay bed 15(?) ^ green sandy clay 67 220- Tto^-- .^.-^'^.:z. green sandy clay, shell Plum Point Marl Member 82 270- •.^•.'/■.■.toV:oJb';(b.- green sand, abundant shells green clayey sand bed 10(?) beds 5-9(?) 113 9-cr-c O ST"©—o o~i! 3 o cr-ft o o a~B •> O © B~© o ©"15 - O T"© O 9—Q © ( B" o o o~o O ©"©" > o ~^~o o en 97 320—1^ o o """o o 0-3 ©—O" o 0~0""0 o ©~! 1 0 o "O"© © s~0 f-© ©tro ©"0-© ©1 o o~a o s~6 © ©"Tjr 0~0 © © ~B" © © ©r^ fc © a~© © o~g s~o »- fir"© o o TJ" r 0 © ©^ 0 Q-ET o ( 370— o"W" © o ^^o ©- 0-^ ' G o T3~o © a~o o ©~Gr o o~o"'o o o~5 o p T3~"0 Q o T3~o o a~s sol ■XT o ©"iro o o" > o is~o O (5~0 o o o"o © "er o & - r"0 o w o ©0 6) 0 "0~TD o 0"T3" O 0~D J-D 0 "S~0 O 0"0 O ©—G> o ©"^ © 0""cr I 3 o er-Q © o "^o o 0 "O © o~u O O O 0 na o cro o o o 0- "CT O <3~I3 O ©—D O ~& 0~© O ^" O O 13 D o 0 0 "C D o o O © Q- (3"T> O "n~0 O (J-© C 1 3~0 o ©—© O © G ~w o €r^ © o "O"© c brown silty clay, becoming more diatomaceous upwards, weathers blocky medium to dark olive-brown silty clay, fresh appearance medium to dark grayish green: phosohatlc debris, brachiopods and small clam molds; small amount glauconite, lowest 0.6 m is sandier; burrowed and bloturbated light to medium tan-olive brown clayey very fine sand, no visible shell material or molds; small amount (1 percent) glauconite, few scattered medium sand grains; upper 0.7 m becomes more clayey; strongly bloturbated, massively bedded dark olive-gray-brown silty day light to medium olive-brown very fine to fine sand, slightly clayey; 1-2percent veryfine glauconite; few medium sand grains and occasional fine pebbles; massive bedding, some bloturbation tan to brown clayey very fine to fine sand; quartz and phosphate pebbles as large as 30 mm. in lower 0.1 m., decreasing in size upwards; lignitic material and shark teeth common in lower 0.2 m; less than 1 percent glauconite; massive bedding .contact of Calvert with Nanjemoy undulating due to burrowing and channeling; heightof burrows about 10 cmjower part of Calvert marked by pebbles, most prevalent in tower 0.2 m., becoming less common upwards Nanjemoy Formation 6.4 dark grayish-green clayey silt with some fine to medium sand; weathered on surface, brown to buff; fossils present only as molds of clams and snails, particularly Turritella; concretions present throughout, relatively branching cylindrically or in layers; glauconite 10-20 percent, muscovite 2-3 percent, heavily bloturbated FIGURE 8.—Geologic section on north bank of Potomac River, 100 yards (91 m) southeast of the mouth of Popes Creek, Charles County, Maryland. This outcrop is the type section of the Popes Creek Sand Member of the Calvert Formation. 46 Shattuck, 1904 ■ 510. .^' ■^»-'.'50 xg 15 14 13 ^2 11 10 9 5 "^ n 3 ^ 0 Plum Point marls CALVERT FORMATION .«■ >o ■ -.^ .^'.'- ~::^:--~-- ■«> -O ■© ^S '^ ■: 0. • .TO 0 ij"' . —B- a a-^ '-» »■ • «~o 0 *-o- 0 9—^ 9 0~« e 0 €-9 e 0 0-» 0 0 "«~ 0 0 O-T 0 0 v-o e o~« 0 0 e o-^o 0 •""» e 0 *«- ' 0 0 *~o 0 e"V 0 0 0 "O" * © "0-© e *—B r 0 0 T5-0 e o~o 0 4 «-©" 0 o"9-o 0 o-y ► 0 0 TJ—© 0 5-0 0 c ■» 00-0 o-o"© o"w- t O"* 0 "WO ©"w-o 0 © 0 © 0 e ff-w ^-o 1 n> 0 "w-© 0 0 w" 3 Fairhaven diatomaceous earth ~2 1SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Gibson, this paper Plum Point Marl Member CAL < m 3) H -n 0 2 > H 0 Fairhaven Z Member Popes Creek Sand Member FIGURE 9.—Chart shouting beds 1-15 and members of the Calvert Formation according to Shattuck (1904) and as treated in this paper. found on the Eastern Shore north of Well 220. But there, the middle part of the Calvert is divi- sible into two facies as seen in wells 224 through 231 (Figures 10-16). The Popes Creek Sand Member continues across the area as a 10 to 20 foot (3 to 6 m) glauconitic sand and gravel, containing quartz and phosphatic pebbles as much as 2 inches (5 cm) in diameter. The Fair- haven Member generally is similar to the beds west of Chesapeake Bay. However, the strata in some of the easternmost wells are less diatoma- ceous and are composed mostly of green clay and sandy clay. These are interbedded with green and brown clays in the farthest northeast well, 231 (Figures 4: loc. 5, and 10). A significant change is seen in the lower part of the Plum Point Marl Member (equivalent to beds 3 through 9 west of the Bay). Nearest the Bay (to the west, southwest, and northwest of Easton; wells 226, 227, 229 (Figures 4: loc. 11, 9, 6; and 11-13), the beds of green sandy clay containing scattered shells are still present. But in wells to the east of a roughly north-south line through Easton, this interval consists of brown clay and silty clay and gray sand interbeds. The interbedded clay and sand are from Vi to 2 inches (0.6 to 5 cm) thick, and the thicker sand beds appear to be crossbedded in the cores. Between the brown clay intervals, which may be 30 to 40 feet (9 to 12 m) thick, are intervals of green clayey sand containing scat- tered shells. This sequence is well developed in Well 224 located approximately 6 miles (9.7 km) southeast of Easton (Figures 4: loc. 10; and 14). The brown clay facies continues northeastward and eastward of Easton (wells 231, 225, and 228; Figures 4: loc. 5, 8, 7; and 10, 15, 16), and appears to be equivalent to the brown clay of the AUoway Clay Member of the Kirkwood Formation of New Jersey as mapped by Isphording (1970). The upper part of the Plum Point Marl Mem- ber (beds 10 to 15 of Shattuck, 1904) is similar over the entire area to the equivalent strata in the NUMBER 53 47 / Depth, (m) Cft: 15 50- Well 231 —-^ ■ -lip-. —■ ^"7 _>o _T:I' o .e~o o o s o o~o o o "o~o G rr B O ©"TS" O O O O ( o o~o o~o o o~o '\ green and brown interbedded clays Fairhaven Member Calvert Formation 46 150- glauconitic sand Eocene (?) FIGURE 10.—Geologic column for Well 231, near Sudlersville, Queen Annes County, Maryland, showing interfmgering of deltaic beds provisionally assigned to Alloway Clay Member of the Kirkwood Formation with inner-shelf sediments of the Plum Point Marl Member of the Calvert. Calvert Cliffs (beds of green clayey sand and clay and a basal sandy shell bed overlain by intervals of scattered to abundant shells). Farther east, Talley (1975) described the strata in a well at Greenwood, Delaware (Figure 4; loc. 4), but did not divide the Chesapeake Group into formations. The lower part of the section in this well consists of shelly, glauconitic sand, which Valia, Khalifa, and Cameron (1977) placed in the basal part of the Calvert. I consider the strata to be an eastward extension of the Popes Creek Sand Member. These strata are overlain by dark gray silty clay, probably equivalent to the Fair- haven Member, and then by brown to gray, commonly shelly silt and sand, probably equiva- lent to the Plum Point Marl Member. In southwestern New Jersey, Miocene strata of the time-equivalent part of the Kirkwood For- mation (Richards and Harbison, 1942), in wells such as NJ1-NJ3 (Figures 4: loc. 1-3; and 17- 19), consist of a basal sand overlain by green silty clay, and then by brown clay with thin interbeds of gray sand. The brown clay contains carbona- ceous fragments (as long as 5 cm) and generally lacks shells. Above the brown clay sequence, silt and medium sand containing abundant shells crop out near Shiloh (Figure 4: loc. 2) and have been penetrated by the NJl well near Cedarville (Figure 4: loc. 3). The stratigraphic sequence is similar to that found in the northeastern part of the Eastern Shore of Maryland. The strata in southwestern New Jersey belong in the Alloway Clay Member as used by Isphording (1970). To the northeast in New Jersey, Isphording reported that the clay grades into his Grenloch Sand Mem- ber (herein adopted), which in turn grades north- eastward into finely laminated, organic-rich, silty sand of his Asbury Park Member (also herein adopted; former Asbury Clay of Kiimmel and Knapp, 1904). 48 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Depth (m) (ft) 0 or Well 226 — ^ - 5— ."—-\<)- — green clayey sand 46 150- Plum Point Marl Member 61 200- 76 250- ; ■ .^-■.^O.-.'Sp.-. ■.•5JD;.' •.: er~o o o~o s> OTW o C'~V> o o S~9- O- ~s o © "W o o o~Er o o a~"© o o~5 o "9 o~(s o s~o o o "ar o o a—© e ^~u" o ~0" & © 'JS~0 0 «~o—o o "o""© o cr"s o 0 - o-o~D~© o Q~S~C O O ^~0 _0 SZD. O CL O" O O ~S" O 0 "0~C 3~© O 'D^'O 6 O G © O O O © O O O S~~D 3~© Q "5~B © 0~0 0""0 O ©"^ O G O © o "O 0 sny o © o o <3~o o «r--o o o o ~w © 0 o~t © j3 e 0—© o ©~5 © a 0—©" o o—B o o "H "o o s~© _© ©"o" 0 OTETJD-© "0—© © ©~C TJ © O T5~0 o (J—© 0 9-^9 ©"O""© o ©^ "©_© O-^~0 © 5~0 0 n»" OB"© 0"0"0^jj ©—o o^ST" o © a o r-0 0 o o © o 6 © -d O 0~D S"T3"^ O <3~0^ © 0~D—O 0~O 0 "0~8 0~0. O 0-© 0 0 "0~0- r© OT5 0 0"©" © O "Ij T3 ©—O 0 O~0 0~o ©i ) &~s e-g-B o o"~o -J .■'■£>.■ green slightly diatomaceous clays green clayey sand phosphate p>ebbles Fairhaven Member Popes Creek Sand Member Calvert Formation 91 300 - Eocene(?) FIGURE 12.—Geologic column for Well 227, near Easton, Talbot County, Maryland, showing basal sands of Popes Creek Sand Member overlain by sediments of inner-shelf environments in remainder of Calvert Formation and small amount of deltaic to prodelta influence in Choptank Formation. fine sand (Reinhardt, Newell, and Mixon, 1980). The lower sand is considered the equivalent of the Popes Creek Sand Member; the diatomaceous clay, the equivalent of the Fairhaven Member, and the upper 82 feet (25 m), the equivalent of the Plum Point Marl Member. A diatomaceous interval in the upper part of the upper sand contains diatoms equivalent to those of beds 14 and 15 in the Calvert Cliffs (Gibson et al., 1980). Whether the equivalent of the entire Plum Point 50 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Depth (m) (ft) 0 0- Well 229 gray clay gray-green sand Pleistocene!?) to Calvert Formation!?) green silty clay 15 50- green slightly diatomaceous clay scattered shells Fairhaven Member Calvert Formation ^ brown sandy clay ; coarse sand, glauconitic Popes Creek Sand Member 30 100- glauconitic sand Eocene (?) FIGURE 13.—Geologic column for Well 229, near Queenston, Queen Annes County, Maryland, showing coarse sand of Popes Creek Sand Member overlain by diatomaceous facies of Fairhaven Member. Marl Member is present here is unknown because biostratigraphic control is lacking. However, equivalents of at least part of each member are found. Although marine diatoms and dinoflagel- lates are present in the Calvert of the Oak Grove core, calcareous fossils are absent here and in most up-dip sections in Virginia. The absence of shells prevents lithologic recognition of the Plum Point Marl Member, although strata of equiva- lent age are present as determined from the dia- toms (Gibson et al., 1980). Lithologically similar sections of the Calvert Formation are found in other parts of the Virginia outcrop belt toward the western edge of the Coastal Plain. Typically, as along the Mattaponi River, the strata contain a basal sand and gravel bed characterized by quartz and phosphatic pebbles and shark teeth and other vertebrate fossils. This is overlain by diatomaceous clay and then by silty clay (Clark and Miller, 1912:126, 135). In down-dip subsurface sections to the east and southeast in Virginia, such as at Jamestown (Fig- ure 4: loc. 24), green sand and clay containing faunas indicative of open marine to partially restricted marine conditions are found. The Cal- vert Formation and equivalents are not found on the western part of the Norfolk arch (Figures 1, 3), but are found on or near the arch toward the coast, as in the Moores Bridge Well at Norfolk (Figure 4: loc. 25). Here the upper part of the Calvert section comprises beds of fossiiiferous green silty clay characteristic of the formation, but the lowermost 11 feet (3.4 m) is an olive clayey phosphatic sand, similar to that of the Pungo River Formation to the south. Thus, the Norfolk area contains the transition from terrig- enous clastic deposition in the Salisbury embay- ment to largely chemical and bioclastic deposi- tion in the Albemarle embayment to the south. ENVIRONMENT OF DEPOSITION.—The BG&E Well is the key section for paleoenvironmental interpretation of the Calvert Formation, because it contains the most detailed information in a single continuous section. From this focal point, the regional patterns for each of the members are discussed. Figure 20 shows sand percentage, spe- cies diversity of benthic Foraminifera, and per- centage of planktonic Foraminifera. The lowermost strata of the Calvert (Popes Creek Sand Member) accumulated during the transgression of the Calvert sea into the Salisbury embayment. Prior to this, the embayment re- ceived widespread glauconitic clay and sand dur- ing much of the early and middle Eocene and more restrictively in the late Eocene. In addition, Oligocene seas invaded part of this embayment NUMBER 53 51 Depth (m 0 Well 224 /p :■.:•. •/O.- -.iS?.-.. . gray clay gray clay and gravel Pleistocene!?) to Miocene!?) Formations bed 17(?) 15 50- :; •'59 • ■.^; :-^.-;£) green clay green sand shell layers brown clay with thin gray sand intert>eds green sand scattered shells bed 10(?) beds 5-9(?) Plum Point Marl Member 61 200- brown clay ^» ^ ^5© ^ ^ green sandy clay oyster bed bed 4 Calvert Formation 76 250- green clayey sand and sandy clay 91 300-. beds Fairhaven Member O SLJ3 ©_© ( fi- O © [i_0 © F> ei .£L O O Q a © 3 Q O © Q_© O Q_S 0_0 0 ra "© © ©_£> 107 350-i.Q-o~«>-Q-2 o <^ ? I © Q_Q^O © fl. 0 Q- i 0 _Q. © 'O\Q_ 0_ 0 -fi a ff CI a e> a pQ ri « a Q o p green diatomaceous clays diatomaceous clays and glauconitic clays interbedded glauconitic sandy limestone Popes Creek Sand Member FIGURE 14.—Geologic column for Well 224, southeast of Easton, Talbot County, Maryland, showing deltaic influence in lower and middle part of Plum Point Marl Member and inner- shelf environments in Choptank Formation. Well 225 0 0 15 50- 30 100- .:^.-.0,;::o.:ci. :••■.:•■ <^-:':-:P'. O G O O O O O £> f^ Q Q O O O o o ts — — ^ brown sand sand and gravel gravel orange sandy clay blue clay shells brown clay, micaceous brown clay, bee sand interbeds FIGURE 15.—Geologic column for Well 225, near Denton, Caroline County, Maryland, showing lesser amount of diatomaceous beds in Fairhaven Member and deltaic influence in Plum Point Marl Member and partic- ularly Choptank Formation. Choptank Formation green clayey sand 46 150- brown clay 61 200- T"^" .^^. ~ ^~^— green sandy clay 2—^2^:-p"-"-Z^. shell layers bed 10(?) 76 250- brown clay Plum Point Marl Member green sandy clay 91 300- brownish-green clayey sand Calvert Formation brown clay i^ro ©"0"o ©■©"© 0 O TJ © 0"0~D O TS~ o o o Gr"n" o o~o green diatomaceous clay 107 350- green sandy clay Fairhaven Member 122 400- glauconitic sand Eocene!?) NUMBER 53 53 Depth Well 228 brown sand 15 50- «£>.-^-.*s ■.^.'•^; blue clay brown sand blue clay brown clay, sand interbeds green clayey sand shells Calvert Formation 30 100-' :^;. glauconitic sand Eocene!?) FIGURE 16.—Geologic column for Well 228, near Skipion, Talbot Couniy, Maryland, showing deltaic influence in lower part of Calvert Formation. (B.W. Blackwelder, pers. comm., 1977; R.K. Ols- son, Rutgers University, pers. comm., 1977). The basal transgressive deposit of the Popes Creek Sand Member in the BG&E Well consists of slightly clayey fine to medium sand with pebbles as large as 11 mm, mollusks including Pecten humphreysii, and reworked clasts of Eocene sedi- ment containing Foraminifera. The base is com- posed of 50 percent sand or more. The lowest sample has abundant planktonic Foraminifera (29 percent) and a relatively high species diversity (25). The most abundant benthic species is Cibi- cides lobatulus, which today lives in agitated inner- to middle-shelf environments. Other abundant species include Valvulinaria floridana, Bolivina mar- ginata, B. paula, and Uvigerina calvertensis. The lower part of the member suggests deposition in a middle-shelf environment at depths as great as 80 meters and open ocean circulation. In the rest of the member, the percentage of sand decreases to between 25 and 35 percent, and the species diversity remains moderately high, in the low 20s. The planktonic percentage decreases to about 2 percent, and Valvulinaria floridana dominates the benthic assemblage. These changes suggest shal- lowing to near-shore sublittoral depths, probably less than 30 meters. This open marine environ- ment of the Calvert transgression is found over the entire Salisbury embayment, as seen in north- ern Virginia, Maryland, Delaware, and south- western New Jersey. The Fairhaven Member also is lithologically uniform and widespread in the embayment from east-central Virginia to northeastern Maryland. In the BG&E Well, the lower part of this member is of shallow sublittoral deposition, characterized by species diversities of 16 to 21 and planktonic percentages of less than 3. The percentage of sand is less (under 25 percent). Some fluctuations from sublittoral depths as great as 30 meters to either shallower marine or possibly restricted marginal environments are seen upward in the Fairhaven in the intervals with low species diversities of 9 to 12 species marked also by the absence of plank- tonic specimens; these intervals occur between others that contain higher diversities of 14 to 20 species accompanied by planktonic specimens, although less than 2 percent. The shallow marine environment for this member continues eastward across the basin except for slightly more silty and sandy and less diatomaceous strata to the east and northeast of Easton, Maryland. In Well 231, brown clay is seen within the diatomaceous unit that, together with the slightly increasing clastic content in that area, suggests the possibility of the beginning of deltaic conditions. The contact between the Fairhaven and the Plum Point Marl members on the western shore 54 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Depth (m) (ft) 0 (Fr Well NJl FIGURE 17.—Geologic column for Well NJl, near Cedarville, Cumberland County, New Jersey, showing interbedding of deltaic and prodelta sediments in the Kirkwood Formation. brown fine to coarse sand 15 SO- SO 100- brown gravelly sand 46 150-1 o C3 o ^> O o gray clay, gravel interbeds blue silty clay gray interbedded sand and clay nnedium sand medium to coarse sand shell layers 61 200-- brown clay Kirkwood Formation 76 250- 107 350- 122 400- green sand abundant shells brown and green clays z\ green silty clay coarse sand to fine gravel 91 300- •:ii--:i-- ■■:iJ^l:i ■ill -^ -: brown clay green sandy clay, shells interbedded with brown clay green silty clay, shelly brown clay green silty clay, some shells medium sand, glauconitic glauconitic sand Eocene!?) NUMBER 53 55 Kirkwood Formation Well NJ2 Depth (m) (ft) 0 OT 15 50- 30 100- 46 150- 61 200--^—_—_—_^ 76 250- brown sand brown sandy gravel brown clayey sand brown fine to medium sand black clayey sand brown clay, laminated "^ blue-gray clay interbedded blue and brown clays green clay shell fragments brown clay, thin beds of green sands interbedded with brown clay glauconitic sand Kirkwood Formation Eocene(?) In the BG&E Well, species diversity drops to 13 to 18 in the equivalents of beds 5 to 9 (sample numbers 30-35), planktonic specimens are ab- sent, and the sand content decreases to as low as 2 percent. At the outcrop, molluscan assemblages in these beds are largely bands of Corbula elevata. Valvulinaria floridana, Caucasina elongata, and Flonlus pizarrense are the dominant benthic species. The faunal characteristics and the high amount of clay suggest deposition in protected to marginal marine environments. The feature that may have sheltered this area is found on the Eastern Shore of Maryland. To the east and northeast of Easton, the interval of beds 5 to 9 contains thick brown clay and thin interbedded sand units, which strongly suggest deltaic influence in the center of Well NJ3 Depth (m) (ft) 0 0- brown sand ^.^-■i^— blue clayey sand, shells brown clay, sandy, shelly laminae 15 SO- greenish brown clays and sand interbeds brown clays brown and blue-gray Interbedded clays SO 100- green and brown interbedded clays FIGURE 18.—Geologic column for Well NJ2, near Shirley, Salem County, New Jersey, showing largely deltaic section of Kirkwood Formation. 46 150-. — brown clay Eocene!?) is marked by a disconformity, which may have been produced by either subaerial or submarine environmental conditions. At the base of the Plum Point Marl Member is an oyster (Pycnodonte percrassa) bed (bed 4) containing a fairly diverse foraminiferal assemblage of 22 to 29 species (dom- inated by Cibicides lobatulus, Valvulinaria floridana, and Hanzawaia concentrica) and several percent planktonic Foraminifera in the lower part. This sandy bed (65 to 77 percent sand) is an open marine deposit formed upon an undulating sur- face, probably in depths of less than 30 meters. brown clay with glauconitic sand interbeds 61 200- glauconitic sand FIGURE 19.—Geologic column for Well NJ3, near Shiloh, Cumberland County, New Jersey, showing largely deltaic sedimentation with marine beds near top of section of Kirk- wood Formation. Sample Number Percent Sand Number of Species of Benthic Foraminifera Percent of Planktonic Foraminifera St. Marys Formation 81 ■ 80 I 79 ■ 78 I 77 ■ 76 ■ 75 ■ 74 73 72 71 70 69 68 67 68 65 64 63 Choptank Formation Plum Point Marl Member 39 I 38 ■ 37 ■ 36 I 35 bl 35a ■ 34 ■ Fairhaven Member Popes Creek Sand Member Nanjemoy Formatior FIGURE 20.—Percentage of sand, number of species of benthic Foraminifera, and percentage of planktonic Foraminifera in total foraminiferal assemblage for samples 5-82 from the Baltimore Gas & Electric Company Well at Calvert Cliffs, Maryland. Location of samples in core is shown in Figure 5. NUMBER 53 57 the Eastern Shore area. The southward outbuild- ing of the delta into the embayment from source areas in northeastern Maryland, western New Jersey, and Pennsylvania, with the accompanying freshwater influence, appears to have cut off open ocean circulation to the west in Maryland, result- ing in an area of fine clastic input and normal to slightly brackish water. Vertebrate evidence to support this interpretation as a protected area is seen in the large number of long-beaked por- poises, typical of estuaries, found in these beds (Whitmore, 1971:32). The area under the influ- ence of the delta is shown in Figure 21. Deltaic environments for the middle Miocene strata on the Eastern Shore were suggested by Gibson (1971). The units of brown clay and silty clay and thin interbeds of crossbedded gray sand as much as 5 cm thick are strongly suggestive of delta-front deposition (Reineck and Singh, 1975:321-338). The interbedded sand and clay do not contain mollusks on the Eastern Shore, but some shells are in brown clay interbeds in southwestern New Jersey. Intervals of fossiiiferous green sand and clay occur between the brown clay intervals in wells on the Eastern Shore (Well 224, Figure 14, for example), suggesting that deltaic pulses alternated with more marine pro- delta to inner-shelf environments. Isphording (1970) proposed that the Alloway Clay Member of the Kirkwood Formation in southwestern New Jersey accumulated in a sublittoral environment. In my opinion, the presence of thick sequences of carbonaceous brown and dark gray clay interbed- ded with thin sand suggests a deltaic, delta-front or prodelta, environment. The absence of shells in the brown clay was attributed by Isphording (1970) to solution by sulfuric acid resulting from weathering of the contained pyrite. However, in the cores from New Jersey wells (NJ1-NJ3), the brown clay does not contain shells or molds even though some of the more permeable interbedded sand does. This suggests a primary or early diage- netic absence of shells in the clay and a deltaic rather than marine origin. Farther east in the Hammond Well (Figure 4: loc. 13), the entire Calvert section is composed of fossiiiferous, glauconitic, gray silty clay, charac- teristic of prodelta and shelf deposition. The area of the Hammond Well appears to be seaward from the deltaic influence, as shown in the paleo- geographic reconstruction (Figure 21). Bed 10 is recognizable across much of the embayment in Maryland as a conspicuous shelly sand containing a high diversity of mollusks (65 species in the Calvert Cliffs, Glaser, 1971:21). West of Chesapeake Bay, it overlies beds 5 to 9 that have low molluscan diversities, and on the Eastern Shore it overlies the brown clay units (Figure 14). Even in southwestern New Jersey a fossiiiferous sand is found on top of the brown clay (wells NJl and NJ3, and outcrops near Shiloh). Several shell beds containing diverse molluscan faunas are in beds 10 to 15 of the Calvert, and these beds also contain moderately high foraminiferal species diversities. Abundant benthic Foraminifera through this interval in- clude Valvulinaria floridana, Caucasina elongata, Cib- icides lobatulus, Hanzawaia concentrica, and Episto- minella sp. Planktonic foraminiferal percentages reach 6 percent. These assemblages suggest re- establishment of open ocean circulation, and dep- osition on sandy bottoms in depths of less than 60 meters. However, periods of restricted marine conditions in the western part of the embayment are reflected in beds 11 and 15 in the BG&E Well (Figures 5, 20); macrofossils are rare, foramini- feral diversities are low, and planktonic forami- nifers are absent. A general increase in sand occurs above bed 11 and continues through the Choptank Formation to beds that indicate re- stricted environments in the uppermost part of the Choptank and St. Marys formations. PUNGO RIVER FORMATION In the Albemarle embayment south of Virginia (Figure 1) a different sedimentary regime of a low clastic environment resulted in the deposition of phosphatic sand, diatomaceous clay, and carbon- ate deposits of the Pungo River Formation (Fig- ure 3). As mentioned above, the area of transition between the Calvert and Pungo River formations 58 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 40"- FiGURE 21.—Paleoenvironments postulated for lower half of Plum Point Marl Member of Calvert Formation and equivalent units, showing deltaic influence on Eastern Shore of Maryland and small amount of clastic sediment in the Albemarle embayment in North Carolina. Line indicates western limit of outcrops. NUMBER 53 59 occurs in southeastern Virginia as seen in the Norfolk Moores Bridge Well (Figure 4: loc. 25). The Pungo River Formation was named by Kimrey (1964) for the sequence of phosphatic sand and clay of middle Miocene age described by Brown (1958). The formation underlies much of the eastern part of the Coastal Plain in North Carolina (Figure 3). No natural outcrops of the formation are known, but it is well exposed in the Lee Creek Mine (Figure 4: loc. 33). The Pungo River is more than 400 feet (122 m) thick in wells drilled on the outer banks in the eastern part of the Albemarle embayment. The middle to inner middle-shelf environments of deposition repre- sented by the formation at its western limits indicate that it originally was more widespread. The thickest part of the Pungo River Forma- tion in North Carolina is composed of phosphatic and diatomaceous clay, phosphatic sand, phos- phatic limestone, and coquina. Kimrey (1965) described the lithofacies in the southern part of the embayment. The beds of clay are light to dark green, and many contain diatoms. The phos- phate content varies from less than 1 to nearly 10 percent. The sand is composed of fine to medium quartz and phosphate grains. Phosphate content usually is 10 to 21 percent in bulk samples (Kim- rey, 1965:9-14). The phosphatic grains are com- posed of collophane and are ovate, smooth, glossy, and brown. Sand-sized bone and tooth fragments also are common. Indurated beds of calcareous clay, phosphatic limestone containing dolomite, and indurated shell beds are scattered through the section. Toward the New Bern arch (southern border of the Albemarle embayment), the upper part of the Pungo River Formation changes to calcareous clay, indurated limestone, phosphatic limestone, and bioclastic debris. North of this area, the equivalent strata are mostly beds of phosphatic and diatomaceous clay with minor phosphatic sand intervals. I herein propose that this upper carbonate unit, including phosphatic limestone, calcareous clay, and coquina in the southern part of the embay- ment, be recognized as the Bonnerton Member of the Pungo River Formation. The type section is the AU-l-GRL core hole in the Aurora quad- rangle, described by Kimrey (1965:17) (Figures 4: loc. 33; and 22). In this core hole, 32 feet (10 m) of white to light gray-green calcareous sand and phosphatic limestone occur from 120 to 152 feet (37 to 46 m). The name is derived from Bonnerton, Beaufort County, North Carolina, where the unit is typically developed. The Bon- nerton is unconformably overlain by the York- town Formation; the contact is marked by bur- rows and small channels containing coarse, black, secondary phosphatic gravels. The lower bound- ary of the Bonnerton is marked by a change to greenish brown phosphatic sand, which contains phosphatic clay and limestone layers. Twelve feet (3.7 m) of Bonnerton was exposed in the initial test pit of Texasgulf Inc., in the northeastern part of the Lee Creek Mine site north of Aurora (Figures 23, 24). This includes units 4 to 7 of Gibson (1967). Unit 7 is a yellow- green sandy coquina; unit 6 is interbedded bio- clastic debris and phosphatic sand; unit 5 is highly fossiiiferous phosphatic limestone; and unit 4 is interbedded phosphatic limestone and phosphatic sand. The lower contact is conforma- ble, and is at the base of the lowest, light-colored limestone bed in Figures 24 and 25. The upper, unconformable contact with the Yorktown For- mation is seen in Figures 23, 24, 26, and 27. As shown by Kimrey (1965, fig. 6), the Bon- nerton Member is extensive in the southern part of the embayment and reaches northwestward as far as Bonnerton. To the west of Bonnerton, apparently, this member was completely eroded away. The southernmost locality is in the Croatan Forest area near Great Lake (Figure 4: loc. 36). Here the Bonnerton consists of about 6 feet (1.8 m) of light yellow-green sandy shell hash in Great Lake Well 181. The Bonnerton thus extends at least 40 miles (64 km) north to south. North of Bonnerton, across the Pamlico River, the carbon- ate facies changes to beds of phosphatic and diatomaceous clay. The Pungo River strata below the Bonnerton Member are dominantly phosphatic sand and moderately phosphatic to nonphosphatic clay, but include thin interbeds of diatomaceous clay 60 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Depth (m) (ft) 36 120- 40 130- 42 140- Well AU-l-GRL l?^?3C=y r^m ■^ .^:: h: llght green diatomaceous clay light green calcareous clay, shell Tragments light grav linnestone sandy, phosphatic sand, clayey, shell fragments calcareous clay and limestone greenish-white phosphatk: limestone no sample gray sand shell fragments Bonnerton Member FIGURE 22.—Geologic column for Well AU-l- GRL, near Aurora, Beaufort County, North Carolina; type well for Belhaven and Bonnerton members. Note largely phosphatic sand section in Belhaven Phosphatic Sand Member and limestone, calcareous clay, and shelly sand in Bonnerton Member. •fe; 46 150- clay, sand, and shell calcarepcs clays, shelly phosphatic limestone greenish-brown phosphatic sand no sample 49 160- greenish-brown fine-grained phosphatic sand Pungo River Formation 52 170- 55 1 BO- greenish-gray phosphatic sand sandy phosphatic limestone phosphatic sand and calcareous clay Belhaven Phosphatic Sand Member SS 190- no sample phosphatic limestone phosphatic clay and sand olive-green calcareous clay 61 200- -I— I • no sample phosphatic sandy limestone greenish-brown phosphatic sand 64 210- :x_i sandy limestone Castle Hayne Limestone NUMBER 53 61 I >v ^<:^- ^i'^^"- '> ^'"^ IS**''^ '^-" FIGURE 23.—Test pit at Texasgulf Inc. Lee Creek Mine. Water level is at middle of Pungo River Formation with 96 feet (29 meters) of section exposed. Arrow near right margin indicates contact between Pungo River and Yorktown formations in face; close-up of face is shown on Figure 24. Two clean faces in center above lowest bench are in lower and middle parts of the Yorktown Formation. Arrow in upper center indicates boulder beds or unit 7 of Gibson (1967), now considered in lower part of Croatan Formation. and phosphatic limestone. I herein assign these highly phosphatic beds to the Belhaven Phos- phatic Sand Member of the Pungo River For- mation. It is so named because it is well developed near Belhaven, Beaufort County, North Carolina (Figure 4: loc. 32). The type section is the same core hole, AU-l-GRL (Kimrey, 1965:18-19; Fig- ure 22), that constitutes the type section of the Bonnerton Member. The Belhaven is 58 feet (18 m) thick (152 to 210 feet (46 to 64 m) in the core hole), and is dominantly medium greenish brown phosphatic sand with some gray-green clay and limestone and dolomite beds. The upper part of this member was exposed in the Lee Creek Mine test pit (Figures 24, 25). The uppermost bed in this member is immediately below the limestone in Figure 25. A slightly indurated, dolomitic, diatomaceous clay bed can be seen at water level in the same figure. The phosphatic sand bed at the top of the Belhaven Member probably is the highest in P2O5 content over the area. The upper contact of the Belhaven is conformable, but the lowermost part is underlain by the Castle Hayne Formation (Eocene). Thus, in the southern part of the basin the Bonnerton Member composes most of the section of the Pungo River. To the north near the Lee Creek Mine and northward toward Belhaven, both members are well developed, and northeast of this area most of the formation is composed of the Belhaven Member. Exact biostratigraphic control is not yet available, but the relationship 62 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY I ■ .•;■•:•- .'•;■. '■ ■••■;/'-'. yj^s^^^ ?■««.,..>--"•• •/ • .■■ '' .'■'?>*H y S3 ^^rj^.-(^--':f^: ^ ^#lf /j-^^-^':r-- FIGURE 24.—Largely carbonate beds of the upper part of the Bonnerton Member of the Pungo River Formation overlain by Placopecten clinlomus beds of the lower Yorktown Formation exposed in north wall of Lee Creek Mine test pit. Arrow indicates unconformity between the two formations. Three channels filled with phosphatic pebbles are visible. Numbers on right border correspond to units used in Gibson (1967). NUMBER 53 63 FIGURE 25.—Contact between dark phosphatic sands of the Belhaven Phosphatic Sand Member and largely carbonate strata of the Bonnerton Member of Pungo River Formation in Lee Creek Mine test pits is at base of lowest light-colored carbonate unit about six feet (1.8 meters) above water level. Two feet (0.6 meters) of diatomaceous clay (unit 2 of Gibson, 1967) of the Pungo River Formation are exposed at water level. Contact between Pungo River and Yorktown formations is at lowest bench, a distance of approximately 20 feet (6 meters) above water level. indicates that there is a partial to entire facies equivalency between the two members in differ- ent parts of the basin. The lithology of the Pungo River changes northward in the Albemarle embayment. The limestone beds diminish and finally disappear, and the phosphatic sand is thinner, more clayey, and less phosphatic. In the core hole at Gatesville, North Carolina (Figure 4: loc. 27), the Pungo River Formation is only 20 feet (6 m) thick, and consists mainly of yellowish green to greenish brown calcareous, phosphatic, clayey sand. The Pungo River Formation continues northward into the southeastern part of the Salisbury em- bayment. Here in the Norfolk Moores Bridge Well, the upper 50 feet (15 m) is olive-green silty clay; the lower 10 feet (3 m), dark-olive phos- phatic clayey sand. Gibson (pp. 359-360) demonstrates that the Pungo River strata in the Lee Creek Mine belong to planktonic foraminiferal zone N8 to lower N9 of Blow (1969:229-234) that are of latest early and earliest middle Miocene age. To the north- east, Abbott and Ernissee (pp. 290-293) found Pungo River strata of this age and also of the younger, zone Nl 1 age. Strata of the intervening zone NIO have not been documented in the area. Whether this is because the important index spe- cies of this relatively short time zone are absent, or whether this is a time of regression in the embayment represented by a disconformity is unknown. ENVIRONMENT OF DEPOSITION.—The deposi- tional environments of the Pungo River Forma- tion were discussed by Gibson (1967) and were determined on the basis of the foraminiferal as- semblages (Gibson, 1968). Gibson concluded that the beds of phosphatic sand of the Belhaven Member formed on the middle to outer shelf (approximately 100- to 200-m water depth). The carbonate and phosphatic clay beds in the Bon- nerton Member formed on the middle to inner 64 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 26.—Unconformable contact (arrow) between Pungo River (light-colored) and York- town (dark-colored) formations in the test pit of the Texasgulf Inc. Lee Creek Mine. Channels in bryozoan hash bed at top of the Bonnerton Member of the Pungo River Formation are filled with phosphate pebbles. Lowermost Yorktown strata contain abundant Placopecten clintonius and medium sand to cobble-sized phosphate grains. Close-up of channel at left is shown in Figure 27. shelf (150 m to less than 70 m in the uppermost bed, Gibson and Buzas, 1973:232). In addition to the faunal evidence, the nature of the phosphatic sand, the organization of its beds extending tens of kilometers, and the in- terbedded limestone and diatomaceous clay led Gibson (1967:638, 642) to consider this phosphate deposit a primary deposit from seawater. The framework of deposition, particularly the depth of water, fit the Kazakov (1937) hypothesis of inorganic precipitation that has been used to explain numerous large phosphatic deposits. A weakness in this hypothesis is that the phosphate mineral found in these deposits does not com- monly form in the laboratory (McConnell, 1958). The presence of large amounts of phosphatic fish bones and teeth may indicate that the postulated upwelling or mixing of currents and resultant high productivity biologically concentrated unu- sual amounts of phosphatic material in the sedi- ments. Another possible method is phosphate replacement of existing carbonate material. Al- though replacement may explain the thin layers of phosphate in the top of underlying carbonate units, such as those in the Castle Hayne Forma- tion, this origin for the major deposit has serious complications, because the molluscan fossils in the phosphatic sand are phosphatic internal molds. This manner of preservation indicates that the shell was still present when the phosphate material was introduced and that the shell was filled, not replaced. The ovoid shape of many of the phosphate pellets is similar to that of fecal pellets; these pellets may have formed by the NUMBER 53 65 addition of phosphate. We do not know the exact genesis of these deposits, but evidence to date strongly indicates that they are primary deposits from a middle- to outer-shelf environment. Another characteristic of the Pungo River strata is the small influx of coarse clastic sedi- ments. The offshore deposits are primarily phos- phatic sand and diatomaceous clay that become more calcareous landward, indicating little clastic influx into the Albemarle embayment from that direction. Thus, the uplift of the Appalachians that affected the northern part of the Coastal Plain at this time, as seen in New Jersey, is not evident here. The presence of clinoptilolite, volcanic glass, shards, and cristobolite in the Pungo River For- mation led Rooney and Kerr (1964) and Gibson (1967) to suggest a volcanic influence in the depositional area. The importance of this influ- ence still is largely unknown. However, this time interval is characterized by unusual deposits all along the Atlantic Coast. The deposits include diatomaceous clay containing shards and clinop- tilolite in the Fairhaven Member of the Calvert Formation in Maryland (Taliaferro, 1933:28; Glaser, 1971:23) and Virginia (Reinhardt, New- ell, and Mixon, 1980); phosphatic and diatoma- ceous clay and sand in North Carolina; and montmorillonite, cristobolite, and attapulgite in South Carolina and Georgia (Ernissee, Abbott, and Huddlestun, 1977). The only region appar- ently unaffected at this time is New Jersey, which was under a strong deltaic influence. The possi- bility was raised by Gibson and Towe (1971) that widespread siliceous deposits of a short timespan could be a result of marine volcanism, which influenced productivity through the addition of nitrogen and phosphorus into the water column. The increase in the levels of these nutrients, in addition to the added silica, could result in abun- dant siliceous organisms in the area, and thus increased deposition and enhanced preservation. Questions have been raised about the role of volcanism in these areas (Weaver and Wise, 1974; FIGURE 27.—Unconformable contact between Placopecten clintonius beds of Yorktown Formation and bryozoan hash beds of the Bonnerton Member of the Pungo River Formation. Coarse phosphate pebbles fill channels in top of Pungo River; generally finer pebbles occur in shell bed of the Yorktown. Pen giving scale is 5 inches (12.7 cm) long. 66 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Gibson and Towe, 1975). Weaver and Wise stressed the finding of siliceous organisms as proof that volcanic influence was not present or impor- tant. However, this missed the main point of the model proposed by Gibson and Towe (1971) that the volcanism largely would cause increased pro- ductivity of siliceous organisms in many environ- mental settings. The finding of these unusual deposits over a relatively short timespan of a few million years requires some explanation. One new piece of evidence, which indicates volcanic influ- ence in the area, is the presence in the Pungo River Formation of large pieces of pumice, as much as 6 inches (15 cm) in diameter. These pieces have been uncovered during washing of the ore at the Lee Creek Mine. The pumice currently is being analyzed to determine the pos- sible origin and significance. CHOPTANK FORMATION The Choptank Formation was named by Shat- tuck (1902) from exposures along the Choptank River on the Eastern Shore of Maryland. Most subsequent work has been more concerned with the extensive exposures along the Calvert Cliffs on the western shore of Chesapeake Bay. The outcrop thickness varies from 45 to 55 feet (14 to 17 m) (Shattuck, 1904, pis. xxx, 5). The contin- uous BG&E core contains approximately 60 feet (18 m) of Choptank. Thicker sections occur in the subsurface to the east. The Hammond Well on the Eastern Shore of Maryland (Figures 4: loc. 13, and 6) contains 125 feet (38 m) (Anderson, 1948:19), and 150 feet (46 m) was found in Well 220 to the southwest of the Hammond Well (Figures 4: loc. 12 and 7); this appears to have been in the axis of the embayment during Chop- tank time. The Choptank Formation in outcrop charac- teristically is quartz sand and silt, shelly in many places, with lesser clayey intervals and indurated limestone layers. The BG&E Well contains mostly muddy and silty fine sand (Figure 5; generally 50 to 70 percent sand. Figure 20). Minor silt and clay intervals are present, along with two indur- ated layers. Two intervals contain numerous mol- luscan shells (beds 17 and 19 in outcrop), and shells occur sparsely in the rest of the formation. Outcrops in the type area of the Choptank are similar to those found west of the Bay. However, two wells on the Eastern Shore contain a different facies of the Choptank Formation. Well 225 (Fig- ure 15), northeast of the type area, contains mostly brown micaceous clay interbedded with scattered shell hash; the rest is green to blue, shelly, clayey sand containing quartz and phos- phate pebbles, thickly interbedded with brown clay and gray sand laminae. In Well 227 (Figure 12), green sandy clay containing shells and sandy shell hash composes most of the formation, but the middle part contains interbedded shell hash and brown clay. To the southeast, in the Ham- mond Well (Figure 6), the Choptank is composed of limy, sparsely glauconitic, shelly, medium sands. Shattuck (1904) divided the Choptank For- mation into a series of zones (here termed "beds") numbered 16 through 20. As in Shattuck's divi- sion of the Calvert Formation, these zones are locally distinguishable by lithologic, not biostra- tigraphic, characteristics. One zone, was defined, however, by the abundance of fossils, particularly mollusks. Gernant (1970) mapped these "zones" in the Choptank Formation and redefined them as lithologic members. The nature of the upper contact of the Choptank is in dispute because Blackwelder and Ward (1976:12, 15) placed Ger- nant's uppermost member, the "Conoy," into an overlying unit, which they informally referred to as the "Little Cove Point" unit. The distribution and thickness of the Choptank Formation is considerably restricted compared to the underlying Calvert Formation. The known distribution extends into northeastern Virginia as shown in Figure 28. In the Oak Grove core hole (Figure 4: loc. 19), located at approximately the known southern limit of the Choptank, slightly diatomaceous clay and silt were found that cor- related with beds 18 and 19 of the Choptank Formation (Gibson et al., 1980). The Choptank strata of bed 18 and 19 age in the Oak Grove core sit upon the upper part of the Calvert Formation, signifying the southward disappearance of the NUMBER 53 67 PENNSYLVANIA FIGURE 28.—Isopachous map of the Choptank Formation and equivalent units (middle Miocene). Contours are in feet; contours are indicated by dashed lines where locations are approximated (modified from Gibson, 1970). lower beds 16 and 17 of the Choptank that still are present at the Nomini Cliffs on the Potomac River. No indication of Choptank-age strata has been found in southern Virginia and North Car- olina. The northern extent of the formation in New Jersey is uncertain because Richards and Harbison (1942) treated the Calvert and Chop- tank formations as a single unit. Because the Choptank Formation was depos- ited in very shallow marine water, its present distribution probably closely reflects the original extent of the Choptank sea (Figure 28). The distribution was restricted by uplift in various areas, including the Albemarle embayment in North Carolina, causing a markedly reduced areal extent of the Choptank seas. ENVIRONMENT OF DEPOSITION.—The Choptank Formation was deposited mostly in shallow marine water (Gibson, 1962:63). Gernant (1970:45-52) also proposed that environments were less than 60 meters deep for the formation as a whole, but they were less than 25 meters deep, and possibly marginal marine, for some beds. Foraminiferal assemblages in the BG&E Well support these interpretations. Foraminiferal species diversity is less than 20 in all samples (most samples contained about 15), and the planktonic foraminiferal content varies from zero to less than 1 percent (Figure 20). These measures of foraminiferal abundance are similar in about 100 samples examined from other localities. On the Eastern Shore in the type area, similar shallow water deposits persist and shell hash is common. Gernant (1971:28) suggested that here much of the formation was deposited in shallower water than that west of Chesapeake Bay. The sediments usually are more than 50 percent sand, which, along with the foraminiferal data, suggests dep- osition in open shallow marine water of 15 to 30 meters depth or less. On the Eastern Shore, evidence of deltaic in- flux into the eastern and northeastern part of the Choptank Formation is present in the subsurface. Well 225 (Figures 4: loc. 8, and 15) contains thick intervals of brown micaceous clay and interbeds of pebbly shell and gray sand that suggest delta- front deposition. To the southwest. Well 227 (Fig- ures 4: loc. 9, and 12), farther away from the delta front, contains mainly shell hash and in- terbeds of brown clay in the middle part. In the Hammond Well (Figures 4: loc. 13, and 6). Chop- tank strata are limy and shelly medium-grained sand containing small amounts of glauconite, indicating shallow sublittoral deposition. The green shelly sand and shell hashes in wells 224 and 226, and the Hammond Well suggest that these strata accumulated in prodelta or inner- shelf environments; they mark the southern limit of deltaic influence (Figures 4, 29). The deltaic outbuilding at this time on the Eastern Shore did not significantly influence deposition in the west- ern part of the embayment, because diverse mol- luscan, ostracode, and foraminiferal assemblages dominated by Cibicides lobatulus, Bolivina paula, Valvulinaria floridana, and Buliminella elegantissima indicate shallow open marine water of normal 68 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 29.—Paleoenvironments postulated for the Chop- tank Formation and equivalent units, showing deltaic influ- ence on Eastern Shore of Maryland and open marine shallow shelf in remainder of embayment. Line indicates limits of formation; line is dashed where limits are inferred. salinity. More open access to the Atlantic Ocean must have prevailed at that time than in middle Calvert time. ST. MARYS FORMATION The St. Marys Formation was named from exposures in St. Marys County, Maryland, par- ticularly those along the St. Marys River near the city of St. Marys (Shattuck, 1902). Other outcrops in St. Marys County are known along the south- ern bank of the Patuxent River and along the western shore of Chesapeake Bay. Outcrops also extend farther north along Chesapeake Bay into Calvert County. A few outcrops of equivalent age strata are in northern Virginia (Ward, pers. comm., 1971); otherwise, the St. Marys in north- ern Virginia and on the Eastern Shore is known only in the subsurface. The St. Marys Formation is geographically restricted as is the underlying Choptank Formation in Maryland and northern Virginia (Figure 30). Shattuck (1904:lxxxv) divided the St. Marys into lithologic "zones" 21 to 24; the lower three zones are exposed in southern Calvert County and the upper one crops out in St. Marys County. A total thickness of 74 feet (23 m) for the four units combined was estimated (Shattuck, 1904: Ixxxv), but no continuous outcrop section or sec- tions exist. As much as 185 feet (56 m) is now known in the Hammond Well (Anderson, 1948:19) on the Eastern Shore (Figure 6). The outcropping St. Marys west of Chesapeake Bay comprises beds of blue clay, sandy clay, and clayey sand, commonly containing shell layers. Clay is more abundant than in the underlying Choptank Formation. The lower percentage of sand is typified in the BG&E Well, where sand is FIGURE 30.—Isopachous map of the St. Marys Formation and equivalent units (upper Miocene). Contours are in feet; contours are indicated by dashed lines where locations are approximated (modified from Gibson, 1970). NUMBER 53 69 less than 50 percent in the lower part of the formation and less than 10 percent in the middle and upper parts (Figure 20). In the Hammond Well (Figure 6) on the Eastern Shore, the strata placed in the St. Marys are coarse sand and fine gravel containing shell fragments and some glau- conite. ENVIRONMENT OF DEPOSITION.—The St. Marys Formation was deposited in a shallow-marine to marginal-marine basin (Gibson, 1962:65; Ger- nant, 1971:28-30). Gernant (1971) suggested a shallow-subtidal to marginal-marine origin for all beds. In the BG&E core, the low species diversity of 13 or less, dominance of the assemblages by Buliminella elegantissima, Buccella mansfieldi, and Cibicides lobatulus, and the absence of planktonic foraminifers in most samples suggest depths of less than 30 meters for the marine beds. Organic rich brownish gray and dark gray clay beds con- taining only 2 percent sand or less suggest brack- ish to restricted marine deposition for at least part of beds 21 and 23. Bed 24, which constitutes the upper part of the St. Marys along the St. Marys River, contains richly fossiiiferous sand and some interbedded nonfossiliferous clay. The shelly units are dominated by a relatively few molluscan species, primarily gastropods. Forami- niferal species diversities range from 15 to 20; the benthic assemblages are dominated by Buccella mansfieldi, Florilus pizarrense, Buliminella elegantis- sima, and Quinqueloculina seminula. Planktonic for- aminifers usually form less than 1 percent of the assemblages, although one sample contains 9 per- cent. These foraminiferal assemblages suggest deposition in shallow marine conditions at depths of less than 30 meters. The nonfossiliferous clay probably represents restricted marginal-marine conditions. The St. Marys strata in the Hammond Well on the Eastern Shore consist of coarse sand and fine gravel containing shell fragments. In New Jersey, the St. Marys fossils occur in beds followed by thick sequences of coarse clastic deposits (Rich- ards and Harbison, 1942:171), which Isphording and Lodding (1969) and Isphording (1976) con- sidered to overlie the Kirkwood Formation con- formably. These medium- to coarse-sand and gravel beds, usually referred to the Cohansey Sand, suggest, as do those of the Hammond Well, that clastic debris was carried in by the same delta system that had been building in the area since middle Calvert time. The large influx of coarse clastic deposits at this time indicates uplift of the Appalachian source area as Gibson (1971) suggested. In the embayment to the west of this rapidly prograding delta sequence, open ocean circulation was cut off for significant periods of time, and deposits of restricted marine to brackish environments were formed as seen in beds 21 and 23 and part of 22. Open marine circulation must have been temporarily re-established during dep- osition of parts of bed 24 slightly to the south. "VIRGINIA ST. MARYS" BEDS The "Virginia St. Marys" beds represented a transition in Miocene depositional patterns in the Salisbury embayment as the axis of depositon moved south of that found in the Choptank and St. Marys formations. Mansfield (1943) divided the St. Marys outcrops in Virginia into three units. Mansfield questionably correlated the low- ermost, stratum A, with beds 21 and 22 of the St. Marys in Maryland. Stratum A is an unfossilifer- ous silty clay found only in northernmost Vir- ginia, and is presumably a marginal-marine or nonmarine unit occurring at the southern extrem- ity of the St. Marys Formation. The overlying unit, termed "zone 1" by Mansfield (1943:6) was correlated with beds 23 and 24 of the St. Marys in Maryland. Zone 1 extends farther south than stratum A, reaching approximately the Rappa- hannock River. Mansfield believed that his high- est unit, zone 2, was younger than the fossiiiferous St. Marys in Maryland. The distribution of zone 2 reflects the continued southward shift in basin location; the zone as reported by Mansfield (1943;6) reached no farther northward than the Rappahannock River, and it extended southward to the James River in southern Virginia. My subsequent field investigations show that zone 2 extends to northern Virginia at the Nomini Cliffs 70 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY MARYLAND'W'jf 100 Miles leoKllometers on the Potomac River and even farther south into northeastern North Carolina along the Meherrin River in the vicinity of Murfreesboro (Figures 4: loc. 28; and 31). The foraminiferal assemblage of zone 2 strata at the Nomini Cliffs indicates a shallow marine environment. Even shallower ma- rine or marginal-marine deposits probably con- tinued into southern Maryland. Most of these strata probably have been stripped by erosion, but possibly some marine and probably some marginal-marine deposits of this age may be found as outliers in Maryland (Stephenson and MacNeil, 1954). As Mansfield's zone 2 of the St. Marys is younger than the type St. Marys, and occurs largely in Virginia, I have used the informal term "Virginia St. Marys" beds to differentiate zone 2 beds from the type St. Marys Formation in Mary- land (Gibson, 1971). Because the geographic dis- tribution and lithology of the "Virginia St. PENNSYLVANIA NEW , JERSE'Bo FIGURE 31.—Isopachous map of the "Virginia St. Marys" beds (upper Miocene). Contours are in feet and are indicated by dashed lines where locations are approximate. FIGURE 32.—Isopachous map of the Yorktown Formation and equivalent strata (contours are in feet): 1 = Cohansey Sand, 2 = Yorktown Formation and "Virginia St. Marys" beds, 3 = Duplin Formation. Formational limits are dashed where approximately located. The dotted line is the approx- imate location of the change between the latter two forma- tions (modified from Gibson, 1970). Marys" beds (Figure 31) are similar to those of the immediately overlying Yorktown (Figure 32), and because they occupy the same depositional basin, I placed the "Virginia St. Marys" beds strata into the basal part of the Yorktown For- mation (Gibson, 1971). Although the "Virginia St. Marys" beds are without question slightly older than the type Yorktown, the units have a similar genesis. This placement of these strata is similar to that by Olsson (1917:3) who recognized an intermediate fauna between the St. Marys and Yorktown faunas and termed this the "Murfrees- boro" stage. Mansfield (1943) and Olsson (1917) placed these beds in the lower part of the Yorktown Formation. The stage name "Murfreesboro" is invalid, however, because of prior usage. NUMBER 53 71 The "Virginia St. Marys" beds are grayish blue to greenish blue clayey sand and sandy clay, commonly very shelly. The beds are 25 to 75 feet (7.6-23 m) thick in the outcrop belt in Virginia and thicken down-dip to more than 200 feet (61 m) in the Norfolk Moores Bridge Well. They generally sit upon the Calvert Formation in the Salisbury embayment in Virginia. In North Car- olina, the "Virginia St. Marys" beds crop out near Murfreesboro (Figure 4: loc. 28). Strata to the east and southeast of Murfreesboro in the subsurface are placed in this unit. They contain a distinctly different foraminiferal assemblage, reflecting a deeper water environment, which is difficult to correlate with the microfaunas of out- crop sections. The exact time of the "Virginia St. Marys" beds is uncertain because of the absence of diag- nostic planktonic foraminifers in outcrop samples. In addition to the late Miocene age for the un- derlying St. Marys Formation, a late Miocene age is suggested by K/Ar dates of 8.7±0.4 my and 6.46±0.15 my on "Virginia St. Marys" beds by Blackwelder and Ward (1976:5). The upper age limit of latest Miocene to earliest Pliocene is drawn from the planktonic foraminiferal place- ment of the lowermost part of the Yorktown Formation into zone N19 (see p.363) and by the K/Ar date of 4.4±0.2 my on lower beds of the Yorktown (Blackwelder and Ward, 1976:8). That considerable late Miocene time is represented by the "Virginia St. Marys" beds is documented by the presence of three consecutive pecten range zones in these strata (Gibson, in prep.) Hiatuses are not unexpected in strata of shallow-marine origin, and at least three are present in the "Virginia St. Marys" beds. This interpretation arises from the sudden appearance of three dif- ferent chronologic subspecies of pectens above undulating surfaces accompanied by slight changes in the sediments. These zones and hia- tuses are best exposed in the James River Valley from Cobham Bay westward to Petersburg. ENVIRONMENT OF DEPOSITION.—Initial deposi- tion in the new center of deposition of the Salis- bury embayment constitutes the "Virginia St. Marys" beds. Deposition of the lower strata took place in shallow marine water of less than 30 meters as interpreted from the low species diver- sities of benthic Foraminifera (less than 15), the dominance of Elphidium excavatum, and the spars- ity of planktonic foraminifers. The distribution of these strata in Virginia is limited to the eastern part of the embayment. Equivalent strata are not known in the Albemarle embayment, which sug- gests that no significant subsidence took place there. Beds of the Chesapecten middlesexensis middlesex- ensis total range subzone overlie the initial depos- its of the "Virginia St. Marys" beds and form much of the middle and upper parts of Mans- field's (1943) zone 2 of the St. Marys in Virginia. These strata still represent shallow-marine envi- ronments of less than 30 meters and contain foraminiferal assemblages similar to those of the underlying beds. It is thought that they accu- mulated during a transgression in the Salisbury embayment, because they are found beyond the known geographic extent of the initial strata of the "Virginia St. Marys" beds. In the highest part of the "Virginia St. Marys" beds, the transgressive sea extended considerably westward, reaching Petersburg (Lieutenant Run, Figure 4: loc. 23) and northward to the Nomini Cliffs (Figure 4: loc. 18). Deposition took place in shallow open marine water of less than 30 meters; the benthic foraminiferal assemblages have spe- cies diversities of 10 to 20 with Elphidium excavatum dominant, and few planktonic foraminifers occur. These strata contain a transitional form of Ches- apecten middlesexensis-C. jeffersonius. Significant downwarping of the northern part of the Albe- marle embayment began at this time because deposits of this age extend to the west of Halifax, North Carolina (Figure 4: loc. 29). Around Hal- ifax, deposition occurred in shallow open marine water of 15 to 30 meters, marked again by the low species diversities of 12 to 20, few planktonic Foraminifera and dominance of Elphidium exca- vatum. To the east, near Murfreesboro, deeper water deposits are found; the species diversity of benthic Foraminifera increases to 45 and plank- tonic specimens increase to 5 to 8 percent. These beds accumulated in open marine water about 30 72 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY to 60 meters deep. According to Bailey (1973:53) these beds, which contained his assemblage I, formed in depths of 22 to 120 meters. YORKTOWN FORMATION The Yorktown Formation and its laterally equivalent formations represent one of the two most widespread transgressions of the Neogene seas in the central Atlantic Coastal Plain, the other having occurred during deposition of the Calvert Formation and equivalents. The York- town Formation is probably the most widespread of the Neogene transgressive units in the central and south-central parts of the Atlantic Coastal Plain. The distribution of the Yorktown Formation and equivalent strata is shown in Figure 32; the distribution of the subdivisions of the Yorktown is shown in Figures 33 through 35. The Yorktown FIGURE 33.—Map of the distribution of the lower Pliocene zone 1 or Placopecten clintonius zone of the Yorktown Forma- tion of Mansfield (1943). Limit of zone is indicated by a dashed line where approximately located. FIGURE 34.—Map of the maximum distribution of the lower and middle parts of zone 2 of the Yorktown Formation of Mansfield (1943) and equivalent strata (lower and upper? Pliocene). 1 = Manfield's Yorktown Formation, 2 = Duplin Formation; dotted line is the approximate location of change between formations. reaches thicknesses of greater than 300 feet (91 m) near Norfolk, Virginia, and probably also beneath the outer banks of North Carolina. The lithology of the Yorktown Formation is variable because the wide geographic distribution reflects a number of environments. Clastic units dominate northern Virginia and southern North Carolina, and indicate that the Piedmont uplift had proceeded southward by this time. Bluish clayey sand is dominant; bluish sandy clay is secondary; molluscan shells are common in most strata. The weathered deposits, particularly the sandier, more permeable, units are buff to yellow. Beds of shell hash, in which broken shells compose well over 50 percent of the sediment, are common and many are crossbedded, representing offshore bars and associated environments. Because these NUMBER 53 73 strata are very permeable, most are highly oxi- dized and some are indurated. The distribution of the Yorktown Formation continued the significant change in depositional pattern seen in the underlying "Virginia St. Marys" beds. The locus of deposition shifted southward from southern Maryland to southern Virginia as the western shore of Maryland was uplifted. This trend toward a more southerly center of deposition began in the "Virginia St. Marys" beds and continued upward through the formation until southern Virginia (including the Norfolk arch) and northern and central North Carolina were covered by the Yorktown sea. The latest "Yorktown Formation" transgression and its equivalents (Waccamaw and Croatan forma- tions) accumulated in a more restricted area cov- ering only the southeastern part of Virginia and VIRGINIA NORTH CAROLINA Beds at Suffolk, witji bed at Uppermost Yorktown with bed at Mt. Gould Duplin marl Biggs Farm at at top top TION one 2 < 2 Beds at Equivalent to the beds at Yorktown Yorktown, Va OWN FO ^ Chama-bearing Chama-bearing bed bed > Zone 1 Zone 1 ST. MARYS Zone 2 ST. MARYS FORMATION (? ) FORMATION Zone 1 r?) FIGURE 35.—Map of the distribution of the upper Pliocene and lower Pleistocene strata. Limit of strata is indicated by a dashed line where approximately located. 1 = "Yorktown" Formation, 2 = Croatan Formation, 3 = James City For- mation of DuBar and Solliday (1963) (=Croatan), 4= Wac- camaw Formation. FIGURE 36.—Correlation chart of the Yorktown and St. Marys formations according to Mansfield (1943). eastern North Carolina. The northern part of the central Coastal Plain from Maryland to southern Virginia was then a positive area. The changes in distribution reflect upwarp in the central and northern Salisbury embayment in Maryland and northern Virginia and signifi- cant downwarping in southern Virginia and the Albemarle embayment in North Carolina. The present definition of the Yorktown For- mation does not imply continuity of deposition in all outcrop sections. The Yorktown Formation as used herein accumulated during early Pliocene through late Pliocene(?) time (Hazel, 1977; see also p.363). Faunal changes and lithologic discon- tinuities are present in the Yorktown Formation as documented by Gibson (1967) and Hazel (1977), and herein illustrated in Figures 24 and 26. Mansfield (1943) divided the Yorktown For- mation in Virginia into two faunal zones, the lower, zone 1, called the Pecten clintonius zone, and the upper, zone 2, called the Turritella alticostata zone. Zone 2 was divided into three units, in ascending order, the C/zawa-bearing bed, the beds at Yorktown, and the beds at Suffolk (Figure 36). 74 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY At the top of the Suffolk unit, he placed the bed at Biggs Farm (exposure not located by later workers), which he considered to be the youngest part of the Yorktown in Virginia. Blackwelder and Ward (1976) proposed four member names for the Yorktown Formation in Virginia that encompass Mansfield's zone 1 and much of his zone 2. These four members are recognizable in the areas of the James and York rivers. Johnson (1969:10) proposed that some of the lithologically distinct beds or members of the Yorktown along the James and York rivers are lateral facies, but biostratigraphic correlation based on the pectens (Gibson, in prep.) supports Mansfield's (1943) view of a vertical succession of these beds. Mansfield (1943) studied the Yorktown strata in North Carolina and compared them with the Virginia succession. The youngest part of the Yorktown in North Carolina recognized by Mansfield was at Mt. Gould Landing along the Chowan River. Mansfield (1943:12) concluded that the strata were younger than any Yorktown in Virginia. This conclusion is supported by Ha- zel (1977) on the basis of ostracodes and by Gibson (in prep.) on the basis of pectens. Gibson (1962) correlated the beds along the Chowan River with the Waccamaw Formation in southern North Carolina on the basis of benthic Forami- nifera; this correlation is supported by Hazel (1977) and Gibson (p.364). These beds are further correlated with the Croatan Formation as used in the Lee Creek Mine and other places in east- central North Carolina on the basis of the ostra- codes (Hazel, 1977), foraminifers (see p.364), and pectens (Gibson, in prep.). Hazel (1977) found strata at Yadkin, Virginia, which he considered younger than any other strata in the Yorktown Formation in Virginia, and probably equal in age to the upper beds along the Chowan River. Strata containing benthic Foraminifera characteristic of latest Yorktown age also occur in the upper part of the Moores Bridge Well at Norfolk (Figure 4: loc. 25) at depths from about 90 to 115 feet (27 to 35 m). These strata are likely equivalents of the Yadkin deposits, and both may be the same age as those originally reported by Mansfield (1943) at Biggs Farm. Beds equivalent to zone 1 of Mansfield's York- town, characterized by the presence o{ Placopecten clintonius, crop out only near Murfreesboro in North Carolina (Figure 4: loc. 28), but they are present in the subsurface southward to the Lee Creek Mine (Figures 23, 24, 27); they also extend to the east (Figure 33). Strata equivalent to zone 2 of Mansfield's zonation extend southward in North Carolina to New Bern (Figures 4: loc. 34; and 34). South of central North Carolina, strata of Yorktown age are placed in the Duplin Forma- tion (Mansfield, 1943) .The historic basis for sep- arating Duplin from Yorktown is the warmer water fauna of the Duplin rather than any strik- ing lithologic change (Mansfield, 1943). Accord- ing to Mansfield (1943:13), the Duplin Formation represents only the younger part of his zone 2 of the Yorktown. The diagnostic species of Duplin pectens and benthic foraminifers confirm its mid- dle and late Yorktown age (see p.363). However, the youngest parts of the Yorktown Formation as exposed along the Chowan River in North Car- olina are younger than the Duplin Formation. In the Salisbury embayment, more than 200 feet (61 m) of Yorktown strata are present in wells in the southern part of the Delmarva Peninsula (Accomack Well, Virginia; Figure 4: loc. 20). Farther north on the Delmarva Peninsula, An- derson (1948:19) provisionally placed 200 feet (61 m) of sand and gravel described from the Ham- mond Well into a nonmarine facies of the York- town Formation. No biostratigraphic data are available to confirm this, however. Richards and Harbison (1942) correlated part of the Cohansey Sand in New Jersey with the Yorktown Formation. The Cohansey is composed of coarse quartz sand and beds of clay and gravel. It contains few fossils, but has been tentatively correlated with the Yorktown. Isphording and Lodding (1969) considered the Cohansey to have accumulated in a regressive phase of late Miocene deposition, essentially a continuation of Calvert and Choptank deposition. This would place the lower part of the Cohansey in the middle Mio- NUMBER 53 75 cene, leaving its youngest beds undated. The presence of shallow-water strata of Miocene age offshore in the Cost B-2 Well (Smith et al. 1976:50) indicates that the Cohansey could be a regressive facies of any or all of the middle Mio- cene through Pliocene onshore units. ENVIRONMENT OF DEPOSITION.—The Albemarle and Salisbury embayments contained open ma- rine, inner- to middle-shelf environments followed by regressive marginal-marine environments dur- ing deposition of the lower and middle parts of the Yorktown Formation and equivalents. The greatest water depths postulated for these strata are 80 to 100 meters (Gibson, 1967:645) and are based on foraminifers. Bailey (1973:53-58) as- signed similar depths (maximum of 120 m) to two intervals on the basis of mollusks. In general, the Yorktown Formation in the Albemarle em- bayment includes older strata deposited in deeper water than those deposited in the southern part of the Salisbury embayment. The Placopecten clintonius zone (zone 1 of Mans- field's Yorktown Formation) is one of the most widespread units in the Yorktown Formation. This zone extends from east-central Virginia southward across the Norfolk arch into the Al- bemarle embayment and reaches southward al- most to New Bern (Figures 4: loc. 34; and 33). The depositional depths of 80 to 100 meters in the Lee Creek Mine (Gibson, 1967), suggest that the zone originally was even more widespread, but has been partly removed by erosion. The P. clintonius zone is the oldest post-Pungo River de- posit in the southern part of the Albemarle em- bayment and is the initial deposit of the York- town Formation in the Lee Creek Mine. This indicates that downwarping occurred here later than in the northern part. The lower several feet of this zone are rich in dark brown to black phosphatic nodules as large as 1 foot (30 cm) in diameter. These nodules are reworked by physical and chemical processes from the phosphatic sand of the Pungo River Formation. They formed in continental, marginal-marine, and marine envi- ronments in post-Pungo River time and were incorporated into the basal strata of the Yorktown Formation. Numerous vertebrate fossils, includ- ing bones of whales, porpoises, and seals, and shark teeth are included in this basal transgressive unit. Molluscan shells, particularly Placopecten clintonius, are abundant (Figures 26, 27). These figures also show the burrowed surface at the top of the Pungo River Formation and the reworked phosphatic nodules. The Placopecten clintonius zone, along with the underlying highest part of the "Virginia St. Marys" beds, represent the deepest water deposition of any outcropping Yorktown strata in North Carolina. The depositional depths for this zone in Virginia were not as great as in the Lee Creek Mine; most were 30 meters or less, as indicated by lower foraminiferal species diver- sity and fewer planktonic foraminifers (Gibson and Buzas, 1973:231). The Yorktown strata above the Placopecten clin- tonius zone in central and northern North Caro- lina are part of a regressional sequence as shown in the Lee Creek Mine by Gibson (1967) and in northeastern North Carolina by Hazel (1971), Bailey (1973), and Gibson (unpub. data). In the Lee Creek Mine, middle beds of the Yorktown accumulated in water of 30 meters or less. In northern North Carolina, strata containing di- verse foraminiferal and molluscan assemblages and common planktonic Foraminifera (Placopecten clintonius zone) grade upward into beds containing molluscan assemblages almost exclusively com- posed of Mulinia congesta and low-diversity fora- miniferal assemblages dominated by Elphidium excavatum (Mansfield's zone 2) with no planktonic specimens. Elphidium excavatum also is character- istic of shallow-marine, lagoon, and sound envi- ronments. These assemblages are associated with laminated clay beds near Murfreesboro and Pal- myra, North Carolina (Figure 4: Iocs. 28, 30), which suggest deposition in a lagoon or sound. The regression or subsequent erosion probably caused the missing section in the upper part of Mansfield's zone 2 of the Yorktown in the Lee Creek Mine (Gibson, 1967, Hazel, 1977). Other Yorktown erosional surfaces, such as the one sep- arating units 2 and 3 in the Lee Creek Mine (Figure 24), are present throughout Virginia and North Carolina. Whether these are submarine or subaerial scour surfaces is not known at present. 76 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Generally, the environments as interpreted from foraminiferal assemblages show little change. In many sections, the molluscan assemblages, partic- ularly the pectens, do change across the scour surfaces and suggest that some time is missing. These scour surfaces help in defining members of the Yorktown Formation in Virginia. The initial deposition of the lower strata of the Yorktown occurred in shallower water than in the southern part of the Salisbury embayment in Virginia. The beds of the Placopecten clintonius zone were deposited in open marine water of less than 30 meters, and these depths were maintained during accumulation of much of the overlying Yorktown strata. The upper beds at Yorktown (zone 2 of Mansfield) contain high-angle, large- set crossbeds of medium to coarse sand and abun- dant shell hash. These are characteristic of barrier bars (Johnson, 1969:12, 24) and reflect a regres- sion of the Yorktown sea in the Salisbury embay- ment. Beds belonging to zone 2 of Mansfield extend westward to Petersburg where they lie upon the uppermost part of the "Virginia St. Marys" beds (Figure 34). The upper beds at Petersburg are blue clay units, some containing bands of Mulinia congesta and foraminiferal assem- blages dominated by 70 percent Elphidium exca- vatum. These upper beds probably were deposited in the lagoon or sound behind the barrier-bar sequence forming to the east. This regression marked the end of Yorktown deposition in much of the Salisbury embayment in Virginia. In the southeastern part, however, a later transgression deposited upper Pliocene to lower Pleistocene strata seen at Yadkin (Hazel, 1977) and at Norfolk (Moores Bridge Well, see p. 74). This latest "Yorktown Formation" transgres- sion covered much of the eastern part of the Albemarle embayment as seen in deposits along the Chowan River, at Terra Ceia, in the upper beds in the Lee Creek Mine, and near James City along the Neuse River (Figures 4: loc. 35; and 35). The beds along the Chowan River were placed in the Yorktown Formation by Mansfield (1943) and are provisionally referred there by Hazel (1977) and Gibson (see p.364). In the south em part of the Albemarle embayment these strata are placed in the Croatan Formation (Hazel, 1977). These beds also indicate a regressional sequence toward the top. In the Lee Creek Mine, the lower strata of the Croatan were deposited in an open marine environment about 30 meters deep; later beds were deposited in open marine water of less than 15 meters depth; uppermost beds accumulated in marginal-marine environ- ments (Gibson, 1967). Bailey (1973) concluded that equivalent beds along the Chowan River accumulated initially in a shallow sublittoral en- vironment, becoming even shallower estuarine, sound, lagoon, and inlet environments when the upper parts were deposited. Thus, at the end of the latest "Yorktown Formation" depositional cycle, a series of marginal-marine, sound, lagoon, and other estuarine environments was found from the Chowan River southward through the Lee Creek area to the Neuse River. Howard (1974) believed that the Croatan Formation near James City on the Neuse River (= James City Forma- tion of DuBar and Solliday (1963); Hazel, 1977) formed in a shallow bay or sound; this would extend the marginal-marine belt to James City (Figures 4: loc. 35; and 35). The crest of the Norfolk arch was covered mostly by marginal-marine deposition during Yorktown time. Occasionally, shallow-marine deposition occurred on part of much of the arch, particularly during deposition of zone 1 of the Yorktown (Clark and Miller, 1912:159). Wells on top of the arch (Figure 37) penetrate thick se- quences of fine-grained sediments, primarily clay, largely devoid of megafossils, but with foramini- feral assemblages dominated by Elphidium exca- vatum. The precise age of these strata is uncertain at this time because of the limited faunas. Erosion and/or nonmarine deposition took place in the northern and western parts of the Salisbury embayment during Yorktown time. In the northeast part of the embayment, coarse clas- tic deposits still were originating in the uplifted Appalachians as seen in the coarse clastic se- quence of the Hammond Well and at Accomack on the Eastern Shore (Figure 4: Iocs. 13, 20). The NUMBER 53 77 Well 217 Depth (m) (ft) 0 0 ; o.• .• o'•'.•■■ ■•"-i?.-.'.'.©:■ brown to gray clay brown sand & gravel Yorktown Formation (?) 15 50 blue clay no fossils blue clayey sand green clayey sand shells Yorktown Formation green sand gravel FIGURE 37.—Geologic column for Well 217, near New Bohemia, Prince George County, Virginia, showing small amount of open marine sediments in section in Yorktown Formation. uplift of the northern source areas had moved southward by Yorktown time to include the ad- jacent Piedmont in Virginia and North Carolina, which supplied clastic sediments to the southern Salisbury and Albemarle embayments. Farther south in North Carolina, the shallow environments persisted as seen in the Duplin Formation and younger part of the Waccamaw Formation. Strata of early Yorktown age are missing in this area. Copeland (1964:229) con- cluded that the Duplin Formation was deposited in depths of 30 to 60 feet (9 to 18 m). Howard (1974:129, 130) considered deposition to have been in water less than 37 meters deep and mostly less than 18. Gibson (unpub. data) found similar depths for many of the exposures. Additional evidence for near-shore deposition in southern North Carolina are the delta lobes, which DuBar et al. (1974:153, 171) attributed to the Duplin Formation. The Waccamaw Formation was deposited on the Cape Fear arch in southern North Carolina during the later "Yorktown Formation" transgression. Deposition occurred in shallow open marine environments (Gibson, 1962:68). Howard (1974) recognized two environments in the Waccamaw in this area, one of shallow open marine (less than 15 m depth and high energy) and the second of lower salinity shallow bay. The shallowness of the Waccamaw environments to the south and the Croatan environments to the north suggests that the intervening area received no marine or marginal-marine deposition (Figure 35). Addendum Since the original submission of this manuscript in 1978, new stratigraphic knowledge has led to several changes in the stratigraphic nomenclature used herein. Ward and Blackwelder (1980) gave the name Eastover Formation to the beds previously and herein termed the "Virginia St. Marys." Gibson (1982) demonstrated that an older cycle of sand and diatomaceous clay beds of the Calvert Formation, termed the Dunkirk beds, underlies the Popes Creek Sand Member along the Patuxent River. Literature Cited Anderson, J.L. 1948. Cretaceous and Tertiary Subsurface Geology. Maryland Department of Geology, Mines, and Water Resources Bulletin, 2:1-113. Bailey, R.H. 1973. Paleoenvironment, Paleoecology, and Stratigra- phy of Molluscan Assemblages from the Yorktown Formation (Upper Miocene-Lower Pliocene) of North Carolina. 110 pages. Ph.D. dissertation, University of North Carolina, Chapel Hill. Berggren, W.A., and J.A. Van Couvering 1974. The Late Neogene. Palaeogeography, Palaeoclimatol- ogy, Palaeoecology, 16(1/2): 1-216. Blackwelder, B.W., and L.W. Ward 1976. Stratigraphy of the Chesapeake Group of Maryland and Virginia. (Guidebook 7b: Northeast-Southeast Sec- tions Joint Meeting 1976.) 55 pages. Arlington, Virginia: Geological Society of America. Blow, W.H, 1969. Late Middle Eocene to Recent Planktonic Fora- miniferal Biostratigraphy. In Proceedings of the First International Conference on Planktonic Microfossils, Ge- neva, Switzerland, 1967, 1:199-422, 54 plates, 43 figures. Leiden: E.J. Brill. Boltovskoy, E., and R. Wright 1976. Recent Foraminifera. 515 pages. The Hague: W. Junk. Brown, P.M. 1958. The Relation of Phosphorites to Ground Water in Beaufort County, North Carolina. Economic Geol- ogy, 53:85-101. Brown, P.M., P.L. Brown, T.E. Shufflebarger, Jr., and J.L. Sanpair 1977. Wrench Style Deformation in Rocks of Cretaceous and Paleocene Age, North Carolina Coastal Plain. North Carolina Department of Natural and Economic Resources Special Publication, 5: 47 pages. Brown, P.M., J.A. Miller, and F.M. Swain 1972. Structural and Stratigraphic Framework, and Spatial Distribution of Permeability of the Atlan- tic Coastal Plain, North Carolina to New York. United States Geological Survey Professional Paper, 796: 79 pages, 59 plates. Cederstrom, D.J. 1945. Geology and Ground-water Resources of the Coastal Plain in Southeastern Virginia. Virginia Geological Survey Bulletin, 63: 375 pages. Clark, W.B., and B.L. Miller 1912. The Physiography and Geology of the Coastal Plain Province of Virginia. Virginia Geological Survey Bulletin, 4:13-222. Clark, W.B., B.L. Miller, L.W. Stephenson, B.L. Johnson, and H.N. Parker 1912. The Coastal Plain of North Carolina. North Caro- lina Geological and Economic Survey, 3: 552 pages. Clark, W.B., G.B. Shattuck, and W.H. Dall 1904. The Miocene Deposits of Maryland. Maryland Geo- logical Survey, Miocene, pages xxi-clv. Copeland, C.W. 1964. Eocene and Miocene Foraminifera from Two Lo- calities in Duplin County, North Carolina. Bulle- tins of American Paleontology, 47(215):205-324, plates 23-43. Dryden, L. 1936. The Calvert Formation in southern Maryland. Pennsylvania Academy of Science Proceedings, 10:42-51. DuBar, J.R., H.S. Johnson, Jr., B. Thom, and W. O. Hatchell 1974. Neogene Stratigraphy and Morphology, South Flank of the Cape Fear Arch, North and South Carolina. In R.Q. Oaks, Jr., and J.R. DuBar, editors, Post-Miocene Stratigraphy, Central and Southern Atlantic Coastal Plain, pages 139-173. Logan: Utah State University Press. DuBar, J.R., and J.R. Solliday 1963. Stratigraphy of the Neogene Deposits, Lower Neuse Estuary, North Carolina. Southeastern Geol- ogy, 4 {4) .213-233. Ernissee, J.J., W.H. Abbott, and P.F. Huddlestun 1977. Microfossil Correlation of the Coosawhatchie Clay (Hawthorn Formation, Miocene) of South Caro- lina, and Its Equivalent in Georgia. Marine Micro- paleontology, 2:105-119. Ferenczi, I. 1959. Structural Control of the North Carolina Coastal Plain. Southeastern Geology, 1:105-116. Gernant, R.E. 1970. Paleoecology of the Choptank Formation (Mio- cene) of Maryland and Virginia. Maryland Geolog- ical Survey Report of Investigations, 12: 90 pages. 1971. Invertebrate Biofacies and Paleoenvironments. Maryland Geological Survey Guidebook, 3:19-30. Gibson, T.G. 1962. Benthonic Foraminifera and Paleoecology of the Miocene Deposits of the Middle Atlantic Coastal Plain. Ph.D. dissertation, Princeton University. 198 pages. 1967. Stratigraphy and Paleoenvironment of the Phos- phatic Miocene Strata of North Carolina. Geolog- ical Society of American Bulletin, 78 (5): 631-650. 1968. Stratigraphy and Paleoenvironment of the Phos- phatic Miocene Strata of North Carolina: Reply. Geological Society of America Bulletin, 79:1437-1448. 1967. 78 NUMBER 53 79 1970. Late Mesozoic-Cenozoic Tectonic Aspects of the Atlantic Coastal Margin. Geological Society of Amer- ica Bulletin, 81:1813-1822. 1971. Miocene of the Middle Atlantic Coastal Plain. Maryland Geological Survey Guidebook, 3:1-15. 1982. Depositional Framework and Paleoenvironments of Miocene Strata from North Carolina to Mary- land. In T.M. Scott and S.B. Upchurch, editors, Miocene of the Southeastern United States. Florida Bureau of Geology Special Publication, 25:1-22. In prep. Miocene and Pliocene Pectinidae (Bivalvia) from the Lee Creek Mine and Adjacent Areas. Gibson, T.G., G.W. Andrews, L.M. Bybell, N.'o. Frederik- sen, T. Hansen, J.E. Hazel, D.M. McLean, R.J. Witmer, and D.S. Van Nieuwenhuise 1980. Geology of the Oak Grove Core, Part 2: Biostra- tigraphy of the Tertiary Strata of the Core. Virginia Division of Mineral Resources Publication, 20:14-30. Gibson, T.G., and M.A. Buzas 1973. Species Diversity: Patterns in Modern and Mio- cene Foraminifera of the Eastern Margin of North America. Geological Society of America Bulletin, 84:217-238. Gibson, T.G., and K.M. Towe 1971. Eocene Volcanism and the Origin of Horizon A. Science, 172:152-154. 1975. Origin of Horizon A: Clarification of a Viewpoint. Science, 188:1221. Glaser, J.D. 1971. Geology and Mineral Resources of Southern Maryland. Maryland Geological Survey Report of Inves- tigations, 15: 84 pages. Hazel, J.E. 1971. Ostracode Biostratigraphy of the Yorktown For- mation (Upper Miocene and Lower Pliocene) of Virginia and North Carolina. United States Geolog- ical Survey Professional Paper, 704: 13 pages. 1977. Distribution of Some Biostratigraphically Diag- nostic Ostracodes in the Pliocene and Lower Pleis- tocene of Virginia and Northern North Carolina. Journal of Research, 5(3): 3 73-388. Howard, J.F. 1974. Neogene Microfaunas in the Cape Fear Arch Area. In R.Q. Oaks, Jr., J.R. DuBar, editors, Post-Miocene Stratigraphy, Central and Southern Atlantic Coastal Plain, pages 123-138. Logan: Utah State Univer- sity Press. Isphording, W.C. 1970. Petrology, Stratigraphy, and Re-definition of the Kirkwood Formation (Miocene) of New Jersey. Journal of Sedimentary Petrology, 40(3):986-997. 1976. Multivariate Mineral Analysis of Miocene-Pli- ocene Coastal Plain Sediments. In Gulf Coast As- sociation Geological Societies, 26th Annual Meeting, Shreveport, Louisiana, pages 326-332. Isphording, W.C, and W. Lodding 1969. Facies Changes in the New Jersey Miocene. In S. Subitsky, editor. Geology of Selected Areas in New Jersey and Eastern Pennsylvania and Guidebook of Excur- sions, pages 7-13. New Brunswick, New Jersey: Rutgers University Press. Johnson, G.H. 1969. Guidebook to the Geology of the York-James Pen- insula and South Bank of the James River. College of William and Mary, Department of Geology Guidebook, 1: 33 pages. Kazakov, A.V. 1937. The Phosphorite Facies and the Genesis of Phos- phorites. In Geological Investigations of Agricul- tural Ores. Leningrad Scientific Institute of Fertilizers and Inseclo-Fungicides Transactions, 142:95-113. Kimrey, J.O. 1964. The Pungo River Formation, a New Name for Middle Miocene Phosphorites in Beaufort County, North Carolina. Southeastern Geology, 5(4): 195-205. 1965. Description of the Pungo River Formation in Beaufort County, North Carolina. North Carolina Department of Conservation and Development, Division of Mineral Resources Bulletin, 79: 131 pages. Kiimmel, H.B., and G.N. Knapp 1904. The Stratigraphy of the New Jersey Clays. New Jersey Geological Survey Final Report, 6:117-209. Maher, J.C. 1965. Correlations of Subsurface Mesozoic and Cenozoic Rocks along the Atlantic Coast. 18 pages, 9 plates. Tulsa: American Association of Petroleum Geologists. 1971. Geologic Framework and Petroleum Potential of the Atlantic Coastal Plain and Continental Shelf. United States Geological Survey Professional Paper, 659: 98 pages. Mansfield, W.C. 1929. The Chesapeake Miocene Basin of Sedimentation as Expressed in the New Geologic Map of Vir- ginia. Washington Academy of Science Journal, 19:263- 268. 1937. Some Deep Wells near the Atlantic Coast in Vir- ginia and the Carolinas. United States Geological Survey Professional Paper, 186-1:159-161. 1943. Stratigraphy of the Miocene of Virginia and the Miocene and Pliocene of North Carolina. In Julia Gardner, Mollusca from the Miocene and Lower Pliocene of Virginia and North Carolina, Part 1: Pelecypoda. United States Geological Survey Profes- sional Paper, 199-A: 1-19. McConnell, D. 1958. The Apatite-like Mineral of Sediments. Economic Geology, 53:110-111. Murray, G.E. 1961. Geology of the A tlanlic and Gulf Coastal Province of North America. 692 pages. New York: Harper and 80 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Brothers. Murray, J.W. 1973. Distribution and Ecology of Living Benthic Foraminifer- ids. 274 pages. New York: Crane, Russak & Com- pany, Inc. Olsson, A. 1917. The Murfreesboro Stage of our East Coast Mio- cene. Bulletins of American Paleontology, 5:155-163. Phleger, F.B., Jr. 1960. Ecology and Distribution of Recent Foraminifera. 297 pages. Baltimore: Johns Hopkins Press. Poag, C.W. 1978. Stratigraphy of the Atlantic Continental Shelf and Slope of the United States. Annual Review of Earth and Planetary Science, 6:251-280. Reineck, H.-E., and I.B. Singh 1980. Depositional Sedimentary Environments. Second edi- tion, 549 pages. New York: Springer-Verlag. Reinhardt, J., W.L. Newell, and R.B. Mixon 1980. Geology of the Oak Grove Core, Part 1: Tertiary Lithostratigraphy of the Core. Virginia Division of Mineral Resources Publication, 20:1-13. Richards, H.G. 1945. Subsurface Stratigraphy of Atlantic Coastal Plain between New Jersey and Georgia. American Asso- ciation of Petroleum Geologists Bulletin, 29:885-955. Richards, H.G., and A. Harbison 1942. Miocene Invertebrate Fauna of New Jersey. Pro- ceedings of the Academy of Natural Sciences of Philadel- phia, 94:167-250, plates 7-22, 9 figures. Rooney, T.P., and P. F. Kerr 1964. Clinoptilolite: A New Occurrence in North Caro- lina Phosphorite. Science, 144:1453. Shattuck, G.B. 1902. The Miocene Formation in Maryland. Science, 15:906. 1904. Geological and Paleontological Relations, with a Review of Earlier Investigations. In W.B. Clark, G.B. Shattuck, and W.H. Dall, The Miocene De- posits of Maryland. Maryland Geological Survey, Mio- cene, pages xxxiii-cxxxvii. Shifflett, E. 1948. Eocene Stratigraphy and Foraminifera of the Aquia Formation. Maryland Department of Geology, Mines and Water Resources Bulletin, 3: 93 pages, 5 plates. Smith, M.A., R.V. Amato, M.A. Furbush, DM. Pert, M.E. Nelson, J.S. Hendrix, L.C. Tamm, G. Wood, Jr., and DR. Shaw 1976. Geological and Operational Summary, Cost No. B-2 Well, Baltimore Canyon Trough Area, Mid- Atlantic OCS. United States Geological Survey, Open- File Report, 76-774: 79 pages. Spangler, W.B. 1950. Subsurface Geology of Atlantic Coastal Plain of North Carolina. American Association of Petroleum Geologists Bulletin, 34:100-132. Spangler, W.B., and J.J. Peterson 1950. Geology of Atlantic coastal Plain in New Jersey, Delaware, Maryland, and Virginia. American As- sociation of Petroleum Geologists Bulletin, 34:1-99. Stephenson, L.W. 1928. Major Marine Transgressions and Structural Fea- tures of the Gulf Coastal Plain. American Journal of Science, series 5, 16:281-298. Stephenson, L.W., and F.S. MacNeil 1954. Extension of Yorktown Formation (Miocene) of Virginia into Maryland. Geological Society of America Bulletin, 65:733-738. Taliaferro, N.L. 1933. The Relation of Volcanism to Diatomaceous and Associated Siliceous Sediments. University of Cali- fornia Publications, Bulletin of the Department of Geolog- ical Sciences, 23(1): 1-56. Talley, J.H. 1975. Cretaceous and Tertiary Section, Deep Test Well, Greenwood Delaware. Delaware Geological Survey, Report of Investigations, 23: 51 pages. Teiflce, R.H. 1973. Stratigraphic Units of the Lower Cretaceous through Miocene Series. In Geologic Studies of the Coastal Plain of Virginia. Virginia Division of Min- eral Resources Bulletin, 83(1): 1-78. Valia, H.S., H. Khalifa, and B. Cameron 1977. Recorrelation of Eocene-Miocene Boundary of Delaware Coastal Plain. American Association of Pe- troleum Geologists Bulletin, 61(5):723-740. Yokes, E.H. 1967. Cenozoic Muricidae of the Western Atlantic Re- gion Part III—Chicoreus {Phyllonotus). Tulane Studies in Geology, 5(3): 133-166. Walton, W.R. 1964. Recent Foraminiferal Ecology and Paleoecology. In J. Imbrie and N.D. Newell, Approaches to Paleo- ecology, pages 151-237, New York: John Wiley and Sons. Ward, L.W., and B.W. Blackwelder 1980. Stratigraphic Revision of Upper Miocene and Lower Pliocene Beds of the Chesapeake Group, Middle Atlantic Coastal Plain. United States Geo- logical Survey Bulletin, 1482-D: 61 pages. Weaver, C.E., and K.C. Beck 1977. Miocene of the S.E. United States. Sedimentary Geology, 17(1/2): 1-234. Weaver, F.M., and SW. Wise, Jr. 1974. Opaline Sediments of the Southeastern Coastal Plain and Horizon A: Biogenic Origin. Science, 184:899-901. Whitmore, F.C, Jr. 1971. Vertebrate Biofacies and Paleoenvironments. Maryland Geological Survey Guidebook, 3:31-36. Age and Correlation of the Yorktown (Pliocene) and Croatan (Pliocene and Pleistocene) Formations at the Lee Creek Mine Joseph E. Hazel ABSTRACT The fossiiiferous beds above the Pungo River Formation (middle Miocene) in the Lee Creek open pit mine in Beaufort County, North Caro- hna, are approximately 70 feet (21.3 m) thick. This thickness includes 46 feet (14 m) that is correlative with the Yorktown Formation of the type area and is referred to that unit, and, above the Yorktown, a fossiiiferous section 23 feet (7 m) thick that is assigned to the Croatan Formation. The 149 species or subspecies of ostracodes identified were from 16 samples from the York- town and Croatan. Coefficients of faunal similar- ity were calculated for all samples, and the re- sulting matrix was subjected to unweighted pair- group cluster analysis. Three major faunal group- ings were delineated. The principal faunal dis- continuity occurs at the Yorktown-Croatan con- tact about 46 feet (14 m) above the base of the Yorktown. The beds below this level belong to the Pterygocythereis inexpectata and Orionina vaughani ostracode assemblage zones. Correlation with other Coastal Plain deposits containing plank- tonic foraminifers indicates that the Orionina vaughani assemblage zone is planktonic foramini- fer zones N19 and N20 in age and that the Pterygocythereis inexpectata assemblage zone may approximate the lowest part of planktonic zone N19 in age. Thus, the Yorktown in the Lee Creek Mine is of early Pliocene age. This is seemingly corroborated by a K/Ar date of 4.4±0.2 my on the Orionina vaughani assemblage zone in Virginia. Joseph E. Hazel, United States Geological Survey, National Center, Reston, Virginia 22092. A third major faunal assemblage is found in the beds of the Croatan Formation, which are referable to the Puriana mesacostalis ostracode as- semblage zone. The upper part of the Croatan can be correlated with rocks in Florida and North Carolina that have been radiometrically dated by the He/U method at about 1.8 to 1.9 mya. A tentative He/U radiometric date of 2.4 mya was obtained for the lower part of the Croatan at the mine. If a date of about 2.0 mya is used for the Pliocene-Pleistocene boundary, the Croatan as used in the mine spans the Pliocene-Pleistocene boundary. Introduction Texasgulfs Lee Creek open pit phosphate mine is on the south bank of the Pamlico River in Beaufort County, North Carolina. The mine has been the subject of considerable interest since it opened in 1963, not only because of the impor- tance of the primary phosphorite deposits but also because it affords access to an exposure of fossiiiferous upper Cenozoic rocks more than 120 feet (36.6 m) thick, an uncommon phenomenon in the Coastal Plain. This paper is concerned with the ostracodes and their biostratigraphy in the beds of the York- town and Croatan formations exposed in the mine walls. This section is interesting not only because of the excellent exposure, but because it is only about 25 miles (40.25 km) north of the Neuse River, the approximate southern limit of the region where the term "Yorktown" is com- 11 82 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY monly used, and it is in the only area where the term "Croatan" (=James City Formation of DuBar and Solliday, 1963) has been used. South of the Neuse, in the Carolinas and in Georgia, the terms "Duplin" and "Waccamaw" are generally used for rocks considered to be at least in part equivalent to the Yorktown and Croatan. The Duplin is known to contain a warmer water fauna than the Yorktown (for example, Mansfield, 1929); thus, there is also a biogeographic bound- ary in the vicinity of the Neuse River, although at present this boundary is poorly understood. ACKNOWLEDGMENTS.—I am grateful to R.M. Forester, P.C. Valentine, and T.M. Cronin, U.S. Geological Survey, and W.A. Berggren, Woods Hole Oceanographic Institution, for critically reading all or parts of the manuscript. I have benefited from discussions with B.W. Black- welder, formerly of the U.S. Geological Survey, and L.W. Ward, U.S. Geological Survey, on up- per Cenozoic stratigraphy in the Virginia-North Carolina region. I thank M.L. Bender, University of Rhode Island, for informative discussions and for allowing publication of helium-uranium dat- ing techniques. W.C. Blow, formerly of the U.S. Geological Survey, and Ellen E. Compton, U.S. Geological Survey, provided excellent laboratory support. Previous Work Gibson (1967) made the first study of the fos- siiiferous rocks in the Lee Creek Mine, assigning them to the Pungo River and Yorktown forma- tions. He measured a 66.4-foot (20.2 m) section of Yorktown in the test pit, in the northeast part of the mine area as shown in Figure 1. Gibson assigned the Yorktown in the mine to the later Miocene, because he concluded that the lower- most beds contained a planktonic foraminifer assemblage consistent with that age. Gibson (1967:638) also pointed out that the presence of iMile 1 Kilometer Garrison Point OPEN ADVANCE Approx. location of long wall ^' April 1972~*"' N \ - \ / PIT /- 1 ' FIGURE 1.—Location of the Texa.sgulf Inc. Lee Creek open pit phosphate mine near Aurora, North Carolina. The samples used in the present study were taken from the southwestern part of the mine. NUMBER 53 83 Placopecten clintonius (Say) in the lowermost beds indicates correlation with Mansfield's (1929, 1943) Placopecten clintonius zone (zone 1) of the Yorktown, and that the presence of Ostrea sculp- turata in the middle part of the section suggests placement in the lower part of Mansfield's Turri- tella alticostata zone (zone 2). On the basis of foraminifer assemblages, Gibson (1967:647) con- cluded that the Yorktown in the mine was depos- ited initially in cool-temperate waters nearly 100 m deep and finally in warmer waters less than 15 m deep. In 1969, Hazel (1971a: 10) referred a 69-foot (21.0 m) section in the southwestern part of the mine to the Yorktown. On the basis of a cluster analysis of 43 Yorktown collections (10 from the Lee Creek Mine), he concluded that the lower few feet of the Yorktown in the mine belonged to the Pterygocythereis inexpectata ostracode assemblage zone, and approximately the upper 12 feet (3.7 m) to the Puriana mesacostalis assemblage zone; the rest of the accessible section between these two zones was placed in the Orionina vaughani assem- blage zone. Hazel (197la:8) assigned the Puriana mesacostalis assemblage zone to the Pliocene and suggested that more of the Atlantic and Gulf Coastal Plain deposits traditionally assigned to the upper Miocene could be Pliocene. Hazel (1971b:373) concluded that the early Yorktown assemblage lived under mild or warm-temperate climatic conditions, that warm-temperate condi- tions prevailed during most of Yorktown time, and that a subtropical marine climate was present in Croatan (his late Yorktown) time. Swain (1974) studied ostracodes from the York- town Formation from various localities in Vir- ginia and North Carolina. Most of the specimens studied by Swain were picked from carbon tet- rachloride floats prepared for the study of fora- minifers by T.G. Gibson of the U.S. Geological Survey. Few species and specimens were available for study, as Swain indicated in his descriptive section, probably because ostracodes (particularly single valves) do not "float" well. Yorktown sam- ples, 500 to 1000 cm^ in size, when picked after only washing or after concentrating the carbonate fraction by using a soap float technique (Howe, 1941; Gibson and Walker, 1967), have yielded hundreds and on occasion thousands of ostracode specimens (this study and Hazel, 1971a). Swain (1974), therefore, would have had difficulty in recognizing the assemblage zones established by Hazel (1971a) because he studied only a small number of ostracode specimens. Swain's (1974) study was further impeded when he tried to use the particular species for which Hazel's (1971a) assemblage zones were named as index or guide fossils, despite the fact that Hazel clearly stated that the zones were assemblage zones. Swain (1974:10) also studied some samples from the Lee Creek Mine and is of the opinion that the large assemblages reported by Hazel (1971a) from the lower part of the Yorktown at the mine are in part reworked from the underly- ing Miocene Pungo River Formation. The writer has processed several samples from the lower part of the Yorktown at the mine and has found no evidence of reworking of ostracodes from the Pungo River into the Yorktown. T.G. Gibson (pers. comm., 1975) has also studied many sam- ples from the lower part of the Yorktown at the mine, and he, too, finds no evidence of reworking. The Pungo River carries a taxonomically distinct and differently preserved assemblage. Swain (1974:10) stated that these lower beds of the Yorktown in the mine are correlative with the middle or upper part of the Yorktown of the other sections he studied. The ostracodes (Hazel, 1971a, and p. 93 herein) and mollusks (Gibson, 1967:638) indicate an obvious correlation with the lower part of the Yorktown of the type area. Most of Swain's samples (1974) were from his "lower Yorktown," and some of these at least are actually from the underlying Eastover Formation of Ward and Blackwelder (1980). Specifically, samples from Swain's lower part of the Yorktown at localities 5, 6, 7, 9, 10, and 11 may be partly or entirely from the Eastover Formation. The Eastover is late Miocene (Ward and Blackwelder, 1980:11) and (or) early Pliocene (Andrews, 1980:19) in age. Swain (1974:9-11) considered the Yorktown Formation to be of late Miocene age; however, he presented no supporting evidence. He assigned 84 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY the beds at Colerain Landing and near Mt. Gould Landing on the Chowan River to the Pliocene(?). Hazel (1971a) assigned these localities to the Pliocene and placed them in the Puriana mesacos- talis assemblage zone. Swain's (1974:11) argu- ment that Puriana mesacostalis cannot be used as a guide for the Pliocene because it "was described from the upper Miocene Duplin Marl" is circular. There is no evidence that the Duplin is Miocene, but there is considerable evidence that it is Pliocene. YORKTOWN FORMATION The Yorktown Formation is poorly understood as a lithostratigraphic unit, because the generally recognized Yorktown sediments contain a variety of lithologies. No extensive petrological exami- nations have been made of the Yorktown, and no extensive detailed mapping has been done at the surface or in the shallow subsurface over any sizable area where the term has been used. With- out such a solid physical stratigraphic framework then, it is understandable, if not excusable, that the term "Yorktown" connotes a biostratigraphic, rather than lithostratigraphic, unit to many workers. In the type area of the Yorktown, which can be defined as the valleys of the York and James rivers and the included peninsula, the Yorktown has generally come to mean the beds containing mollusks referable to Mansfield's (1929, 1943) Placopecten clintonius and Turritella alticostata zones (the so-called zones 1 and 2, respectively). These two zones are in need of redefinition and revision in the light of modern biostratigraphic thought; nonetheless, the Yorktown has been recognized in an area of about 17,000 square miles (44,030 square kilometers) as a unit of perhaps substage magnitude. An "ian" ending was added to the Yorktown by Malkin (1953:767), who considered it a sub- stage of the upper Miocene. She did not, however, study much of the Yorktown or discuss the distri- bution and correlation of this "substage" in any detail. Such nomenclatorial procedures have been followed before (for example, Murray, 1961). However, addition of "ian" endings to ostensibly lithostratigraphic terms in order to make their biostratigraphic use legitimate, although not nec- essarily confusing, is cumbersome and nomencla- torially undesirable; therefore, it is not followed here. In North Carolina, rocks younger than the Yorktown of the type area have been included in the Yorktown by Clark et al. (1912), Mansfield (1943), MacNeil (1938), Hazel (1971a), and at the Lee Creek Mine by Gibson (1967) and Hazel (1971a). Gibson (1970) and Swain (1974) have also placed deposits demonstrably older than the classic Yorktown in the Yorktown Formation. Lithologic units 1-4 of Figure 2 are correlative with the Yorktown Formation of the type area, and their lithologies do not differ from those of beds assigned to the Yorktown in northern North Carolina and Virginia. Therefore, units 1-4 (equivalent to units 1-5 of Gibson, 1967) are assigned to the Yorktown Formation. CROATAN FORMATION Units 5-7 (equivalent to Gibson's, 1967, units 6-9) in the mine have been referred to the James City Formation of DuBar and Solliday (1963) by DuBar, Solliday, and Howard (1974:109). The James City was proposed as a substitute for the Croatan Formation of Dall (1892) by DuBar and Solliday (1963:215, 228), because Dall did not adequately define the Croatan Formation in his original work. Dall's collections apparently con- tained both Pliocene and Pleistocene species, and he did not designate a type section. Mansfield (1928:135) reviewed the situation and restricted the name Croatan "to those beds on or near the Neuse River which are of Pliocene age." Accord- ing to DuBar and Solliday (1963:223) and DuBar, Solliday, and Howard (1974:106), how- ever, the beds to which Mansfield restricted the name probably represent a Pleistocene unit con- taining reworked Pliocene fossils. DuBar and Solliday (1963:228) selected as the type locality of their James City Formation the NUMBER 53 85 outcrops on the Neuse River just below the town of James City, which is about 12 miles (19.2 km) from the town of Croatan. Mansfield (1936) had previously included these outcrops in the Croatan but did not tie the name to a type section. How- ever, MacNeil (1938:19) did suggest this locality as the type section for the Croatan. Because this is in the area of the Croatan people (the unit was not named for the town) and MacNeil has indi- cated a type section, the writer sees no reason not to adhere to the rules of priority and retain the term "Croatan," suppressing the term James City. Attention is called, however, to the clarify- ing efforts of DuBar and Solliday (1963) and particularly DuBar, Solliday, and Howard (1974). Dall used the term "Croatan beds," and Mans- field referred to the "Croatan Sand," even though several lithologies are present in the Croatan (see DuBar, Solliday, and Howard, 1974). The writer believes that the unit should be referred to as the Croatan Formation. Collections and Analyses The Lee Creek Mine was visited again by the writer in April 1971, and a 69-foot (21.0-m) sec- tion of Yorktown was measured in the southwest area of the mine. Collections were made from most of the beds that were inaccessible in the middle part of the formation in 1969 (Hazel, 1971a: 10). These and the original collections were supplemented by collections made by L.W. Ward of the U.S. Geological Survey in 1972. The stra- tigraphic position of the various collections is indicated in Figure 2. The 149 species or subspecies of ostracodes occurred in the 16 samples used for multivariate analysis. Three Yorktown samples, one from the Croatan, and two from unit 8, which may repre- sent the Flanner Beach Formation (Pleistocene), were barren of ostracodes and are indicated by X in Figure 2. The samples were compared in Q- mode (samples compared on the basis of species content) by calculating Otsuka similarity coeffi- cients between all samples and performing an unweighted pair-group cluster analysis (UPGM) on the resulting matrix. These techniques, as applied to biostratigraphy, have been described by Hazel (1970, 1971a). To minimize environ- mental or preservational differences between samples, the range-through method of calculation was used (see Cheetham and Deboo, 1963); that is, a species was counted as present in a sample for the purposes of the calculation of the similarity coefficient if it occurred in samples on either side of but not in the sample in question. The Q-mode dendrogram resulting from the cluster analysis is also illustrated in Figure 2. In order to ascertain which species were pri- marily responsible for the groupings seen in Q- mode, an R-mode analysis (taxa compared with each other on the basis of the samples in which they occur) using the Otsuka coefficient was per- formed on all those species or subspecies that occur in more than one sample but not in all samples (a total of 85 taxa). The results of this R- mode analysis are presented in Figure 3. ALPHABETICAL LIST OF SPECIES The 149 ostracode species found in the Yorktown Formation at the Lee Creek Mine are listed alphabetically. The number to the left of the name is the computer code number assigned to the species. The numbers to the right of the name indicate occurrences in the samples of Figure 2. The plates and figures illustrating the species are also indicated. Cod( i Species Sample Illustration 238 Actinocythereis captionis 2, 10-16 PI. 8: figs. 1, 2, 4 241 A. dawsoni 1,3,5-7, 9 PI. 9: fig. 186 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Code Spec ies Sample Illustration 239 A. marylandica 1, 5,6,9 PI. 8: fig. 3 240 A. mundorffi, sma 11 form 2, 3, 5-8, 16 PI. 9: fig. 2 243 Acuticythereis laevissima 7,8 318 Anchistrocheles sp C 11 9 Aurila laevicula 6,8 PI. 15 fig. 4 14 Bairdoppilala triangulata 1,6,8-16 27 Bensonocythere blackwelden 1,6,8,9, 11-13 PI. 35 figs. 1,2,4; PI. 37: fig. 4 32 B. bradyi 11 PI. 34 figs. 1, 2; PI. 38 figs. 2, 4 24 B. calverli 1 PI. 34 fig. 5 19 B. gouldensis 4,8,9, 11, 15 PI. 34 figs. 3, 4; PI. 37: figs. 2, 3 30 B. ncespitensis 12, 14 PI. 33 figs. 1-4 25 B. rugosa 3, 8, 10 PI. 32 figs. 3, 4; PI. 37: fig. 1 21 B. trapezoidalis 1, 5,9 PI. 32 figs. 1, 2 18 B. whitei 10-16 PI. 35 fig. 3 33 Bensonocythere sp. M 15 319 Bensonocythere sp. OO 11 320 Bensonocythere sp. PP 10, 13 329 Bensonocythere sp. QQ 10 330 Bensonocythere sp. RR 13 44 Bythocythere sp. E 14 47 Campylocythere laeva 5,6,9-16 55 Caudites paraasymmetncus 13-15 PI. 12 figs. 2-4 56 Cneslocythere? sp. 16 217 Cushmanidea cf C. seminuda 14 57 Cypride IS sp. B 15, 16 62 Cytherella sp. A 5, 12 321 Cy there lla sp. B 1 63 Cylherelloidea sp. A 14, 16 106 Cytheridea campwallacensis 1, 4, 6 PI. 1: fig. 4; PI. 2: figs. 1,3,4 111 C. carolinensis 10-15 PI. 3 107 C. virginiensis 2,3,5-15 PI. 1: fig. 3; PI. 2: fig. 2 105 Cytheridea aff C. virginiensis 3,6, 7 PI. 1: figs. 1, 2 250 Cytheromorpha? curta 13 68 C. incisa 9, 11, 13 PI. 21 figs. 1,2; PI. 23: figs. 5, 6 70 C. macroincisa 11, 15 PI. 22 figs. 1-5 67 C. suffolkensis 9, 11, 12 PI. 23 figs. 1-4 66 C. warneri 1, 5-10 PI. 22 fig. 6 322 Cytheromorpha sp. I 1 71 Cytheropteron lalqt inensis 1, 2, 5-9, 12 PI. 9: fig. 4 72 C. ? yorktownensis 1, 3, 5,8-12, 14-16 PI. 9: fig. 3 317 Cytherura elongata 1, 3, 9, 15 132 C. forulala 1, 5,8,9, 12-15 75 C. howei 1, 5, 8, 12, 13, 15 77 C. reticulata 9, 10, 13-15 80 C. wardensis 6, 9, 15 193 Cytherura sp. 5,8 323 Cytherura sp. A A 13, 15 340 Cytherura sp. BB 13 76 Cytherura sp. D 1, 3, 5, 7-9 83 Cytherura sp. L 8-10, 14 84 Cytherura sp. M 13, 15 85 Cytherura sp. N 9, 12-16 194 Cytherura sp. U 13 316 Cytherura sp. W 13 NUMBER 53 Code Species Sample Illustration 90 Echinocythereis leecreekensis 10-12 PI. 36: figs. 1-3; PI. 38: fig. 3 89 E. planibasalis 1, 5,6,8,9 PI. 36: fig. 4 91 Eucythere declivis 1, 10, 12 92 E. gibba 5, 6, 9, 11-13 94 E. triangulata 11, 13 96 Eucythere sp. F 5 331 Hermanites ascitus 8,9 PI. 11: figs. 1-3 115 Hirschmanma? hespera 10-12, 14 PI. 20: figs. 1,2; PI. 21: figs. 3,4 114 H.? quadrata 10-14 PI. 20: figs. 3, 4 23 Hulingsina amencana 2, 4, 6-14, 16 156 H. glabra 10, 11, 13, 14 22 H. rugipustulosa 1,9-13, 15, 16 128 Hulingsina sp. C 1,9, 10, 12-15 82 Hulingsina sp. F 11-15 41 Hulingsina sp. R 5 61 Hulingsina sp. U 7 121 Leptocythere nikraveshae 14 207 Leptocythere sp. E 13 326 Leptocythere sp. F 11 127 Loxoconcha edentonensis 12, 13 PI. 24: figs. 2, 4 136 L. matagordensis 13-16 126 L. purisubrhomboidea 12 131 L. reticularis 6,8-10, 12, 13 125 Loxoconcha sp. C 1-3,5,9 130 Loxoconcha sp. H 3-7, 12-16 133 Loxoconcha sp. M 5,8,9 69 Loxoconcha sp. S 1 216 Loxoconcha sp. T 14, 16 13 Malzella conradi, angulate form 8-10 PI. 14 figs. 1, 2, 4 12 M. evexa 1, 3, 5, 6,8-14, 16 PI. 14 fig. 3; PI. 15: figs. 1-3, 5 139 Microcytherura choctawhatcheensis 1,6-16 PI. 29 fig. 3 140 M. expanda 10, 11, 15, 16 PI. 30 figs. 1-3 149 M. minuta 3 PL 31 figs. 1-3 138 M. similis 9-16 PI. 29 fig. 4; PI. 30: fig. 4; PI. 31: fig. 4 141 Microcytherura sp. D 13 145 Microcytherura sp. H 1,5,8-10 147 Microcytherura sp. M 12 257 Microcytherura sp. P 15 312 Microcytherura sp. R 10, 11, 15 151 Muellerina bassiounii 10-13, 16 PI. 16: figs. 1, 4; PI. 18: fig. 6 153 M. blowi 8-13 PI. 17: figs. 1,3; PI. 18: fig. 2 157 M. canadensis petersburgensis 1-3, 5-9 PI. 16: fig. 2; PI. 18, figs. 1, 3 150 M. ohmerti 1,3, 5, 7-16 PI. 16: fig. 3 152 M. wardi 1,3,9-14, 16 PI. 17: figs. 2,4; PI. 18: figs. 4,5 160 Muellerina sp. P 4 163 Murrayina barclayi 7,9 PI. 11: fig. 4 162 M. macleani 1-5, 7 PI. 10: figs. 1-4 164 Murrayina sp. E 7 309 Neocaudites angulatus 9, 15 PI. 6: figs. 2-4 328 N. subimpressus 8 PI. 5: fig. 4 166 N. trip lis triatus 6,8,9 PI. 6: fig. 1 167 N. vanabilus 11, 12 PI. 5: figs. 1-3; PI. 7: fig. 1 172 Orionina vaughani 1, 5,6,8-16 PI. 12 fig. 187 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Code Species Sample Illustration 54 Palaciosa minuta 13 PI. 13: figs. 1, 3, 4 182 Paracyprideis sp. C 5-7,9 184 Paracypris sp. B 5 189 Paracythendea altila 1,6,9-16 PI. 28: fig. 4 188 P. cronini 5,6,8-10, 12, 13 PI. 28: figs. 1, 2; PI. 29: fi 223 P. mucra 13 PI. 29: fig. 2 190 P. rugosa 13, 15 PI. 28: fig. 3 327 Paracytheridea sp. F 9 192 Paracytheroma stephensoni 13-15 196 Paradoxostoma delicata 13, 15 198 Paradoxostoma sp. E 11 200 Paranesidea? laevicula 13 201 Paranesidea sp. B 8 98 Peratocytheridea bradyi 6, 7, 10-14 , 16 99 P. sandbergi 3-7, 9, 10, 12-16 PI. 4: figs. 1-3 324 P. setipunctata 15 PI. 4: fig. 4 104 Peratocytheridea sp. J 14 143 ''Pontocythere'' sp. I 10, 13, 15, 16 PI. 38: fig. 1 230 ''Ponlocythere'" sp. G 1, 3, 5-7 155 ''Pontocythere" sp. J 9, 13 205 Propontocypris sp. D 3, 5, 7 175 Proteoconcha gigantica 13 173 P. jamesensis 9 PI. 25: figs. 1, 2; PI. 27: fi 178 P. mimica 9 206 P. multipunctata, sensu lato 1, 5,6, 8-1 5 8 P. tuberculata 13 177 Proteoconcha sp. Z 1 215 Pseudocytheretta burnsi 1, 2,4-11, 13-15 PI. 24: figs. 1, 3 224 Pterygocythereis inexpectata 1-7 PI. 7: fig. 3 227 Puriana carolinensis 1,3,6-16 PI. 27: figs. 1, 3, 4 228 P. convoluta 10, 13, 14 PI. 26: figs. 1, 2, 4 229 P. mesacostalis 13, 15 PI. 25: fig. 4 226 P. rugipunctata 6, 9, 12, 12 PI. 25: fig. 3; PI. 26: fig, 3 165 Radimella confragosa 10-16 PI. 13: fig. 2 236 Sclerochilus sp. B 11, 14 15 Thaerocythere carolinensis 12 PI. 19: figs. 1, 3, 4 78 T. schmidtae 10, 12, 14 PI. 19: fig. 2 16 Xestolebens ventrostriata 10, 11, 13, 14 169 Xestoleberis sp. E 12, 15 NUMERICAL COMPUTER CODE LIST OF SPECIES Code Species 8 Proteoconcha tuberculata (Puri, 1960) 9 Aurila laevicula (Edwards, 1944) 12 Malzella evexa, new species 13 M. conradi (Howe and McGuirt, 1935), angulate form 14 Bairdoppilata triangulata Edwards, 1944 15 Thaerocythere carolinensis, new species 16 Xestolebens ventrostriata Swain, 1951 18 Bensonocythere whitei {Swam, 1951) 19 B. gouldensis, new species 21 B. trapezoidalis (Swain, 1974) 22 Hulingsina rugipustulosa (Edwards, 1944) Code Species 23 H. amencana (Cushman, 1906) 24 Bensonocythere calverti (Ulrich and Bassler, 1904) 25 B. rugosa, new species 27 B. blackwelden, new species 30 B. ricespitensis, new species 32 B. bradyi, new species 33 Bensonocythere sp. U 41 Hulsingsina sp. R 44 Bythocythere sp. B 47 Campylocythere laeva Edwards, 1944 54 Palaciosa minuta (Edwards, 1944) NUMBER .53 89 Code Species 55 Caudites paraasymmetricus, new species 56 Cneslocythere? sp. 57 Cyprideis sp. B 61 Hulingsina sp. U 62 Cytherella sp. A 63 Cy there lloidea sp. A 66 Cytheromorpha warneri Howe and Spurgeon, 1935 67 Cytheromorpha suffolkensis, new species 68 Cytheromorpha incisa, new species 69 Loxoconcha sp. S 70 Cytheromorpha macroincisa, new species 71 Cytheropteron talquinensis Puri, 1954 72 C?yorktownensis (Malkin, 1953) 75 Cytherura howei (Puri, 1954) 76 Cytherura sp. D 77 C. reticulata Edwards, 1944 78 Thaerocythere schmidtae (Malkin, 1953) 80 Cytherura wardensis Howe and Brown, 1935 82 Hulingsina sp. F 83 Cytherura sp. L 84 Cytherura sp. M 85 Cytherura sp. N 89 Echinocythereis planibasalis (Ulrich and Bassler, 1904) 90 E. leecreekensis, new species 91 Eucythere declivis (Norman, 1865) 92 E. gibba Edwards, 1944 94 E. triangulata Puri, 1954 96 Eucythere sp. F 98 Peratocytheridea bradyi (Stephenson, 1938) 99 P. sandbergi, new species 104 Peratocytheridea sp. J 105 Cytheridea aff. C. virginiensis (Malkin, 1953) 106 C. campwallacensis, new species 107 C. virginiensis (Malkin, 1953) 111 C. carolinensis, new species 114 Hirschmanma? quadrata, new species 115 H.? hespera, new species 121 Leptocythere nikraveshae Morales, 1966 125 Loxoconcha sp. C 126 L. purisubrhomboidea Edwards, 1953 127 L. edentonensis Swain, 1951 128 Hulingsina sp. C 130 Loxoconcha sp. H 131 L. reticularis Edwards, 1944 132 Cytherura forulata Edwards, 1944 133 Loxoconcha sp. M 136 L. matagordensis Swain, 1955 138 Microcytherura similis (Malkin, 1953) 139 M. choctawhatcheensis (Puri, 1954) 140 M. expanda, new species 141 Microcytherura sp. D 143 ''''Pontocythere" sp. I 145 Microcytherura sp. H 147 Microcytherura sp. M 149 M. minuta, new species Code Species 150 Muellerina ohmerti, new species 151 M. bassiounii, new species 152 M. wardi, new species 153 M. blowi, new species 155 ''Pontocythere" sp. j 156 Hulingsina glabra (Hall, 1965) 157 Muellerina canadensis petersburgensis, new subspecies 160 Muellerina sp. P 162 Murrayina macleani Swain, 1974 163 M. barclayi McLean, 1957 164 Murrayina sp. E 165 Radimella confragosa (Edwards, 1944) 166 Neocaudites trip lis triatus (Edwards, 1944) 167 N. variabilus, new species 169 Xestoleberis sp. E 172 Orionina vaughani (Ulrich and Bassler, 1904) 173 Proteoconcha jamesensis, new species 175 P. gigantica (Edwards, 1944) 177 Proteoconcha sp. Z 178 P. mimica Plusquellec and Sandberg, 1969 182 Paracyprideis sp. C 184 Paracypris sp. B 188 Paracytheridea cronini, new species 189 P. altila Edwards, 1944 190 P. rugosa Edwards, 1944 192 Paracytheroma stephensoni (Puri, 1954) 193 Cytherura sp. 194 Cytherura sp. U 196 Paradoxostoma delicata Puri, 1954 198 Paradoxostoma sp. E 200 Paranesidea.^ laevicula (Edwards, 1944) 201 Paranesidea sp. B 205 Propontocypris sp. D 206 Proteoconcha multipunctata, sensu lato 207 Leptocythere sp. E 215 Pseudocytheretta burnsi (Ulrich and Bassler, 1904) 216 Loxoconcha sp. T 217 Cushmanidea cf. C. seminuda (Cushman, 1906) 223 Paracytheridea mucra Edwards, 1944 224 Pterygocythereis inexpectata (Blake, 1929) 226 Puriana rugipunctata (Ulrich and Bassler, 1904) 227 P. carolinensis, new species 228 P. convoluta Teeter, 1975 229 P. mesacostalis (Edwards, 1944) 230 "Pontocythere" sp. G 236 Sclerochilus sp. B 238 Actinocythereis captionis, new species 239 A. marylandica (Howe and Hough, 1935) 240 A. mundorffi (Swain, 1951), small form 241 A. dawsoni (Brady, 1870) 243 Acuticythereis laevissima Edwards, 1944 250 Cytheromorpha.^ curta Edwards, 1944 257 Microcytherura sp. P 309 Neocaudites angulatus, new species 312 Microcytherura sp. R 90 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Code Species 316 Cytherura sp. W 317 C. elongata Edwards, 1944 318 Anchistrocheles sp. C 319 Bensonocythere sp. OO 320 Bensonocythere sp. PP 321 Cytherella sp. B 322 Cytheromorpha sp. I 323 Cytherura sp. AA 324 Peratocytheridea setipunctata (Brady, 1869) 326 Leptocythere sp. F 327 Paracytheridea sp. F 328 Neocaudites subimpressus (Edwards, 1944) 329 Bensonocythere sp. QQ 330 Bensonocythere sp. RR 331 Hermanites ascitus, new species 340 Cytherura sp. BB Biostratigraphy at the Lee Creek Mine Figure 2 shows two major clusters of samples, labelled I and II, and four principal subclusters, labelled A-D (sample 16 tends to cluster at a low level because of relatively low diversity; the same is true to a lesser extent of sample 1). Samples 1-9 from the Yorktown Formation are faunally more similar to each other than to samples 10-16 from the Croatan Formation. Thus, the major faunal discontinuity is be- tween samples 9 and 10, and the Yorktown-Croa- tan contact may represent the major hiatus in the section above the base of the Yorktown. This conclusion is contrary to that of Welby and Leith (1969) who stated that the major break in the part of the mine section treated in the present study is between Gibson's (1967) units 2 and 3 of the Yorktown, that is, between samples 6 and 7 of this study. Clusters A, B, C, and D indicate that recogniz- able faunal packages are found within each of the major clusters. Samples 1-7 from the lower 12 feet (3.6 m) of the Yorktown (units 1-3 of Figure 2) group together, and samples 8 and 9 from unit 4 from the upper part of the formation form another cluster. The latter two samples are each composites of samples containing similar faunas taken at virtually the same stratigraphic position but at different times. The interval below sample 8 and above sample 7 contained only a few poorly preserved ostracodes, none of which could be identified. Sample 10 was taken from a 4-foot (1.2-m) interval in a 12-foot (3.6-m) thick bed of differ- entially indurated burrowed sand (unit 5) as- signed to the Croatan; the irregularly shaped, indurated sandstone blocks are scattered through- out the unit but concentrated at the top. This sample groups faunally with those from the lower part (unit 6) of the very macrofossiliferous bed above. Samples from the upper part of this bed (unit 7) form subcluster D. The two parts of the bed are differentiated by the number of large mollusk shells that they contain: the lower part (unit 6) contains many, the upper part (unit 7), few. Small shells are abundant in both, but com- minuted shell is more common or predominant in unit 7. The upper few feet of section below nonmarine Pleistocene ("Cherry Point" unit of DuBar, Solliday, and Howard, 1974) is a dark blue unfossiliferous sandy clay (unit 8) in undu- lating contact with unit 7; its stratigraphic rela- tionship is unclear, and it is questionably referred to the Flanner Beach Formation (upper Pleisto- cene). Opposite the R-mode dendrogram in Figure 3, the ranges of the 85 taxa in the 16 samples are given. These taxa, as well as those that occur in all or only one of the samples, are listed alpha- betically and by code number (pp.85-90). At the top left of Figure 3 is a dendrogram summarizing the Q-mode results seen in Figure 2. For conven- ience in discussion, the R-mode subclusters of Figure 3 are labelled A-P, and the major clusters, I-IV Cluster I consists of species that occur chiefly in the Yorktown in the mine, and clusters II and IV are composed of species occurring principally in the Croatan. Cluster III is com- posed of species that occur in both the Yorktown and Croatan. Subcluster D consists of species restricted to the samples of Q-mode subcluster A. R-mode sub- cluster C contains species that mostly occur throughout Q-mode subcluster A and extend into B and, in part, also into C, whereas R-mode subcluster B represents species that are found in the stratigraphically higher samples of Q-mode subclusters A and B. Subcluster H consists of species that occur NUMBER 53 91 ■25359 T" Otsuka coefficient ^VNINO ^ 1^0 II 40 50 60 70 80 90 100 1 1 1 1 \ I 1 ■F 1— -^ rH C Unit 9 Unit 8 } Unit 7 Unit 6 Unit 5 "Cherry Point" Flanner Beach Formation SCALE 1ETERS 5-rl6 ''^^'^ 4- 3- 2- :— . — 25375 4 25358 COVERED c= EXPLANATION Indurated sand 25374 ( tJnit 4 24885 Clay <3.- — -^ Pebbles Pungo River Formation JL. ^f .ej. Unit 3 Unit 2 Unit I A A AAA A A Small and large macrofossils Phosphate FIGURE 2.—Location of collections, general lithology, and results of Q-mode cluster analysis of the Yorktown and Croatan samples collected in the southwestern part of the Lee Creek Mine. Units 1-9 are the major lithologic units; the 5-digit numbers to the immediate right of the strati- graphic column are USGS Cenozoic locality numbers; the collection points are indicated by dots, except for nonmicro- fossiliferous samples, which are indicated by X; the individ- ual and composite samples (1-16) used in the multivariate analysis are indicated to the left of the column. The faunal relationships are indicated by the dendrogram, which was obtained by an unweighted pair-group cluster analysis of a matrix of Otsuka similarity coefficients. This procedure dem- onstrated two major clusters, indicated by the numerals I and II, and four principal subclusters, A-D, with a less clearly marked subcluster, E. R - mode ( Otsuka coefficient 70 _L_ Q-mode (summary) ~n , E ID IC IB 161514131211109 8 7 6 5 4 3 2 1 330 228 78 16 156 115 153 83 165 151 140 143 18 114 309 312 111 138 85 82 70 236 67 68 155 90 167 94- 30 127 99 90 100 p 1 80 _L_ V N, X 1-00 60 50 40 30 FIGURE 3.—Results of an R-mode analysis of 85 ostracode species occurring in two or more but not all of the 16 samples of Figure 2. The ranges of the species in the samples are indicated in the body of the figure; a summary of the Q- mode analysis is given at the top left. The major clusters (I-IV) and subclusters (A-P) are arbitrarily labelled for convenience in discussion. The numbers in the vertical col- umn to the left of the dendrogram are the computer code numbers assigned to each species, which are numerically listed in the text. IV 240 107 238 40 I 1 07 ' .c38 I 23 I L 19 98 47 80 188 92 62 226 131 216 63 136 57 196 229 323 192 190 84 55 169 224 162 230 106 205 105 125 89 21 239 241 157 145 66 91 71 27 25 166 163 182 133 193 243 9 13 331 130 ' ^ III II NUMBER 53 93 through most of the section in the mine, and subcluster G, those that occur in the upper part of the Yorktown and the Croatan. Subclusters I through P of cluster IV contain species that are present in the samples of Q-mode clusters C and D, with some occurrences in Q- mode subcluster B. R-mode cluster II comprises species that are largely restricted to the upper part of the Croatan and are, therefore, responsible for Q-mode subcluster D. Relative Stratigraphic Positions of Lee Creek Beds CORRELATIONS WITH LOCALITIES TO THE NORTH Samples from the Yorktown and Croatan for- mations in the Lee Creek Mine have high simi- larity values with those from localities farther north in North Carolina and Virginia (Figure 4) and can be assigned to the three assemblage zones proposed by Hazel (1971a). Samples 1-4 from units 1 and 2 (Figure 2) can be confidently placed in the Pterygocythereis inex- pectata assemblage zone (Figure 4). The faunal change from the Pterygocythereis inexpectata assem- blage zone to the younger Orionina vaughani assem- blage zone is one of gradation, apparently mostly climatically controlled (Hazel, 1971b:372), and samples 5, 6, and 7 are intermediate in composi- tion. For example, they have a combined average similarity of 84.1 with samples 2, 3, and 4, and 81.1 with samples 8 and 9. Samples 5, 6, and 7 cluster with samples 2, 3, and 4 when the Lee Creek samples are analyzed separately, but they cluster with those of the Orionina vaughani assem- blage zone when Yorktown samples from other areas are added (Hazel, 197la:2-7). Samples 8 and 9 from unit 4 represent the Orionina vaughani assemblage zone. Sample 9 rep- resents beds not sampled in the previous study (Hazel, 197la:7). However, the presence of the species Murrayina barclayi (McLean, 1957), Echino- cythereis planibasalis (Ulrich and Bassler, 1904), and Actinocythereis dawsoni (Brady, 1870), suggests that unit 4 is no younger than the middle Orionina vaughani assemblage zone. The equivalent of the uppermost part of the classic Yorktown of the type area is seemingly missing at Lee Creek. If sample 10, which is from a 3- or 4-foot (0.9- or 1.2-m) interval in the upper middle part of unit 5, is representative of the assemblage of the entire unit, then this unit is early Puriana mesacos- talis assemblage zone in age, as based on its faunal similarity with the overlying samples and on the biostratigraphic fidelity values of the contained species (Hazel, 197la:5, 6). Units 6 and 7 are also placed in the Puriana mesacostalis assemblage zone; the distinct cluster (D) formed by samples 13-15 suggests that the assemblage zone is divisible, but this very probably only reflects ecological differ- ences between the upper and lower part of the bed. Units 5-7, as far as is known, are younger than deposits in the type area of the Yorktown, except for the one locality at Yadkin, Virginia, assigned to the Yorktown by Hazel (1971a, fig. 3), which has a P. mesacostalis assemblage zone. Also, the beds assigned to the Croatan here are correlated with those cropping out along the Cho- wan River in North Carolina in the vicinity of Colerain and Mt. Gould landings, which have been assigned to the Yorktown Formation by various authors. The beds along the Chowan River and those of the Puriana mesacostalis assem- blage zone locality at Yadkin must be investi- gated further to ascertain which formation (s) is represented. (These beds have recently been as- signed to the Chowan River Formation, named by Blackwelder (1981b) after this project had been completed.) CORRELATIONS WITH LOCALITIES TO THE SOUTH Many Yorktown and Croatan species are pres- ent in sediments of similar age to the south, although the ostracode assemblages are in general somewhat different in their overall aspect. Some of these species have limited stratigraphic ranges, and some concurrent range zones are useful in recognizing the chronozones of the three assem- blage zones of Hazel (1971a) and in correlating the Yorktown with formations in the Carolinas south of the Neuse River and in Georgia and Florida. 94 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TIME (M.Y.) Planktonic foraminifer zones N22 (part) Calcareous nannofossil zones NN19 (part) Ostracode assemblage zones Va & N. C. S. E. Virginia Chowan River Area N. Carolina Lee Creek Mine N. Carolina Upper Part Southeastern N. Carolina Northeastern S. Carolina Waccamaw Formation West Florida South Florida Caloosahatchee Formation (type) Orionina vaughani Yorktown Formation Mt. Gould and Colerain Yorktown Formation fflfflmmEMmfnfflim Bear Bluff Lower Part Formation of Dubar and Mill others (1974) Duplin Format! Yorktown Formation Jackson Bluff Formation Caloosahatchef at St. Petersburg [D moid] 111 m 'Pinecrest" / beds '' Tamiami Formation Pterygo-" cvthereis inexpeciaia FIGURE 4.—Suggested correlation of the Yorktown and Croatan formations at the Lee Creek Mine with other Coastal Plain lithostratigraphic units and with planktonic foraminifer and nannofossil zonations (time-scale and planktonic organism zones from Berggren and van Couvering, 1974). The planktonic zonations are shown for comparative purposes. Zone-diag- nostic planktonic microfossils are generally rare in the formations indicated in the figure. Care must be taken, however, in using certain kinds of species for the purpose of correlation in the Atlantic Coastal Plain. During Yorktown and Croatan time, the marine climate changed from possible mild-temperate to subtropical conditions in the region of Yorktown and Croatan outcrop (Hazel, 1971b:373), perhaps in response to the closing of the Isthmus of Panama, which would affect the Gulf Stream system (Berggren and Hollister, 1974:158, 175; Emiliani, Gartner, and Lidz, 1972; Casey, McMillen, and Bauer, 1975). A climatic shift in the same direction is to be expected in the southern part of the Atlantic Coastal Plain. Cryophilic species, then, should be expected to have longer stratigraphic ranges in the north than in the south; the reverse would be true for ther- mophilic species. Cognizance of the paleoclima- tologic framework can be very important in bio- stratigraphic interpretation. A large collection of samples from Coastal Plain units to the south has been made, but the studies are incomplete. A multivariate analysis of the data is planned that, it is hoped, will lead to the establishment of regionally useful assemblage zones, such as were proposed for the Virginia- northern North Carolina region (Hazel, 1971a). In addition, work is in progress on a paper delin- NUMBER 53 95 eating the most useful ostracode range and con- current range zones of the Pliocene and Pleisto- cene of the middle Atlantic Coastal Plain. This study and those of previous workers (Edwards, 1944; Pooser, 1965; Swain, 1968; Puri, 1953b; Puri and Vanstrum, 1971) suggest certain corre- lations, which are summarized below. Beds referable to the Pterygocythereis inexpectata ostracode assemblage zone, which in the mollus- can zonation of Mansfield (1929, 1943) would approximate the Placopecten clintonius zone (zone 1), are either uncommon or difficult to recognize in the southern Atlantic Coastal Plain. The Ray- sor Marl of Cooke (1936), known from one local- ity on the Edisto River, South Carolina, and now generally included in the Duplin Formation has been said to contain a fauna of this age. Washings from the original Raysor Marl collection of Cooke (1936) contain Pterygocythereis inexpectata (Blake, 1933), Pseudocytheretta burnsi (Ulrich and Bassler, 1904), Malzella evexa, new species; Actinocythereis marylandica (Howe and Hough, 1935), Cytherop- teron'^ yorktownensis (Malkin, 1953), Cytheridea vir- giniensis (Malkin, 1953), Muellerina ohmerti, new species, Muellerina wardi, new species, and other species. This temperate assemblage is of early Yorktown age (chronozone of the Pterygocythereis inexpectata assemblage zone or lower Orionina vaughani assemblage zone). The Duplin Forma- tion near Magnolia, North Carolina, contains a large ostracode assemblage suggestive of a middle or late Orionina vaughani assemblage zone age. It should be noted here that the statement attrib- uted to the writer in Berggren and Van Couver- ing (1974:125), that the Duplin correlates with rocks containing planktonic foraminifers of zone N12 (middle Miocene), contains an unfortunate typographical error; it should read "N19" (Pli- ocene) rather than "N12." Units 3, 4, and the upper part of 2 of the Yorktown in the Lee Creek Mine are placed in the Orionina vaughani assemblage zone. In these beds, Neocaudites triplistriatus (Edwards, 1944), Neo- caudites angulatus, new species, and N. subimpressus (Edwards, 1944), first appear, and several typical Yorktown forms (subclusters B and C of Figure 3) are last seen. The former all occur farther south, but few of the latter do. Malzella conradi (Howe and McGuirt, 1935), which ranges as high as the lower part of the Puriana mesacostalis assemblage zone, occurs in the lower Duplin, Jackson Bluff, Red Bay, and Tam- iami formations. Murrayina barclayi McLean, 1957, which occurs in the Orionina vaughani assemblage zone and older units, has been traced as far south as Orlando, Florida, where it occurs in rocks of Jackson Bluff age (also see Pooser, 1965:60). Ben- sonocythere rugosa, new species, occurs in the York- town and the lower part of the Croatan; in the Duplin Formation, and the Bear Bluff Formation of DuBar et al. (1974:156) in the Carolinas; and the Tamiami Formation in the subsurface of southern Florida. The equivalent of the Bear Bluff Formation of North and South Carolina of DuBar et al. (1974) is probably represented at Lee Creek by the lower part of the Croatan Formation (unit 5). Ostra- codes have been studied from the Bear Bluff Formation at Calabash, North Carolina, and from the subsurface near Bayboro, South Caro- lina. The Bear Bluff assemblage is very similar to that of the overlying Waccamaw Formation but contains the typical Yorktown-Duplin forms, Ben- sonocythere rugosa, new species, and Malzella conradi (Howe and McGuirt, 1935). The two species are known to extend only into the lower part of the Puriana mesacostalis assemblage zone. Both occur in sample 10 from the lower part of the Croatan, with an assemblage that is otherwise very similar to that of the upper part of the Croatan. According to DuBar et al. (1974:156-157) the Bear Bluff occurs primarily in the subsurface and is unconformable with the overlying Waccamaw and possibly with the underlying Duplin Forma- tion, where the latter is present; they indicated that the Bear Bluff macrofauna is transitional between that of the Duplin and Waccamaw for- mations. Units 6 and 7 of the Croatan in the Lee Creek Mine are placed in the Puriana mesacostalis assem- blage zone and contain an evolutionarily ad- vanced form (larger, more elongated) of Loxocon- 96 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY cha edentonensis (Swain, 1951), which is also known from the Waccamaw Formation. Caudites para- asymmetncus, new species (=C. sellardsi of Swain, 1968), occurs in unit 7. This distinctive species has been found previously only in the Waccamaw Formation and in its type locality of the Caloo- sahatchee Formation. It appears then that at least the upper part of the Croatan at the mine corre- lates with part or all of the Waccamaw and the type Caloosahatchee (DuBar, 1974:220, DuBar et al., 1974:164). CORRELATION OF THE YORKTOWN AND CROATAN FORMATIONS WITH A TIME SCALE The correlations suggested in the above discus- sion were based on benthic species endemic to North America. Because none of these taxa occur in stratotypes of the upper Tertiary stages of western Europe, they provide no data for direct correlation with European deposits (Waller, 1969:92). Estimates of age for the Yorktown and Croatan samples, therefore, must be based on (1) finding in these formations more mobile orga- nisms that have a wider distribution and may occur in Europe, or (2) biostratigraphic correla- tion with rocks that do contain such organisms, or (3) obtaining radiometric dates for the Lee Creek or demonstrably correlative deposits. Planktonic foraminifers are not diverse in the Yorktown at the Lee Creek Mine above the basal bed (Gibson, 1967:638; Akers, 1972:34). Gibson (1967:637) indicated that the planktonic assem- blage from the lowermost beds was of late Mio- cene aspect; he later (1971:10) concluded that the beds belonged to Blow's (1969) zone N16. This zone, according to Berggren and Van Couvering (1974, figs. 5, 11), is late Miocene in age and lasted from about 10.5 to 8.5 million years ago (mya). However, only one of the species listed by Gibson (1967:637) would be inconsistent with a younger age, and two of the species would be inconsistent with a pre-Pliocene age. Akers (1972) identified 17 species of planktonic foraminifers from the Yorktown at Rice's pit in Hampton, Virginia (loc. 11 of Hazel, 1971a: 11). He placed the Yorktown at the pit in zone N19, pointing out that the assemblage is essentially the same as that of the Jackson Bluff Formation of western Florida. The ostracode assemblage in the Yorktown at Rice's pit indicates placement in the Orionina vaughani assemblage zone (Hazel, 1971a, fig. 3). The Tamiami Formation in the subsurface of the Miami area also carries a zone N19 plank- tonic foraminifer assemblage (identified by M. Ruth Todd, U.S. Geological Survey), as well as ostracodes indicating correlation with the Jackson Bluff Formation (at least the Ecphora zone of the Jackson Bluff) and the Yorktown. Present in the Tamiami Formation are common Orionina vaugh- ani assemblage zone constituents such as Malzella conradi (Howe and McGuirt, 1935), Malzella evexa, new species, Orionina vaughani (Ulrich and Bassler, 1904), Bensonocythere rugosa, new species, Actinocy- thereis dawsoni (Brady, 1870), Cytheropteron? york- townensis Malkin, 1953, and Puriana rugipunctata (Ulrich and Bassler, 1904). According to Berggren (1973), zone N19 lasted from about 4.8 to 3.3 mya. The lowermost Yorktown beds at the Lee Creek Mine, however, are clearly older than those at Rice's pit (Hazel, 1971a, fig. 3). The Red Bay Formation of Puri and Vernon (1964) in western Florida belongs in planktonic zone N17, according to Akers (1972:13). The ostracode and molluscan assemblages of Puri and Vernon's Red Bay Formation (=Arca zone of older literature) also indicate that Red Bay is older than the Jackson Bluff Formation. Based on such mollusks as Chesapecten middlesexensis (Mansfield, 1929; Ward and Blackwelder, 1975; Druid Wilson, pers. comm., 1973), rocks below the Tamiami Formation in the subsurface of southern Florida can be correlated with the Red Bay and with the Eastover Formation of Virginia, which is stratigraphically below the Yorktown. The ostracode Otikocythere redbayensis (Howe and Brown, 1935) is known only from Puri and Ver- non's Red Bay Formation and from subsurface deposits on the Eastern Shore of Maryland judged to belong to the lower Eastover Formation. This evidence suggests that the Yorktown is younger than zone N17 which, according to Berggren (1972a, fig. 7; 1973), was about 8.5 to 5.0 mya. Berggren (1973) also presented evidence that the NUMBER 53 97 Miocene-Pliocene boundary is at about 5.0 mya, and that this, for all practical purposes, is equiv- alent to the base of zone N18. These findings lead to the conclusion that the lowermost beds of the Yorktown at the Lee Creek Mine are at least as young as zone N18 and are therefore Pliocene in age. Recently, Andrews (1980) has found the dia- tom species Thalassiosira oestrupii (Ostenfeld) in the upper part of the lower Eastover Formation (upper Claremont Manor Member). According to Andrews (1980:20, 22) this indicates that at least the upper part of the Claremont Manor Member and the overlying Cobham Bay Member of the Eastover Formation are also early Pliocene in age. On the basis of calcareous nannofossils, Akers and Koeppel (1973) concluded that the Yorktown at the Lee Creek Mine was the chronostrati- graphic equivalent of planktonic foraminifer zone N20 rather than N19. However, more recent work indicates that N20 is equivalent to upper N19 (for example, Poore, 1979). Glauconite, from a sample containing an as- semblage typical of the Orionina vaughani assem- blage zone (collected 18 feet (5.5 m) above the beach and 1.3 miles (2.1 km) below the mouth of Grove Creek on the left bank of the James River, James City County, Virginia), gives a K/Ar age of 4.4±0.2 my. This indicates an early Pliocene age for the sample and apparently corroborates the biostratigraphic dating of the Yorktown as early Pliocene (zone N19). Using the He/U method, Bender (1973) dated corals from the Caloosahatchee Formation of southern Florida. Five of the dates were based on specimens taken from the upper part of the Ca- loosahatchee (Bee Branch Limestone or Ayers Landing Members of DuBar, 1958) in the type area of the formation, and a sixth was based on a specimen from farther north at St. Petersburg. The five dates from the type area average 1.84 my and have an observed range of 1.78 to 1.89 mya. The upper part of the Croatan in the mine biostratigraphically correlates with the type Ca- loosahatchee and the Waccamaw Formation. The Caloosahatchee sample from St. Petersburg, which was dated at 2.53 my, contains a molluscan assemblage (Druid Wilson, pers. comm., 1974) indicative of an age younger than the "Pinecrest" beds of Olsson (1964), and several of the distinc- tive forms found in the Caloosahatchee in its type area are conspicuously absent. M.L. Bender, University of Rhode Island, in cooperation with the U.S. Geological Survey, has dated corals from the Waccamaw, Croatan, York- town, and other upper Cenozoic Coastal Plain units using the He/U technique (see Bender, 1973; Blackwelder, 1981a: 17, 24; 198 lb: 10). The results for some samples are important to the present study. Coral from the lower part of the Croatan (upper part of unit 5 from the north wall of the mine) gives a date of about 2.4 my. Coral from sample 18 of Hazel (1971a: 11; fig. 3) from the "Yorktown" near Mt. Gould Landing, North Carolina, gives a date of about 1.91 my. Hazel (197la:7) suggested that the "York- town" beds near Mt. Gould correlate with what is termed unit 6 in the Croatan Formation in the present study. The fossiiiferous beds from the lower part of the exposure at Colerain Landing are younger than the Yorktown Formation of the type area and the Lee Creek Mine, and older than those from near Mt. Gould; therefore, they are most probably correlated with the lower part of the Croatan (unit 5) of the mine. The radiometric data (Bender, 1973; in litt., 1975) coupled with the biostratigraphy suggest that the Croatan Formation and its correlatives were deposited between about 1.5 and 2.6 mya and that the contact between units 5 and 6 in the mine may approximate 2.0 my. The Yorktown Formation probably was deposited between about 2.6 and 4.8 mya. The youngest part of the Yorktown in the mine is apparently no younger than the middle Orionina vaughani assemblage zone. A 3.7- to 4.8-my age range for the Lee Creek Yorktown is not unreasonable. In connection with the placement of the Pli- ocene-Pleistocene boundary in the mine, it should be noted that there is considerable controversy as to the radiometric age of the Pliocene-Pleistocene boundary. The Pleistocene should be recognized in a manner similar to all other series of the 98 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Phanerozoic, that is, by correlation of localities with the type area by whatever techniques that give temporally meaningful correlations. There is little logic to the argument that the Pleistocene be recognized by climatic deterioration. At present, many authors accept the appear- ance of the planktonic foraminifer Globorotalia truncatulinoides, which apparently first appears in the type area of Calabria near the base of the Calabrian Stage, as evidence of the beginning of the Pleistocene. Berggren et al. (1967) presented the results of a study of a deep-sea core from the south-central Atlantic, in which micropaleonto- logic and paleomagnetic analyses were per- formed. Berggren indicated that the evolutionary transition from G. tosaensis to G. truncatulinoides occurred in this borehole and that the first evo- lutionary appearance of G. truncatulinoides was at 500 cm, paleomagnetically dated at 1.85 my. Later refinement by Berggren and others (Berg- gren and van Couvering, 1974:88) placed this event at 1.8 my, thereby providing the basis for this commonly cited date for the beginning of the Pleistocene. Parker (1973:280) also studied the foraminifers of the same Atlantic core and opined that at least some of the G. tosaensis specimens identified by Berggren were referable to a variant of G. crassa- formis called ''ronda.'' However, G. tosaensis and G. truncatulinoides occur in the core, and Parker's findings in effect indicate only that the Pliocene- Pleistocene boundary championed by Berggren is some 85 cm below where Berggren (1967) placed it. This suggests a revision of the date for the Pliocene-Pleistocene boundary from 1.8 to 2.0 mya. Not all workers are willing to accept that Glo- borotalia truncatulinoides is useful in marking the beginning of the Pleistocene or that the Pleisto- cene began at about 2.0 mya. In open-ocean sediments, however, abundant G truncatulinoides does seem to be a useful criterion, although plank- tonic foraminifers are generally not abundant in the outcropping sublittoral upper Cenozoic de- posits of the Atlantic Coastal Plain. If the suggested placement of the Pliocene- Pleistocene boundary in Figure 4 is correct, then the unconformity at about 2.8 my in the Atlantic Coastal Plain may correlate with the first cooling event documented in deep-sea cores (Beard, 1969; Berggren, 1972b). Locality Data The stratigraphic position and U.S. Geological Survey number of the Lee Creek Mine samples used in this study are indicated in Figure 2. The comparative material from elsewhere in the York- town Formation has been given by Hazel (1971a). Comparative material from farther south in the Coastal Plain in North Carolina, South Carolina, Georgia, and Florida consists of more than 300 U.S. Geological Survey collections from the Du- plin, Waccamaw, Bear Bluff, Jackson Bluff, Tam- iami, and Caloosahatchee formations and Ols- son's "Pinecrest" beds (Hazel, 1977; Cronin and Hazel, 1980; Cronin, 1980). Systematics Because of the large number of new species involved in the study, the writer has presented the systematic part of the paper as follows: With one exception, Peratocytheridea setipunctata (Brady, 1869), only the new species group taxa are treated in formal systematics. For these, a differential diagnosis, but no description as such, is presented. The diagnoses are supplemented by what the writer considers to be generally excellent scan- ning-electron photomicrographs presented as stereopairs. This approach is taken because the writer believes there is considerable redundant and nondiagnostic information in most ostracode species descriptions. Features that are general characteristics of the genus or family and those that can be clearly observed on photomicrographs need not be described. I believe that the diagnoses presented here, coupled with the illustrations, will be sufficient to indicate my concept of the taxa to other workers. Most of the species previously described from the Yorktown and Croatan and some that are left in open nomenclature are also illustrated. An alphabetical listing and a numerical computer NUMBER 53 99 code list of the taxa used in the computer analyses are presented (pp. 85-90); in the alphabetical list, occurrence data are followed by number of the plate and figure in which each species is illus- trated. A checklist of taxa treated formally in this report also follows, reflecting the hierarchic clas- sification used. Although all the ostracode subfamilies found in the Yorktown and Croatan have been studied and the species delineated by the author, descrip- tions have not been prepared for some of them; these new taxa are listed in open nomenclature. Major groups in this category are the loxocon- chids (except for Hirschmannia), and the cythe- rurids, Hulingsina, Cushmanidea, Neocytherideis, Leptocythere. This study is a contribution from a U.S. Geo- logical Survey program to document the Pliocene and Quaternary ostracodes of the Atlantic conti- nental margin (Hazel, 1967, 1968a, 1970, 1971a, 1971b, 1975a, 1975b; Hazel and Valentine, 1969; Valentine, 1971). The specimens used to illustrate the species were selected from the Yorktown and Croatan formations at the Lee Creek Mine, as well as from various other formations and modern samples. The locality data for the illustrated spec- imens are given in the figure descriptions. All illustrated specimens are deposited in the USNM collections of the National Museum of Natural History, Smithsonian Institution, Washington, DC. Abbreviations used in the tabulations of the dimensions are as follows: N, number of speci- mens measured; M, mean; sd, standard devia- tion; OR, observed range; and V, coefficient of variation. CHECKLIST Order PODOCOPIDA Miiller, 1894 Suborder PODOCOPA Sars, 1865 Superfamily CYTHERACEA Baird, 1850 Family CYTHERIDEIDAE Sars, 1925 Subfamily CYTHERIDEINAE Sars, 1925 Genus Cytheridea Bosquet, 1852 Cytheridea campwallacensis, new species Cytheridea carolinensis, new species Genus Peratocytheridea, new genus Peratocytheridea setipunctata (Brady, 1869) Peratocytheridea sandbergi, new species Family TRACHYLEBERIDIDAE Sylvester-Brad- ley, 1948 Subfamily TRACHYLEBERIDINAE Bradley, 1948 Tribe TRACHYLEBERIDINI Bradley, 1948 Genus Actinocythereis Puri, 1953 Actinocythereis captionis, new species Genus Neocaudites Puri, 1960 Neocaudites vanabilus, new species Neocaudites angulatus, new species Tribe PTERYGOCYTHEREIDINI Puri, 1957 Genus Pterygocythereis Blake, 1933 Pterygocythereis alophia, new species Subfamily HEMICYTHERINAE Puri, 1953 Tribe AURILINI Puri, 1974 Genus Malzella, new genus Malzella evexa, new species Tribe ECHINOCYTHEREIDINI Hazel, 1967 Genus Echinocythereis Puri, 1953 Echinocythereis leecreekensis, new species Tribe ORIONININI Puri, 1974 Genus Caudites Coryell and Fields, 1937 Caudites paraasymmetricus, new species Tribe COQUIMBINI Ohmert, 1968 Genus Muellerina Bassiouni, 1965 Muellerina ohmerti, new species Muellerina canadensis petersburgensis, new sub- species Muellerina bassiounii, new species Muellerina wardi, new species Muellerina blowi, new species Tribe THAEROCYTHERINI Hazel, 1967 Genus Thaerocythere Hazel, 1967 Thaerocythere carolinensis, new species Genus Hermanites Puri, 1955 Hermanites ascitus, new species Genus Puriana Coryell and Fields, 1953 Puriana carolinensis, new species Subfamily CAMPYLOCYTHERINAE Puri, 1960 Tribe CAMPYLOCYTHERINI Puri, 1960 Genus Proteoconcha Plusquellec and Sandberg, 1969 Proteoconcha jamesensis, new species Tribe LEGUMINOCYTHEREIDINI Howe, 1961 Genus Bensonocythere Hazel, 1967 Bensonocythere bradyi, new species Bensonocythere blackwelden, new species Bensonocythere gouldensis, new species Bensonocythere ricespitensis, new species Bensonocythere rugosa, new species Family CYTHERIDAE Baird, 1850 Subfamily CYTHERINAE Baird, 1850 Tribe CYTHERINI Baird, 1850 Genus Cytheromorpha Hirschmann, 1909 Cytheromorpha incisa, new species 100 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Cytheromorpha macroincisa, new species Cytheromorpha suffolkensis, new species Genus Microcytherura Muller, 1894 Microcytherura minuta, new species Microcytherura expanda, new species Family LOXOCONCHIDAE Sars, 1865 Genus Hirschmanma Elofson, 1941 Hirschmanma? hespera, new species Hirschmanma? quadrata, new species Family PARACYTHERIDEIDAE Puri, 1957 Genus Paracytheridea Muller, 1894 Paracytheridea cromni, new species Genus Cytheridea Bosquet, 1852 Cytheridea campwallacensis, new species PLATE 2: FIGURES 1, 3, 4 Anonocytheridea floridana (Howe and Hough).—Malkin, 1953: 784, pi. 79: figs. 29, 30. Cytheridea sp. B.—Hazel, 197la:6, table 1, species 73.— Swain, 1974:14, pi. 1: figs. 11, 12 [not pi. 1: fig. 4]. Cytheridea campwallacensis Hazel, 1977:378, figs. 3, 5f, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Many shallow nar- row fossae containing normal pore canals, thus possessing relatively smoother valve surfaces than Cytheridea virginiensis. On eight well-preserved specimens of C. campwallacensis, very short spines at the anterior and a single short spine in each valve at the posterior were observed. In contrast, C. virginiensis has 6 or 7 anterior spines on each valve and 2 spines connected by an extension of the valve between the spines (or a tab with spines at either end) at the posterior only in the right valve. The pattern of fossae is distinct from that shown by Cytheridea virginiensis (Plates 1,2). Sexual dimorphism is strong in C campwallacensis and weak in C. virginiensis. HOLOTYPE.—A female right valve (Plate 2: figure 4), USNM 172619, from the lower part of the Orionina vaughani assemblage zone of the York- town Formation on the James River, Virginia (sample VA-7 of Malkin, 1953, pi. 79: fig. 29). ETYMOLOGY.—From Camp Wallace, Virginia, on the James River, where the species occurs commonly. DIMENSIONS (in microns).—The height statistics are biased toward right valves; 14 of the 18 specimens preserved well enough to measure were right valves. Female Male Length Height Length Height N 11 11 7 7 M 825 480 879 477 sd 47 32 9 - OR 775-900 450-550 862-888 450-520 V 5.7 6.7 1.1 - AGE RANGE.—Early Pliocene. DISTRIBUTION.—Lower and middle part of the Yorktown, Pterygocythereis inexpectata and lower Or- ionina vaughani assemblage zones in Virginia and North Carolina. Thirty-five specimens were found. Cytheridea carolinensis, new species PLATE 3 Cytheridea sp. G.—Hazel, 1971a:6, table 1, species 76. Cytheridea carolinensis Hazel, 1977:376, figs. 3, 5e, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Smaller than C vir- giniensis, and with a more anterior anterodorsal angle, more weakly pitted surface, and fewer normal pore canals. Much smaller than C. camp- wallacensis, and the arrangement of normal pores is distinctly different. At least 5 anterior denticles; none were observed at the posterior. Three to 4 low ridges parallel the anterior margin in contrast to C. campwallacensis and C. virginiensis, in each of which there are only two. HOLOTYPE.—A female left valve (Plate 3: fig- ure 1), USNM 191357, from the Puriana mesacos- talis assemblage zone of the Croatan Formation at the Lee Creek Mine, North Carolina (sample 15). DIMENSIONS (in microns).— Female Male Length Height Length Height N 9 9 2 2 M 617 361 - sd 23 25 - OR 588-650 325-400 638-650 350-362 V 3.8 7 _NUMBER 53 101 AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Upper part of the Yorktown Formation in Virginia and the Croatan Forma- tion in North Carolina. More than 20 specimens have been found. REMARKS.—The surface pits are wider and deeper on the older (early Pliocene) specimens assigned to the species. Genus Peratocytheridea, new genus Peratocytheridea Hazel, 1977, figs. 3, 7h; table 1 [nomen nu- dum]. TYPE SPECIES—Cytheridea setipunctata Brady, 1869. DIFFERENTIAL DIAGNOSIS.—In adult specimens, ventral margin is concave in the anterior half; carapace widest in the posterior half; the hinge holomerodont (with reversal of hingement in some species); dorsal adductor muscle scar elon- gated toward the anterodorsal and posteroventral areas. REMARKS.—Distinct morphologic differences have been recognized between the North Ameri- can Miocene to Holocene species referred to Hap- locytheridea and the early Tertiary and late Cre- taceous species referred to the genus (Morkhoven, 1963:278-281; Hazel, 1968b: 126). The genus Peratocytheridea is proposed to accommodate some of these late Cenozoic species. The species here referred to Peratocytheridea are more broadly rounded at the posterior than those referred to Hap locy theridea. They are widest in the posterior half, whereas Hap locy theridea is com- pressed posteriorly. The ventral margin of Pera- tocytheridea is concave in the anterior half and that of Haplocytheridea, in the posterior half. The dorsal adductor muscle scar in Peratocytheridea is elon- gated toward the anterodorsal and posteroventral areas. In at least the type species of Hap locy theridea, the dorsal adductor is not elongated. Under high magnification small denticles can be seen on the outer margin of the selvage, as well as at the outer margin of the valve at the anterior and posterior. This last characteristic may or may not be diag- nostic. The following species are assigned to Perato- cytheridea : Cytheridea setipunctata Brady, 1869; Pliocene to Holocene. C. (Hap locy theridea) bradyi Stephenson, 1938; Pliocene to Holocene. C (H.) wadei Stephenson, 1938; Pliocene. Cytheridea kirkbii Brady, 1866; Holocene. C. subovata Ulrich and Bassler, 1904; late Oligo- cene to Miocene. Hap locy theridea bassleri Stephenson of Puri (1953b, pi. 3, figs. 1-3); Miocene. H. placentiaensis Teeter, 1975; Holocene. H. texana Stephenson, 1944; late Oligocene. Peratocytheridea sandbergi, new species; Pliocene. The taxonomic positions of the Miocene Carib- bean species assigned to Hap locy theridea by van den Bold (1965), the late Oligocene Gulf Coast forms studied by Butler (1963) and Poag (1974), and the younger early Miocene species "//." mans- fieldi (Stephenson, 1938) and ^'H.'' gardnerae (Ste- phenson, 1938) are not clear at present (see Sand- berg, 1964b). The species here assigned to Peratocytheridea have a combined chronostratigraphic range of upper Oligocene to Holocene. Hap locy theridea is known from the Upper Cretaceous to the upper Eocene and possibly Oligocene (Poag, 1972:68; 1974:49). The well-known species, Cytheridea setipunctata Brady, 1869, is chosen as the type species. The soft parts for Peratocytheridea setipunctata have been illustrated by Sandberg (1970, figs. 3, 5, 7, 9). Peratocytheridea setipunctata (Brady, 1869), new combination PLATE 4: FIGURE 4 Cytheridea setipunctata Brady, 1869:124, pi. 14: figs. 15, 16. Cytheridea (Hap locy theridea) ponderosa Stephenson, 1938:133, pi. 23: fig. 10; pi. 24: figs. 1, 2. Cytheridea (Leplocytheridea) sulcata Stephenson, 1938:139, pi. 23: fig. 2. Cytheridea puncticillata Brady.—Tressler and Smith, 1948:11 [partim], pi. 1: fig. 2 [two juvenile specimens on USNM slide 87319 are probably Peratocytheridea setipunctata; as Sandberg (1964b:362) points out, the male specimen in the collection is a Cyprideis]. 102 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Haplocythendea bassleri Stephenson.—Swain, 1955:617 [par- tim], pi. 59: fig. 9a [not pi. 59: fig. 9b, which is a Cyprideis amencana (Sharp, 1908)]. [?] Haplocythendea bassleri (Stephenson).—Puri and Hulings, 1957, fig. 11. Haplocythendea cf. H. ponderosa (Stephenson).—Curtis, 1960:482, pi. 2: fig. 9. Haplocythendea ponderosa (Stephenson).—Curtis, 1960:486, pi. 3: fig. 1. Haplocythendea bassleri (Stephenson).—Curtis, 1960:486, pi. 3: fig. 2. [?] Haplocythendea ponderosa (Stephenson).—Puri, 1960:110. [?] Haplocytheridea cf. H. nodosa (Stephenson).—Puri, I960: 110. Cyprideis floridana Puri, 1960:100, pi. 2: fig. 5. [?] Anomocytheridea cf. A. floridana (Howe and Hough).— Benda and Puri, 1962, pi. 3: fig. 32. Haplocytheridea gigantea Benson and Coleman, 1963:27, pi. 3: figs. 10-14, fig. 14. Haplocythendea setipunctata.—Sandberg, 1964a:507, pi. 3: fig. 12; 1964b:361, pi. 1: figs. 10-14, pi. 2: figs. 1-4 [not Haplocytheridea setipunctata (Brady)].—Williams, 1966:21, figs. 5-11, 16 [=Peratocytheridea bradyi (Stephenson), 1938]. —Morales, 1966:34, pi. 2: figs. 3a-c.—Hulings, 1967:643, fig. 3q.—Grossman, 1967:64a, pi. 11: figs. 4, 7, pi. 16:figs. 13-18 [not Haplocythendea setipunctata (Brady)].—Engle and Swain, 1967:413, pi. 2: fig. 15 [=Cyprideis amencana (Sharpe, 1908)].—Swain, 1968:7, pi. 1: figs. 5a-c; pi. 7: figs, la, b [=Peratocytheridea sandbergi, new species].—King and Kornicker, 1970:29, pi. 4: figs. 2a, b, pi. 13: figs. 9, 10, pi. lb: figs. 7, 8.—Sandberg, 1970, figs. 5, 7, 9 [soft parts].—Krutak, 1971:16, pi. 2: figs. 6a, b.—Valentine, 1971, pi. 2: figs. 48, 49 [not Haplocythendea setipunctata (Brady)].—Swain, 1974:12, pi. 9: fig. 16 [in part]. DIMENSIONS.—The illustrated female right valve is 1010 microns long and 610 high. AGE RANGE.—Pliocene to Holocene. DISTRIBUTION.—Chesapeake Bay to Laguna Terminos, Mexico, Puerto Rico, and the Baha- mas in the Holocene, primarily in estuaries and lagoons. Pleistocene to Holocene, Virginia-North Carolina region; Pliocene to Holocene in Florida. In the present study, the species is known only by one valve from sample 15 in the Croatan For- mation. Peratocytheridea sandbergi, new species PLATE 4: FIGURES 1, 2, 3 Haplocythendea bassleri (Stephenson).—Pooser, 1965:43, pi. 3: figs. 4-9. Haplocythendea setipunctata (Brady).—Swain, 1968:7, pi. 1: figs. 5a-c; pi. 7: figs, la, b; 1974:12, pi. 9: fig. 16 [in part]. ""Haplocythendea" sp. B.—Hazel, 1971a:6, table 1, species 72. Peratocytheridea sandbergi Hazel, 1977, figs. 3, 7h, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from the closely related Peratocytheridea setipunctata by its smaller size, subtle differences in arrangement of normal pores, and shape of the opaque areas of the valves as seen in transmitted light; normal pores of P. sandbergi are circular in outline whereas a mixture of circular and elongated pores is found in P. setipunctata. The smallest specimens of P setipunctata known to the writer occur in upper Pleistocene deposits in North Carolina and Vir- ginia. The females in these samples average about 962 microns in length and the males 1038 mi- crons. Both sexes of the living and fossil P. seti- punctata from Florida are consistently longer than 1000 microns. Peratocytheridea sandbergi females from the Pterygocythereis inexpectata assemblage zone and the higher Orionina vaughani assemblage zone of the Yorktown Formation average 852 microns in length; males are the same length. In the Croatan Formation, P sandbergi females are about the same size, but the males are somewhat smaller (about 781 microns). The anterodorsal angle in P. sandbergi is less acute in both females and males than in P setipunctata. HOLOTYPE.—A female left valve (Plate 4: fig- ure 1), USNM, 172663, from the Yorktown For- mation (sample 4). ENTYMOLOGY:—Named in honor of P.A. Sand- berg, University of Illinois. DIMENSIONS (in microns).—Pooled data; there are some changes in dimension in the males through time; see above. Male Female Length Height Length Height N 10 10 9 9 M 844 535 811 457 sd 24 24 41 31 OR 812-875 488-562 750-850 412-488 V 2.8 4.5 5.1 6.7 AGE RANGE.—Early Pliocene to Pleistocene. NUMBER 53 103 DISTRIBUTION.—Upper Pterygocythereis inexpec- tata assemblage zone through Puriana mesacostalis assemblage zone in Virginia and North Carolina. Waccamaw and Duplin formations in South Car- olina. Genus Actinocythereis Puri, 1953 TYPE SPECIES.—Cythere exanthemata Ulrich and Bassler, 1904. Actinocythereis captionis, new species PLATE 8: FIGURES 1, 2, 4 Cythereis exanthemata var. gomillionensis Howe and Ellis.—Ed- wards, 1944:521, figs. 31, 32. Actinocythereis exanthemata.—Puri, 1953a: 179, pi. 2: figs. 4, 6; 1953b:252, pi. 13: fig. 7.—Hulings, 1966:55, fig. 8h; 1967:655, fig. 7k.—Swain, 1968:14, fig. 12, pi. 2: fig. 5a-f. [Not Cythere exanthemata Ulrich and Bassler.] Actinocythereis exanthemata gomillionensis (Howe and Ellis).— McLean, 1957:83, pi. 10: figs. 2a-d.—Swain, 1974:31, pi. 4: fig. 22. Actinocythereis gomillionensis (Howe and Ellis).—Williams, 1966:30, figs. 6a-c, 24. Actinocythereis sp. B.—Hazel, 1971a:6, table 1, species 42. Actinocythereis aff. A. gomillionensis (Howe and Ellis).—Valen- tine, 1971, pi. 1: figs. 39, 40, 44, 48. Actinocythereis exanthemata exanthemata (Ulrich and Bassler).— Swain, 1974:30, [partim], pi. 5: fig. 2. Actinocythereis captionis Hazel, 1977:379, figs. 3, 8c, table 1 [nomen nudum].—Cronin and Hazel, 1980:18, figs. 6g, h [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Actinocythereis exanthemata (Ulrich and Bassler, 1904) by its smaller size, characteristic arrange- ment of bullate tubercles of the ventral and me- dian rows, higher and more evenly rounded pos- terior, and smoother surface between tubercles. Distinguished from A. gomillionensis (Howe and Ellis, 1935) by its smaller size, different arrange- ment of bullate tubercles in the ventral row, less inflated carapace, and ornamental details of the muscle node. The posterior bullate tubercles of the ventral and median rows in A. captionis are set close together and at nearly the same angle, whereas in A. gomillionensis they are more en eche- lon. HOLOTYPE.—A female carapace (Plate 8: figure 1), USNM 172461, from the Holocene Carolinian faunal province off Cape Fear, Holocene sample 2251; lat. 33°42.7' N, long. 78°45.0' W, 12 m depth. ETYMOLOGY.—Latin, captio, deception. DIMENSIONS.—The holotype measures 750 X 400 microns. AGE RANGE.—Early Pliocene to Holocene. DISTRIBUTION.—Cape Cod to Florida in the Holocene, common constituent of inner sublit- toral Pliocene and Pleistocene assemblages of the Atlantic Coastal Plain from Delaware to Florida. Actinocythereis captionis is a mild-temperate to sub- tropical species. Genus Neocaudites Puri, 1960 TYPE SPECIES.—Neocaudites neviami Puri, 1960 (=male o{ N. triplistriatus (Edwards, 1944). Neocaudites variabilus, new species PLATE 5: FIGURES 1-3; PLATE 7: FIGURE 1 Trachylebens? cf. T? trip It striata (Edwards).—Swain, 1951:37, pi. 6: figs. 2, 3. Orionina lienenklausi (Ulrich and Bassler).—Puri, 1953b:254, fig. 8d, pi. 12: fig. 14. Costa sp., aff. C. triplislnata.—HaW, 1965:33, pi. 7: fig. 8. Neocaudites sp. A.—Valentine, 1971, pi. 3: figs. 38, 42. Neocaudites variabilis Hazel.—Cronin and Hazel, 1980:23, fig. 8h [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Larger, more pro- duced at the posterior, and with a less well-de- veloped ventrolateral carina than Neocaudites tri- plistriatus (Edwards, 1944); surface may vary from nearly smooth to coarsely reticulate, whereas N. triplistriatus never has a well-developed reticulum. Distinguished from Neocaudites subimpressus (Ed- wards, 1944) by its larger size, more produced posterior, and absence of the concentric rows of large fossae paralleling the anterior. HOLOTYPE.—Female right valve (Plate 7: fig- ure 1), USNM 190494, from the Duplin Forma- tion near Magnolia, North Carolina, USGS 23639. ETYMOLOGY.—Latin, variabilus, changeable; 104 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY with reference to the variation of the reticulum. DIMENSIONS.—The holotype measures 650 X 338 microns; a female right valve measures 663 X 338 microns; a male paratype measures 725 X 375 microns. AGE RANGE.—Pliocene to Holocene. DISTRIBUTION.—Yorktown, Duplin, Norfolk, Waccamaw formations (Pliocene and Pleisto- cene) in the North Carolina-Virginia region; Jackson Bluff Formation (Pliocene) in Florida. In the Holocene, the species has been found in 6 samples off North and South Carolina. REMARKS.—The range of variation in this form is great and more than one species group taxon may well be present in what is here referred to Neocaudites vanabilus. However, specimens are not common in any one sample, and more material will be required to determine whether there are consistently different morphotypes present and what is their stratigraphic and geographic distri- bution. Neocaudites angulatus, new species PLATE 6: FIGURES 2-4 DIFFERENTIAL DIAGNOSIS.—Distinguished from Neocaudites triplistriatus (Edwards, 1944) by its larger size, more prominent ventrolateral carina, and by presence of a few large fossae rather than several small ones at the posterodorsal termina- tion of the median carina. HOLOTYPE.—Female carapace (Plate 6: figure 4), USNM 172742, from the Waccamaw Forma- tion at Old Dock, North Carolina (locality NC-4 of Swain, 1968). ETYMOLOGY.—Latin, angulatus, with angles; with reference to the pronounced ventrolateral carina, which in end view gives the carapace a more angulate form than is seen in other species. DIMENSIONS.—The holotype measures 675 X 375 microns; the illustrated right valve measures 725 X 362 microns. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Yorktown, Croatan, Duplin, and Waccamaw formations in North Carolina; it is not common in any sample. Genus Pterygocythereis Blake, 1933 TYPE SPECIES.—Cythereis jonesii Baird, 1850 Pterygocythereis alophia, new species PLATE 7: FIGURES 2, 4 Pterygocythereis cornuta amencana.—Puri, 1953b:261 [partim], pi. 13: figs. 2, 4 Pterygocythereis sp., aff. P. americana.—Benson and Coleman, 1963:22, pi. 5: figs. 2, 3 [partim].—Swain, 1968:19, fig. 18, pi. 2: figs. 7a-d. Pterygocythereis sp., cf P. howei.—Hulings, 1967:641, figs. 2f, 3m. Pterygocythereis sp. A.—Valentine, 1971:8. DIFFERENTIAL DIAGNOSIS.—Distinguished eas- ily from Pterygocythereis inexpectata by its smaller size and absence of fluted crests. Distinguished from P. miocemca by its smaller size, less sloping dorsal outline, and position, size, and shape of the tab on the posterior of the ala; in P. alophia, this structure is small and is a blunt spine, whereas in P. miocemca it is a broad thin tab. HOLOTYPE.—Female right valve (Plate 7: fig- ure 4), USNM 172642, from Holocene sample 1861 in Raleigh Bay off North Carolina, lat. 34°45.6' N, long. 75°44.6' W, at 41 m depth. ETYMOLOGY.—Greek, lophos, crest, with the neg- ative prefix a with reference to the absence of a crest. DIMENSIONS.—The holotype is 780 microns long and 415 high. AGE RANGE.—Early Pliocene to Holocene. DISTRIBUTION.— Orionina vaughani and Puriana mesacostalis assemblage zones in North Carolina, Bear Bluff Formation of DuBar et al. (1974:156), and Waccamaw Formation in the Carolinas, Jackson Bluff Formation in west Florida, and Tamiami Formation and "Pinecrest" beds of Ols- son (1964) in south Florida. From off Virginia to at least off southwestern Florida in the Holocene. REMARKS.—Van den Bold (1967) separated the uncrested Miocene and younger Coastal Plain and Caribbean species of Pterygocythereis from the crested Pterygocythereis amencana-P inexpectata lin- eage, referring the former to his P. miocenica. All the Pliocene and younger uncrested specimens NUMBER 53 105 seem to belong to one species, P. alophia, whereas the upper and middle Miocene forms belong to P. miocenica. Genus Malzella, new genus Malzella Hazel, 1977:376, 379; figs. 3, 6c, d; table 1 [nomen nudum]. TYPE SPECIES.—Malzella evexa, new species. DIFFERENTIAL DIAGNOSIS.—Dorsal precaudal ridge much better developed than in Radimella and continuous with the posterior subdorsal ridge. Dorsal ridge, which forms dorsal outline in lateral view, divided into anterior and posterior parts by intersection with short ridge homologous with dorsal end of anterosubdorsal ridge of Radi- mella. Carapace more evenly rounded dorsally and ventrally than in Radimella. Internal features as in Radimella except that posterior tooth is di- vided into 2 to 5 lobes. Surface more costate and more coarsely pitted than in Aurila; denticulate caudal process consistently present in Malzella, variably so in Aurila. Size within species quite variable, coefficients of variation near 7.0 com- mon in populations. REMARKS.—The following species are consid- ered to be representatives oi Malzella: "Aurila''^ bellegladensis Kontrovitz, 1978; Pleisto- cene. Hemicythere conradi Howe and McGuirt, 1935; late Miocene to Pliocene. Malzella evexa, new species; Pliocene to early Pleis- tocene. Aurila floridana Benson and Coleman, 1963, Pleis- tocene to Holocene. Aurila conradi littorala Grossman, 1965; Holocene. '^Aurila'", new species aff. A. conradi (Howe and McGuirt, 1935) of Howe and van den Bold (1975); early Holocene. Malzella is a common constituent of Pliocene, Pleistocene, and Holocene deposits of the middle and southern Atlantic Coast and the Gulf Coast. As a fossil, at least, it extended to Venezuela (Aurila conradi conradi of Rodriguez, 1969). The oldest species known is Malzella conradi (Howe and McGuirt, 1935), which occurs in rocks as old as late Miocene (Red Bay Formation of Puri and Vernon, 1964, in Florida). Although most com- monly found in modern and fossil marine shelf deposits, some species tolerate brackish and hy- persaline conditions. ETYMOLOGY.—Malzella is named in honor of Dr. Heinz Malz of the Senckenberg Museum, Frankfurt am Main, Germany. Malzella evexa, new species PLATE 14: FIGURE 3; PLATE 15: FIGURES 1-3, 5 Hemicythere conradi.—Edwards, 1944:518, pi. 86: figs. 17, 18.—Swain, 1951:42, pi. 6: figs. 9-12.—Puri, 1953:176, pi. 2: figs. 1, 2.—Brown, 1958:65, pi. 6: fig. 17. [Not Hemicythere conradi Howe and McGuirt, 1935.] Aurila conradi conradi (Howe and McGuirt).—Pooser, 1965:48, pi. 17: figs. 1, 2, 12, 13.—Swain, 1968:23, pi. 5: figs. 7a-i. Radimella floridana (Benson and Coleman).—Hazel, 1971a:6, table 1, species 9. Malzella evexa Hazel, 1977:376, 379, figs. 3, 6d, table 1 [nomen nudum].—Cronin and Hazel, 1980:15, fig. 4g [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Malzella conradi by its more evenly rounded dor- sum, more posteroventrally expanded ventrolat- eral carina, and by details of the arrangement of pores and fossae; distinguished from M. floridana by its smaller fossae and wider muri and less expanded ventrolateral carina. HOLOTYPE.—Female left valve (plate 14: figure 3), USNM 172653, from the Puriana mesacostalis assemblage zone of the "Yorktown" near Mt. Gould Landing on the Chowan River, North Carolina, USGS 24897 (sample 20 of Hazel, 1971a). ETYMOLOGY.—Latin, evexus, rounded at the top. DIMENSIONS.—The holotype measures 738 X 475 microns; the illustrated male left valve meas- ures 675 X 400 microns. AGE RANGE.—Early Pliocene to early Pleisto- cene. DISTRIBUTION.—Yorktown and Croatan for- mations in North Carolina; Bear Bluff of DuBar et al. (1974), Duplin, and Waccamaw formations in North and South Carolina; "Pinecrest" beds 106 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY of Olsson (1964) and Jackson Bluff, Tamiami, Caloosahatchee, and Bermont (of DuBar, 1974) formations in Florida. REMARKS.—Most of the fossil specimens oi Mal- zella from other than the Yorktown Formation referred to M. conradi are actually M. evexa; this apparently has resulted from most workers iden- tifying M. conradi by use of Edwards' (1944, pi. 86, figs. 17, 18) illustrations of specimens that represent the early, smaller form of M. evexa. Malzella evexa has not been found north of North Carolina. In table 1 of Hazel (1971a: 6), as the result of a clerical error, this taxon is indicated as occurring in sample 31, which is from the only known outcrop of the Puriana mesacostalis assem- blage zone in Virginia. The species does not occur in sample 31. The calculations in the paper are not affected. The specimens illustrated by Swain (1951:42; pi. 6, figs. 9-12) oi Hemicythere conradi Howe and McGuirt are Malzella evexa; however, Swain's (1951:43) unillustrated specimens, USNM 560753 and 560760, represent M. conradi (Howe and McGuirt, 1935). Populations of the species that have large fe- males, such as the holotype, occur in Puriana mesacostalis assemblage zone deposits. Genus Echinocythereis Puri, 1953 TYPE SPECIES.—Cythereis garretti Howe and McGuirt, 1935. Ejchinocythereis leecreekensis, new species PLATE 36: FIGURES 1 3, PLATE 38: FIGURE 3 Buntonia? ci. B.? garretti.—Swain, 1951:39 [partim], pi. 3: fig. 6. Echinocythereis evax.—Brown, 1958:65, pi. 6: fig. 12. Echinocythereis garretti.—Swain, 1968:15, fig. 13, pi. 4: fig. 12. Echinocythereis sp. A.—Valentine, 1971:6, table 1. Echinocythereis leecreekensis Hazel, 1977:378, figs. 3, lOh, table 1 [nomen nudum].—Cronin and Hazel, 1980:25, fig.. 9f [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—A large, very in- flated species with an evenly rounded posterior and, on well-preserved specimens, a prominent posteroventral spine on each valve. Distinguished from Echinocythereis margaritifera by its larger size, more evenly rounded posterior, better developed posteroventral spine, and more obvious concen- tric arrangement of spines. Distinguished from E. planibasalis by its spinosity, lack of a reticulum, and absence of the ventrolateral alar-like row of spines present in E. planibasalis. HOLOTYPE.—Female right valve (plate 36: fig- ure 1), USNM 191495, Croatan Formation, USGS 25359 (sample 10). DIMENSIONS.—The holotype measures 1013 mi- crons in length and 650 in height; the illustrated female left valve measures 1000 X 675, and the illustrated male right valve, 1013 X 575 microns. AGE RANGE.—Late Pliocene to Holocene. DISTRIBUTION.—Croatan Formation (Puriana mesacostalis assemblage zone) in North Carolina; Bear Bluff of DuBar et al. (1974) and Waccamaw formations in North and South Carolina; Pli- ocene deposits in cores taken offshore of Jackson- ville, Florida. In the Holocene, seemingly conspe- cific specimens have been found off the Atlantic Coast from the southwest side of the Georges Bank to Florida. Genus Caudites Coryell and Fields, 1937 TYPE SPECIES.—Caudites medialis Coryell and Fields, 1937. Caudites paraasymmetricus, new species PLATE 12: FIGURES 2-4 Caudites sellardsi (Howe and Neill).—Swain, 1968:22, pi. 6: figs. 2a, b. Caudites sp.—Puri and Vanstrum, 1971, fig. 4. Caudites asymmelricus Hazel, 1977:378, 380, figs. 3, 7f, table 1 [nomen nudum], Caudites paraasymmetricus Hazel.—Cronin and Hazel, 1980:6- 22, figs. 2, 8i; table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Caudites sellardsi (Howe and Neill, 1935) by the asymmetry of the valves in which a well-devel- oped vertically oriented carina, forming a right angle with the ventrolateral carinae, is present in the posterior area in the left valves but not in right valves where the ventrolateral carinae curve upward toward the posterodorsal area, and also NUMBER 53 107 by the position of the short dorsolateral carina, which is positioned more dorsally in C. paraasym- metricus than in C. sellardsi. Caudites para- asymmetricus can be distinguished from Caudites rectangularis (Brady, 1869), which has a similar valve asymmetry, by the virtual absence of any median carina in the former. Caudites paraasym- metricus is very similar to C. asymmelricus Pokornyi (1970), but the caudal process is more centrally located than in the latter and the ventral carina extends farther toward the anterior. DIMENSIONS.—The holotype measures 650 X 325 microns; the illustrated female right valve measures 625 X 312 microns; the illustrated male right valve measures 588 X 300 microns. AGE RANGE.—Early Pleistocene. REMARKS.—In some 30 specimens of Caudites paraasymmetricus available for study, all the right valves possessed one type of ornamentation and the left valves another. A few of the relatively shorter and more elongated specimens are inter- preted to be males; they possess the same valve asymmetry as those that must be the females. DISTRIBUTION.—Upper part of Croatan For- mation and the Waccamaw Formation in North Carolina; Caloosahatchee Formation in Florida. Genus Muellerina Bassiouni, 1965 TYPE SPECIES.—Cythere latimarginata Speyer, 1863. Muellerina ohmerti, new species PLATE 16: FIGURE 3 Trachyleberis? martini (Ulrich and Bassler).—Swain, 1951:29, pi. 3: figs. 8, 15. [?] Trachyleberis cf. T.? micula (Ulrich and Bassler).—Swain, 1951:29, fig. 3L. Trachyleberis martini (Ulrich and Bassler).—Malkin, 1953:793, [partim], pi. 82: fig. 10. Murrayina martini (Ulrich and Bassler).—Puri, 1953b:256, fig. 8e, f, pi. 12: figs. 11-13.—McLean, 1957:86, pi. II: figs. la-c, 2a, b, 3a-d; 1966:68, pi. 22: fig. 2. Murrayina micula (Ulrich and Bassler).—Williams, 1966:31, fig. 18-7, 25a, b. Murrayina canadensis (Brady).—Hulings, 1966:55, figs. 4f-h, 8g; 1967:654, figs. 4n, 7h. Muellerina lienenklausi (Ulrich and Bassler).—Hazel, 1967:21, pi. 3: figs. 3-6, 11; pi. 7: figs. 1, 4, 5, 7.—Swain, 1968:16 [partim], pi. 3: figs. 2a, b, ?c, e, f, ?g, h; 3a, c [not figs. 2d, 4a, = Muellerina bassiounii, new species; not figs. 3d, 4b, = M. wardi, new species; not fig. 3b, = a new species incom- pletely studied at present].—Hazel, 1968b: 1266, table 1. Muellerina aff. M. lienenklausi (Ulrich and Bassler).—Hazel, 1970, table 1.—Valentine, 1971, pi. 3: figs. 36, 40, table 1. Muellerina sp. A.—Hazel, 197la:6, table 1, species 28. Muellerina micula (Ulrich and Bassler).—Swain, 1974:38, pi. 7: figs. 1, 3-8; ?pl. 7: fig. 2. DIFFERENTIAL DIAGNOSIS.—Distinguished by its broad, evenly rounded anterior, acute anterodor- sal angle, and relatively well-developed pair of carinae posterior to the muscle node. Slightly depressed area is located posterocentrally and contains several fossae; depression is open towards the anteroventral area, and a row of fossae extend from the depression to about midlength. Mueller- ina ohmerti is smaller, more quadrate, and more broadly rounded at the anterior than is M. cana- densis (Brady, 1870). Muellerina wardi, new species also possesses a depression in the posterocentral area, but it is not open towards the anteroventral area. Muellerina ohmerti is also similar to M. bassi- ounii, new species, which also has a posterocen- trally located area of fossae, but which is dissected by short carinae. Shape and distribution of sur- face fossae also are important in distinguishing M. ohmerti from other species. In the anterior part, particularly, the fossae of M. ohmerti tend to be circular in outline and discrete. In M. bassiounii and M. wardi, many of the homologous fossae are coalesced, and others are elongated. Some of the soft parts of M. ohmerti have been illustrated (as M. lienenklausi) by Hazel (1967; pi. 7, figs. 1,4,5,7). HOLOTYPE.—Female left valve (Plate 16: figure 3), USNM 112741, from Holocene (sample 1287 of Hazel, 1967, 1970; Valentine, 1971). This spec- imen was also illustrated by Hazel (1967, pi. 3: fig. 4). ETYMOLOGY.—Named in honor of Dr. Wolf Ohmert, the author of the Subfamily Coquimbi- nae. DIMENSIONS.—The means for length and height for 22 Holocene female specimens are 682 microns and 377 microns, respectively, with observed ranges of 600-725 and 350-400 microns. The means for length and height for 16 Holocene 108 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY male specimens are 648 microns and 330 microns, respectively, with observed ranges of 625-675 and 312-362 microns. See also Hazel (1967:21). AGE RANGE.—Late Miocene (?) and Pliocene to Holocene. REMARKS.—Prior to 1967, when the writer re- ferred this species to Muellerina lienenklausi (Ulrich and Bassler, 1904), specimens of the taxon were being consistently assigned to another species in another family (Murrayina martini (Ulrich and Bas- sler, 1904)). However, although M. lienenklausi is undoubtedly congeneric, the type specimen is broken, and the Calvert Formation (lower and middle Miocene) has yielded specimens of more than one species oi Muellerina. At the present time it cannot be determined with complete certainty which of the Calvert or younger forms is M. lienenklausi. Therefore, M. ohmerti is proposed for the species referred to by the writer as M. lienen- klausi in 1967, and Muellerina sp. A in 1971. The specimens of Muellerina illustrated by Swain (1968, PI. 3) are all on a single hole micro- slide in the National Museum of Natural History. Four species are present on the slide. The queried identifications in the synonymy above result from not being able to determine in all cases which specimen represents which of Swain's illustra- tions. DISTRIBUTION.—From the Gulf of Maine to Florida in the Holocene; Virginia to Florida in the Pliocene; Pleistocene in Virginia, North Car- olina, and western Atlantic submarine canyons. Specimens possibly referable to M. ohmerti have been found in the upper Miocene so-called "St. Marys" Formation of Virginia. Muellerina canadensis petersburgensis, new subspecies PLATE 16: FIGURE 2, PLATE 18: FIGURES 1, 3 [?] Trachyleberis martini (Ulrich and Bassler).—Malkin, 1953: 793 [partim], pi 82: fig. 11. Muellerina aff. M. canadensis (Brady). — Hazel, 1971a:6, table 1, species 61. Muellerina canadensis petersburgensis Hazel, 1977:376, figs. 3, 10b, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Muellerina blowi by details of the reticulum, par- ticularly in the posterocentral area. Fossae in the posterocentral area of the valve in Muellerina can- adensis petersburgensis are arranged in distinct rows; in Muellerina blowi, the pits are more randomly arranged. Muellerina canadensis petersburgensis is con- sistently smaller than its descendant subspecies M. canadensis canadensis; otherwise, the two taxa are very similar, and no other consistent morpho- logic differences have been noted. Fifteen female specimens oi Muellerina canadensis petersburgensis have a mean length of 657 microns, whereas the mean length of 22 female specimens of M canadensis is 764 microns (Hazel, 1967:22). The student's t computed in comparing the means is 8.77, with 35 degrees of freedom, a value significant at less than 0.001. HOLOTYPE.—Female left valve (Plate 16: figure 2), USNM 167408, from the lower part of the Orionina vaughani assemblage zone in the York- town Formation at Petersburg, Virginia (sample 38 of Hazel, 1971a). ETYMOLOGY.—From the city of Petersburg, Virginia. DIMENSIONS (in microns).— Male Female Length Height Length Height N 15 11 2 2 M 657 344 sd 34 14 OR 612-700 323-362 625-675 312-325 V 5.1 4.1 - AGE RANGE.—Late Miocene to Pliocene. DISTRIBUTION.—Yorktown and Croatan for- mations in the Pterygocythereis inexpectata, Orionina vaughani, and lower part of the Puriana mesacostalis assemblage zones in Virginia and North Carolina. Upper Miocene so-called "St. Marys" Formation in Virginia. Muellerina bassiounii, new species PLATE 16: FIGURES 1, 4, PLATE 18: FIGURE 6 Muellerina lienenklausi (Ulrich and Bassler).—Swain, 1968:16 [partim], pi 3: figs. 2d, 4a. Muellerina sp. D.—Hazel, 1971a:6, table 1, species 7. NUMBER 53 109 N M sd OR V Muellerina bassiounii Hazel, 1977:378, figs. 3, lOd, table 1 [nomen nudum].—Cronin and Hazel, 1980:19, fig. 6e [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Posterocentral area dissected by 3 or 4 vertically oriented, sinuous carinae; in M. wardi, which is morphologically close to M. bassiounii, in the same area on the valves there is a spatulate fossa, narrowest dor- sally and either open or closed dorsally by bor- dering carinae. In the lower anterocentral area in M. bassiounii, usually 2 fossae coalesce to form a T-shaped compound fossa or a C-shaped fossa open towards the posterior. Muellerina bassiounii is also smaller than M. wardi. HOLOTYPE.—Female left valve (Plate 16: figure 4), USNM 167410, from the type locality of the Croatan Formation on the Neuse River, North Carolina. ETYMOLOGY.—Named in honor of Mohamed el Amin Ahmed Bassiouni of Cairo, Egypt, the au- thor of the genus Muellerina. DIMENSIONS (in microns).— Male Length 5 602 6 600-612 0.9 Height 5 295 7 288-300 2.32 Female Length Height N 16 16 M 642 330 sd 24 13 OR 588-675 312-350 V 3.8 3.9 AGE RANGE.—Late Pliocene to early Pleisto- cene. DISTRIBUTION.—Puriana mesacostalis assemblage zone in the "Yorktown," Croatan, and Wacca- maw formations in North Carolina. Muellerina wardi, new species PLATE 17: FIGURES 2, 4; PLATE 18: FIGURES 4, 5 Murrayina martini (Ulrich and Bassler).—Brown, 1958:65, pi. 3: fig. 3. Muellerina lienenklausi (Ulrich and Bassler).—Swain, 1968:16 [partim], pi 3: figs. 3d, 4b. Muellerina sp. E.—Hazel, 1971a:6, table 1, species 17. Muellerina wardi Hazel, 1977:376, figs. 3, 10a, table 1 [nomen nudum].—Cronin and Hazel, 1980:19, fig. 6c [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Muellerina bassiounii by a rather well-defined spa- tulate fossa located posterocentrally that is either closed or open dorsally (the narrow end); there is either a short isolated carina in the center of the fossa or the carina may be connected with the carina forming the anterior border of the fossa. In M. bassiounii, there is no clearly defined fossa in this area of the valve, but several vertically oriented, sinuous carinae. Muellerina wardi is sig- nificantly larger than M. bassiounii. HOLOTYPE.—Female left valve (Plate 17: figure 4), USNM 167409, from the Puriana mesacostalis assemblage zone in the "Yorktown" Formation near Mt. Gould Landing, North Carolina (sample 20 of Hazel, 1971a). ETYMOLOGY.—Named for L.W. Ward, U.S. Geological Survey, who supplied material used in this study. DIMENSIONS (in microns).— Male Female Length 2 625 Height 2 288-300 Length 16 683 19 650-725 2.7 Height 16 338 11 325-350 3.3 AGE RANGE.—Pliocene to early Pleistocene. OCCURRENCE.—Orionina vaughani and Puriana mesacostalis assemblage zones in Virginia and North Carolina. The species has been found in the Yorktown, Duplin, Croatan, Bear Bluff of DuBar et al. (1974), and Waccamaw formations in Virginia and the Carolinas, the "Pinecrest" beds of Olsson (1964) in south Florida, and Pliocene from cores offshore from Jacksonville, Florida. Muellerina blowi, new species PLATE 17: FIGURES 1,3; PLATE 18: FIGURE 2 Muellerina sp. F.—Hazel, 1971a:6, table 1, species 21. Muellerina blowi Hazel, 1977:376, figs. 3, 10c, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Differs from M. ohmerti, M. wardi, and M. bassiounii in that the no SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY valves are evenly convex in the posterocentral area with no large fossae or prominent carinae present. Muellerina canadensis (Brady, 1870) is sim- ilar in this respect, but the fossae of that species are aligned in distinct rows in the posterior, whereas in M. blowi there is only indistinct align- ment of a few of the fossae. In M. blowi, fossae are relatively smaller and the intervening murae wider than in other Muellerina species. Sexual dimorphism is weak, and sexing of individual specimens is difficult. HOLOTYPE.—Female left valve (Plate 17: figure 1), USNM 167411, from the Orionina vaughani assemblage zone of the Yorktown Formation at Suffolk, Virginia (sample 29 of Hazel, 1971a). ETYMOLOGY.—Named for Warren Blow, for- merly of the U.S. Geological Survey, who sup- plied valuable material and technical assistance during this study. DIMENSIONS.—Assumed females range from 625 to 688 microns in length and 325 to 350 microns in height. The assumed males range from 588 to 650 microns in length and 275 to 312 microns in height. AGE RANGE.—Pliocene to early Pleistocene. OCCURRENCE.—Pterygocythereis inexpectata, Orion- ina vaughani, and Puriana mesacostalis assemblage zones in Virginia and North Carolina, Yorktown, "Yorktown," and Croatan formations. Genus Thaerocythere Hazel, 1967 TYPE SPECIES.—Cythereis crenulata Sars, 1865. Thaerocythere carolinensis, new species PLATE 19: FIGURES 1, 3, 4 Trachyleberis?, cf T.? angulata (Sars).—Swain, 1951:29, pi 3: figs. 9-12. Hemicythere schmidtae Malkin.—Brown, 1958:66, pi 3: fig. 1. Thaerocythere sp.—Swain, 1974:40, pi. 7: fig. 16. Thaerocythere carolinensis Hazel, 1977:377, 378, figs. 3, 6f, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished by the presence of a low, wide, smooth tubercle located in the posterocentral area just behind and above the broad muscle tubercle; posterior ter- mination of the dorsolateral carina is also marked by a large swelling, and another is present in the posterocentral area. In these characteristics, T. carolinensis differs from the related species T. schmidtae (Malkin, 1953) in which the surface is regularly pitted and there are no prominent tu- bercles other than the muscle tubercle. HOLOTYPE.—Female carapace (Plate 19: figure 4), USNM 172657, from the Puriana mesacostalis assemblage zone in the "Yorktown" Formation at Colerain Landing, North Carolina (sample 14 of Hazel, 1971a). DIMENSIONS (in microns).—The data are pooled; the holotype measures 750 X 450 microns. Female Male Length Height Length Height N 10 10 4 4 M 735 446 679 409 sd 32 19 19 11 OR 700-800 425-488 675-712 400-425 V 4.3 4.4 2.7 2.9 AGE RANGE.—Late Pliocene to early Pleisto- cene. DISTRIBUTION.—Puriana mesacostalis assemblage zone in the "Yorktown" Formation in the Cho- wan River area of North Carolina and the Croa- tan Formation at the Lee Creek Mine. Swain (1974:41) indicated that Thaerocythere carolinensis occurs in the type Yorktown Formation in Vir- ginia (his sample S-5), referable to the Onomna vaughani assemblage zone. However, the writer has never seen this species in deposits of this age, and it is not present in the original material from collection S-5, which is in U.S. Geological Survey collections. Genus Hermanites Puri, 1955 TYPE SPECIES.—Plermama reticulata Puri, 1953. Hermanites ascitus, new species PLATE 11: FIGURES 1-3 Hermanites ascitus Hazel, 1977, figs. 3, 6h, table 1 [nomen nudum].—Cronin and Hazel, 1980:19, fig. 6f [ncmen nudum!. NUMBER 53 111 DIFFERENTIAL DIAGNOSIS.—Smaller than Her- manites reticulatus and with a less well-developed reticulum, which is nonfoveolate as compared with H. recitulatus; median carina traversing pos- terior half of valves also serves to distinguish the species; in H. ascitus, the ventrolateral carina is above the ventral outline in lateral view, whereas in H. reticulatus, this carina forms the ventral outline. HOLOTYPE.—Female left valve (Plate 11: figure 1), USNM 172745, from the Duphn Formation on the left bank of the Lumber River, near Lum- berton, North Carolina, 1.5 miles (2.4 km) south of intersection of routes 211 and 74. Sample taken 1 foot (0.3 m) below water level in gray shelly sand. DIMENSIONS.—The holotype is 600 microns long and 350 microns high; the male left valve measures 550 X 288 microns and the female right valve 575 X 312 microns. AGE RANGE.—Late Miocene to Pliocene. DISTRIBUTION.—Known from samples 8 and 9 of the present study, the Duplin Formation of North Carolina near Lumberton and at the Robe- son Farm near Tar Heel, Bear Bluff Formation of DuBar et al. (1974) near Bayboro, North Car- olina, and the upper part of the so-called "St. Marys" Formation of Virginia. Genus Puriana Coryell and Fields, 1953 TYPE SPECIES.—Favella puella Coryell and Fields, 1937 (= Cythereis rugipunctata var. gatunensis Coryefl and Fields, 1937). DIFFERENTIAL DIAGNOSIS.—Small, with a rela- tively strongly convex dorsum and concave ven- ter. Arrangement of carinae and short tubercles on the surface is characteristic. Behind the muscle node is a carina oriented parallel to the length; the ridge may be whole or divided into two parts. Anterior to the muscle node are two intersecting carinae forming a Y. In some populations, the surface carinae in P. carolinensis are quite thick relative to the intercarinal areas. Sexual dimor- phism indistinct; what are probably males are slightly smaller and slightly more elongated than females. A similar undescribed species is found in the Holocene off the Atlantic Coast, but that taxon is without the carina posterior to the muscle node, and although there is a homologous struc- ture to the Y-shaped compound ridge, the "Y" form is lost; in addition, P. carolinensis does not possess undercut ridges in the posterior half of the valves as does the Holocene species. Puriana mes- acostalis (Edwards, 1944) possesses a carina pos- terior to the muscle node also, but P carolinensis is much smaller and less quadrate. HOLOTYPE.—A female right valve (Plate 27: figure 1), USNM 172469, from the "Yorktown" Formation near Mt. Gould Landing, North Car- olina (sample 20 of Hazel, 1971a). ETYMOLOGY.—From North Carolina. DIMENSIONS (in microns).— Female Male Length Height Length Height N 12 12 2 2 M 554 291 - - sd 18 11 - - OR 525-575 275-312 500-525 250-275 V 3.2 3.7 - Puriana carolinensis, new species PLATE 27: FIGURES 1, 3, 4 Favella rugipunctata (Ulrich and Bassler).—Malkin, 1953:79, pi. 82, fig. 24. Puriana mesicostalis (Edwards) [sic].—Swain, 1968:19, pi. 5, fig. 13. Puriana sp. D.—Hazel, 1971a:6, species 10 (add samples 5, 6, 7, and 19. Puriana carolinensis Hazel, 1977:376, figs. 3, 8b, table 1 [nomen nudum].—Cronin and Hazel, 1980:15, fig. 4b [nomen nudum]. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Pterygocythereis inexpectata, Orion- ina vaughani, and Puriana mesacostalis assemblage zones; Yorktown, Croatan, Duplin, and Wacca- maw formations in North Carolina; Bear Bluff of DuBar et al. (1974) and Waccamaw formations in South Carolina; Cancellana and Ecphora zones of the Jackson Bluff Formation in western Flor- ida; "Pinecrest" beds of Olsson (1964), and Tam- iami Formation in southern Florida. 112 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Subfamily CAMPYLOCYTHERINAE Puri, 1960 In 1967, the writer considered the Campylo- cytherinae Puri, 1960, synonymous with the Leg- uminocythereididae Howe, 1961, at the subfam- ily level and placed this taxon in the Hemicy- theridae. This is no longer believed to be correct. The Leguminocythereinae and Campylocytheri- nae are still believed to be closely related, but now the writer considers them as separate tribes within the subfamily Campylocytherinae. The genera Campylocythere and Acuticythereis, both proposed by Edwards (1944), and Proteocon- cha Plusquellec and Sandberg, 1969, were mono- graphed by Plusquellec and Sandberg in a very useful paper. The reader is referred to Plusquellec and Sandberg (1969) for diagnoses and morpho- logic details of the Campylocytherini. These taxa, plus the less well understood southern forms of Climacoidea Puri, 1956, and Reticulocythereis Puri, 1960, constitute the North American Campylo- cytherini. The genera Leguminocythereis Howe and Law, 1936, Triginglymus Blake, 1951, Anticythereis van den Bold, 1946, and Bensonocythere Hazel, 1967, constitute the Leguminocythereidini. The last ge- nus is the only representative of the subfamily to occur in the middle Miocene to Holocene inter- val. Bensonocythere seems to be endemic to the Atlantic Coast of North America, where it is found in sediments deposited in cold-temperate to subtropical waters. Chrysocythere Ruggieri, 1962, which was in- cluded with this general group by the writer in 1967 (p. 26), is actually a thaerocytherine, and Basslerites Howe, 1937, is perhaps a buntonid. Lemocythere Howe, 1951, which may be a senior synonym of Pseudocytheromorpha Puri, 1957, may belong to the Leguminocythereidini but is in need of more study. Genus Proteoconcha Plusquellec and Sandberg, 1969 TYPE SPECIES.—Proteoconcha proteus Plusquellec and Sandberg, 1969 (= Acuticythereis nelsonensis Grossman, 1967). Proteoconcha jamesensis, new species PLATE 25: FIGURES 1, 2; PLATE 27: FIGURE 2 Proteoconcha sp. E.—Hazel, 1971a:6, species 19. DIFFERENTIAL DIAGNOSIS.—Higher relative to its length than most other species of Proteoconcha, except P. gigantica (Edwards, 1944), a larger and otherwise distinctive species. More tumid in the posteroventral area than other species of the ge- nus. Males much less acuminate at the posterior than other species, except P. gigantica and P. redbayensis (Puri, 1953). Surface can be smooth or pitted, the pitting being mainly in the posterior half of the valves. Arrangement of normal pore canals (all sieve type) is most like that seen in P. nelsonensis (Grossman, 1967) but differs in detail. HOLOTYPE.—Female left valve (Plate 25: figure 1), USNM 167377, in the Orionina vaughani assem- blage zone of the Yorktown Formation, USGS 24622 (sample 32 of Hazel, 1971a). ETYMOLOGY.—From the James River of Vir- ginia. DIMENSIONS.—The holotype is 700 microns long and 425 microns high. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.^Onorzma vaughani and Puriana mesacostalis assemblage zones in Virginia and North Carolina. Genus Bensonocythere Hazel, 1967 (= Prodictyocythere Swain, 1974) TYPE SPECIES.—Leguminocythereis whitei Swain, 1951. REMARKS.—When this genus was proposed in 1967, the writer had no idea of its persistence in Miocene to Holocene Atlantic Coast deposits. In that paper, only the northern Holocene assem- blages from the Atlantic shelf were studied, and only three species of Bensonocythere were recog- nized. At that time, Miocene, Pliocene, or Pleis- tocene Atlantic Coast assemblages had not been studied in any detail, and the same was true of Holocene assemblages south of New Jersey. At least 17 species are present in the Pliocene NUMBER 53 113 to Holocene deposits of the Middle Atlantic Coast. Other species, as yet not fully studied, are found in the middle and upper Miocene deposits. Bensonocythere, the coquimbine genus Muellerina, and the thaerocytherine Puriana, characterize and make distinctive the sublittoral fossil and living cold-temperate to subtropical assemblages of the western North Atlantic. The word "persistence" was used above pur- posely, because often in contrast to Muellerina and Puriana, individual species oi Bensonocythere seldom dominate an assemblage or even number among the abundant forms of it. However, in sublittoral marine to polyhaline assemblages several Benson- ocythere species are commonly present. Identification is not easy, as is also the case with the Campylocytherini. Size and shape are, of course, important; however, many species are similar in this respect. The normal pore-canal positions are one very diagnostic feature of Ben- sonocythere species; however, because most species have a coarse reticulum, the pores are not readily discernible at normal working magnifications in incident or transmitted light. Furthermore, the fossae on fossil and Holocene specimens often are filled with sediment. Fortunately the fossae them- selves are covariant with the normal pores; thus, these can be homologized from specimen to spec- imen to delineate species. The height and width of the muri between the pits are variable, but the positions of these are also consistent among con- specific individuals. The recently proposed genus Prodictyocythere (Swain, 1974), is, in this writer's opinion, a syn- onym of Bensonocythere. Bensonocythere bradyi, new species PLATE 34: FIGURES 1, 2; PLATE 38, FIGURES 2, 4 Bensonocythere sp. R.—Hazel, 1971a, table 1, species 1. Bensonocythere bradyi Hazel, 1977:376, figs. 3, 9g, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Bensonocythere ricespitensis by its larger size and details of the reticulum; for example, the three coalescing fossae immediately anterior to the sub- central tubercle are not present or are ill defined in B. ricespitensis; the circular arrangement of pits just below the anterior to the subcentral tubercle is also different in B. ricespitensis. In size and shape, B. bradyi is similar to B. gouldensis, but the pattern of fossae is distinctly different. HOLOTYPE.—Male right valve (Plate 34: figure 1), USNM 167380, from the Puriana mesacostalis assemblage zone on the Chowan River near Mount Gould Landing, North Carolina (sample 17 of Hazel, 1971a). ETYMOLOGY.—Named for the 19th- and early 20th-century English ostracode worker, G.S. Brady. DIMENSIONS.—The holotype is 850 microns long by 375 microns high; the illustrated male left valve measures 850 X 400 microns; the illus- trated female right valve (interior view) measures 900 X 425 microns. DISTRIBUTION.—Common in the Orionina vaughani and Puriana mesacostalis assemblage zones in Virginia and North Carolina and rare in the Pterygocythereis inexpectata assemblage zone. REMARKS.—In the Puriana mesacostalis assem- blage zone, specimens assigned to this species possess wider muri relative to the fossae than in populations from the older zones. Compare plates 34 and 38. Bensonocythere blackwelderi, new species Plate 35: FIGURES 1, 2, 4; PLATE 37: FIGURE 4 Bensonocythere sp. J.—Hazel, 1971a, table 1, species 75. Bensonocythere sp. K.—Hazel, 1971a, table 1, species 17. Bensonocythere sp. G.—Valentine, 1971, pi 1: fig. 23, table 1. DIFFERENTIAL DIAGNOSIS.—Distinguished from Bensonocythere ricespitensis by its more evenly rounded anterior and posterior; more convex (particularly in females) rather than straight dor- sum; the arrangement of fossae and muri over the surface also distinguishes B. blackwelderi from B. ricespitensis, particularly that in the posterior half of the valves and in the anteroventral area. Ben- sonocythere blackwelderi is smaller than B. bradyi, and the arrangement of fossae is distinctly different. HOLOTYPE.—Female left valve (Plate 35: figure 114 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 4), USNM 167398, from the Orionina vaughani assemblage zone in the Yorktown Formation in Virginia near Tormentor Creek, a tributary to the James River in Isle of Wight County, Virginia (sample 33 of Hazel, 1971a). ETYMOLOGY.—Named in honor of B.W. Black- welder, formerly of the U.S. Geological Survey. DIMENSIONS.—The holotype is 781 microns long and 425 microns high. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Pterygocythereis inexpectata (rare), Orionina vaughani, and Puriana mesacostalis assem- blage zones in Virginia and North Carolina. The forms with particularly wide muri have been found only in the older two zones in the Yorktown Formation. The specimen illustrated by Valen- tine (1971) as Bensonocythere sp. G from upper Pleistocene deposits near Yadkin, Virginia, is probably a reworked Orionina vaughani assemblage zone form. JBensonocythere gouldensis, new species Plate 34: FIGURES 3, 4; PLATE 37: FIGURES 2, 3 Leguminocythereis whitei Swain, 1951:43 [partim], pi. 3: fig. 18; pi. 4: fig. 1.—Brown, 1958:63, pi, 6: fig. 10. Bensonocythere sp. B.—Hazel, 1971a, table 1, species 44. Bensonocythere gouldensis Hazel, 1977:376, figs. 3, 4a, table 1 [nomen nudum].—Cronin and Hazel, 1980:17, fig. 5a [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from the closely related Bensonocythere blackwelderi by its less evenly rounded anterior and by the slightly different pattern of fossae, particularly in the posterior half of the valves. In B. gouldensis, the first interior carina paralleling the anterior valve margin is close to but distinctly inside the anterior and anteroventral outline, whereas in B. black- welderi, this carina tends to form the outline; the muri are narrower in B. gouldensis than in B. blackwelderi. HOLOTYPE.—Female left valve (Plate 37: figure 2), USNM 167385, from the Puriana mesacostalis assemblage zone of the "Yorktown" Formation on the Chowan River, North Carolina, near Mt. Gould Landing (sample 19 of Hazel, 1971a). ETYMOLOGY.—From Mt. Gould, North Caro- lina. DIMENSIONS.—The female holotype measures 758 X 375 microns. Females are relatively shorter but absolutely larger; the largest measured female is 875 microns long. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Yorktown and Croatan for- mations in Virginia and North Carolina; Duplin Formation in North Carolina. Bensonocythere ricespitensis, new species Plate 33 Leguminocythereis whitei Swain, 1951:43 [partim], pi. 3: fig. 16. Bensonocythere sp. N.—Hazel, 1971a, table 1, species 79. Bensonocythere ricespitensis Hazel, 1977:376, figs. 3, 4c, table 1 [nomen nudum].—Cronin and Hazel, 1980:17, fig. 5b [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Bensonocythere bradyi by its smaller size and char- acteristic arrangement of fossae, particularly in the area below and slightly anterior to the muscle node where there is a circular pattern of about 6 fossae; three of these are in a row subparallel to the greatest valve length, the anterior fossae slightly higher than the posterior one; below these are 3 or 4 fossae in a semicircle, open dorsally. Bensonocythere sapeloensis (Hall, 1965) has a similar arrangement of fossae in this part of the valve, but B. ricespitensis is larger, and the fossae pattern in the posterior of the valves is quite different. HOLOTYPE.—Female left valve (Plate 33: figure 1), USNM 167394, from the Orionina vaughani assemblage zone at Mr. William Rice's marl pit in the Yorktown Formation at Hampton, Vir- ginia, USGS 24907 (sample 36 of Hazel, 1971a). DIMENSIONS.—The female holotype measures 750 X 400 microns. Males are about the same length but not as high. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Orionina vaughani and Puriana mesacostalis assemblage zones in Virginia and North Carolina in the Yorktown and Croatan NUMBER 53 115 formations; Waccamaw Formation in North Car- olina. Bensonocythere rugosa, new species PLATE 32: FIGURES 3, 4; PLATE 37: FIGURE 1 Bensonocythere sp. I.—Hazel, 1971a, table 1, species 51. Bensonocythere rugosa Hazel, 1977:376, 378, figs. 3, 4b, table 1 [nomen nudum].—Cronin and Hazel, 1980:17, fig. 5c [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Bensonocythere whitei (Swain, 1951) by its concave dorsum, convex venter, more oblique anterior and details of the reticulum. Distinguished from a morphologically similar undescribed late Plio- cene to Holocene species (Bensonocythere sp. M. of Hazel, 1971a), which does not occur at Lee Creek, by its larger size, weaker sexual dimorphism, and details of the reticulum, particularly in the pos- terior part of the valves where the fossae are difficult to homologize. In both B. rugosa and Bensonocythere sp. M. of Hazel (1971a), the fossae are developed mainly between major vertically oriented muri and separated by minor muri; in B. rugosa, these minor muri are more weakly developed. HOLOTYPE.—Female left valve (Plate 32: figure 4), USNM 167382, from the Orionina vaughani assemblage zone of the Yorktown Formation at Petersburg, Virginia, USGS 24909 (sample 39 of Hazel, 1971a). ETYMOLOGY.—Latin, rugosus, wrinkled. DIMENSIONS.—The holotype measures 688 X 388 microns; the other illustrated female right valve measures 700 X 350 microns; the illustrated male right valve measures 675 X 325 microns. AGE RANGE.—Pliocene. DISTRIBUTION.—Yorktown Formation in Vir- ginia and North Carolina; lower part of Croatan and Duplin formations in North Carolina; Bear Bluff Formation of DuBar et al. (1974) in North and South Carolina; Tamiami Formation in Flor- ida. Genus Cytheromorpha Hirschmann, 1909 TYPE SPECIES.—Cythere fuscata Brady, 1869. Cytheromorpha incisa, new species PLATE 21: FIGURES 1, 2; PLATE 23: FIGURES 5, 6 Cytheromorpha sp. D.—Hazel, 1971a:6, species 48. Cytheromorpha incisa Hazel, 1977, figs. 3, 9a, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Cytheromorpha macroincisa by its smaller size, strongly developed reticulum with large fossae, wider but slightly shorter posteroventral incised area, more elongate posterior hinge tooth, and details of normal pore arrangement; distin- guished from Cytheromorpha warneri Howe by the presence of the posteroventral incised area, which is absent in C. warneri warneri Howe and Spurgeon, 1935, and only weakly developed in C. warneri newportensis Williams, 1966. HOLOTYPE.—Female left valve (Plate 21: figure 1), USNM 172560, from the Puriana mesacostalis assemblage zone of the "Yorktown" Formation, near Mt. Gould Landing, North Carolina, USGS 24895 (sample 18 of Hazel, 1971a). ETYMOLOGY.—Latin, incisus, cut into, with ref- erence to the posteroventral incised area of the valves. DIMENSIONS.—The female holotype measures 550 microns long and 300 microns high; the illustrated male measures 563 X 275 microns. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Throughout the Yorktown and Croatan formations in Virginia and North Car- olina. Cytheromorpha macroincisa, new species PLATE 22: FIGURES 1-5 Cytheromorpha macroincisa Hazel, 1977:377, 378, figs. 3, 9c, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Cytheromorpha incisa by its pitted rather than reti- culate surface, by the more narrow but longer incised area in posterior part of the ventrolateral surface of the valves; and by details of normal pore arrangement. Distinguished from C. suffolk- ensis by its more broadly rounded posterior, dif- ferent arrangement of punctae and small carinae, 116 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY and the different shape, placement, and angle of the incised area, which is set farther forward in C macroincisa. HOLOTYPE.—Female left valve (Plate 22: figure 1), USNM 172562, from the Puriana mesacostalis assemblage zone of the "Yorktown" Formation near Mt. Gould Landing, North Carolina, USGS 24895 (sample 18 of Hazel, 1971a). ETYMOLOGY.—Latin, macro and incisus, long and cut, with reference to the narrow elongated in- cised area in the valves. DIMENSIONS.—The holotype is 613 microns long and 325 microns high; the illustrated female right valve measures 600 X 312 microns. AGE RANGE.—Pliocene. DISTRIBUTION.—Puriana mesacostalis assemblage zone in North Carolina, with one possible occur- rence in the Orionina vaughani assemblage zone on the Virginia Eastern Shore; "Yorktown" and Croatan formations on the Chowan River in North Carolina. Cytheromorpha suffolkensis, new species PLATE 23: FIGURES 1-4 Cytheromorpha sp. C.—Hazel, 1971a:6, species 56. Cytheromorpha warneri Howe and Spurgeon.—Swain, 1974 [partim], pi 4: fig. 2. Cytheromorpha suffolkensis Hazel, 1977:376, figs. 3, 9d, table 1 [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Cytheromorpha macroincisa by its smaller size, differ- ently shaped and positioned incised area on ven- trolateral surface, which is set farther to the pos- terior in C suffolkensis. Surface punctae similar in size to C. macroincisa, but the arrangement of punctae and small carina at anterior and poste- rior is different. Easily distinguished from C. incisa by its punctate rather than reticulate surface. HOLOTYPE.—Female left valve (Plate 23: figure 1), USNM 172551, from the Orionina vaughani assemblage zone of the Yorktown Formation at Suffolk, Virginia, USGS 24814 (sample 29 of Hazel, 1971a). ETYMOLOGY.—From Suffolk, Virginia. DIMENSIONS.—The holotype is 525 microns long and 275 microns high; the illustrated male right valve measures 550 X 288 microns. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Orionina vaughani and Puriana mesacostalis assemblage zones in Virginia and North Carolina, Yorktown, "Yorktown" on the Chowan River, and Croatan formations. Genus Microcytherura Miiller, 1894 (= Tetracytherura Ruggieri, 1952) TYPE SPECIES.—Microcytherura nigrescens Muller, 1894. Microcytherura minuta, new species PLATE 31: FIGURES 1-3 Microcytherura sp. O.—Hazel, 1971a, table 1, species 84. Microcytherura minuta Hazel, 1977:376, figs. 3, 5b, table 1 [nomen nudum].—Cronin and Hazel, 1980:21, fig. 7g [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Smaller than other known species of Microcytherura; evenly rounded at the anterior and posterior, whereas M. expanda is more obliquely rounded and expanded (or inflated) at the posterior; reticulum developed with very fine pits in the fossae, in contrast to M. expanda, M. similis, and M. choctawhatcheensis, which have larger pits in the fossae. HOLOTYPE.—Female left valve (Plate 31: figure 3) USNM 172750, from the Orionina vaughani as- semblage zone in the Yorktown Formation at Williamsburg, Virginia (sample 45 of Hazel, 1971a). DIMENSIONS.—The holotype is 488 microns long and 375 high; the illustrated female right valve measures 525 X 262 microns; and the illus- trated male left valve is 513 X 250 microns. DISTRIBUTION.—Yorktown Formation, Pterygo- cythereis inexpectata (one record) and Orionina vaughani assemblage zones, in Virginia and North Carolina; "Yorktown" on the Chowan River, Croatan, and Duplin formations and Puriana mes- acostalis assemblage zone in North Carolina. NUMBER 53 117 Microcytherura expanda, new species PLATE 30: FIGURES 1-3 Microcytherura expanda Hazel, 1977:377, 378, figs. 3, 5c, table 1 [nomen nudum].—Cronin and Hazel, 1980:21, fig. 7f [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Microcytherura similis by its inflated and broadly rounded posterior and better developed reticu- lum. The intensity of the development of the reticulum is similar in M. choctawhatcheensis, but M. expanda in general has fewer pits present in the fossae, the positions of the fossae are different, and the posterior is more broadly rounded and inflated. Larger than M. minuta and with a higher posterior and pits in the fossae. HOLOTYPE.—Female left valve (Plate 30: figure 2), USNM 172752, from the "Yorktown" For- mation on the Chowan River, North Carolina, near Mt. Gould Landing, USGS 24897 (sample 20 of Hazel, 1971a). ETYMOLOGY.—Latin, expando, spread out. DIMENSIONS.—The holotype is 575 microns long and 312 high; the illustrated male left valve measures 550 X 300 microns; the illustrated fe- male right valve is 537 X 300 microns. AGE RANGE.—Late Pliocene to early Pleisto- cene. DISTRIBUTION.—"Yorktown" Formation on the Chowan River, North Carolina; Croatan and Waccamaw formations also in North Carolina. Genus Hirschmannia Elofson, 1941 TYPE SPECIES.—Cythere viridis Miiller, 1785. Hirschmannia? hespera, new species PLATE 20: FIGURES 1, 2; PLATE 21: FIGURES 3, 4 Hirschmannia? sp. B.—Hazel, 197la:6, species 15. DIFFERENTIAL DIAGNOSIS.—Easily distin- guished from Hirschmannia viridis by its more elon- gated valves. Distinguished from H..^ quadrata by the fact that it is relatively much higher at the anterior than the posterior; the surface is covered with small punctae, whereas the surface of H..^ quadrata is covered with relatively large fossae with second-order reticulation separated by low muri. HOLOTYPE.—Female left valve (Plate 20: figure 1), USNM 172575, from sample 14 of the present study. ETYMOLOGY.—Latin, hesperus, western, with ref- erence to this being a western Atlantic member of the genus. DIMENSIONS.—The holotype is 475 microns long and 250 microns high; the illustrated female right valve (Plate 20: figure 2) measures 488 X 250 microns. AGE RANGE.—Pliocene. DISTRIBUTION.—Orionina vaughani and Puriana mesacostalis assemblage zones in Virginia and North Carolina, Yorktown, "Yorktown," and Croatan formations. Hirschmannia? quadrata, new species PLATE 20: FIGURES 3, 4 DIFFERENTIAL DIAGNOSIS.—Distinguished from Hirschmannia.'^ hespera by its less tapered posterior, more inturned ventral margin when viewed in- ternally, and presence of a surface ornamentation of fossae with secondary reticulation separated by lower muri. The last characteristic, coupled with a straighter ventral outline, distinguishes the spe- cies from H. viridis. HOLOTYPE.—Female left valve (Plate 20: figure 3), USNM 172571, from Puriana mesacostalis as- semblage zone of the "Yorktown" Formation near Yadkin, Virginia, USGS 24905 (sample 31 of Hazel, 1971a). ETYMOLOGY.—Latin, quadratus, squared. DIMENSIONS.—The holotype is 475 microns long and 250 microns high; the illustrated male right valve, USNM 172573, measures 488 X 250. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Onomna vaughani and Puriana mesacostalis assemblage zones of Virginia and North Carolina, including the subsurface of the Virginia Eastern Shore. 118 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Genus Paracytheridea Miiller, 1894 TYPE SPECIES.—Paracythendea depressa Miiller, 1894. Paracytheridea cronini, new species PLATE 28; FIGURES 1, 2; PLATE 29: FIGURE 1 Paracytheridea nodosa.—Swain, 1951:51 [partim], pi. 3: fig. 22. Paracytheridea sp. A.—Hazel, 1971a:6, table 1, species 40. Paracytheridea mucra.—Swain, 1974:20, pi. 1: fig. 23. Paracytheridea edwardsi Hazel, 1977:376, figs. 3, 7a, table 1 [nomen nudum].—Cronin and Hazel, 1980:23, fig. 8d [nomen nudum]. DIFFERENTIAL DIAGNOSIS.—Distinguished from Paracytheridea mucra Edwards, 1944, by its smaller size, narrower posterior, better developed reticu- lum, and much less well-developed ala. Distin- guished from P. altila Edwards, 1944, by its less pointed posterior, narrower muri, presence of a weakly developed sulcus, more broadly rounded anterior, and position of and trend of various carinae. Distinguished from P. rugosa by its less well-developed reticulum, more rounded poste- rior, and position of and more weakly developed carinae. HOLOTYPE.—Female left valve (Plate 29: figure 1), USNM 172759, from the type locality of the Duplin Formation at Natural Well near Magno- lia, North Carolina, USGS 23639. (See Edwards, 1944.) ETYMOLOGY.—Named in honor of T.M. Cronin, U.S. Geological Survey. DIMENSIONS.—The holotype is 575 microns long and 300 microns high. AGE RANGE.—Pliocene to early Pleistocene. DISTRIBUTION.—Yorktown Formation in Vir- ginia and North Carolina, Duplin and Croatan formations in North Carolina, Bear Bluff For- mation of DuBar et al. (1974) in South Carolina. Literature Cited Akers, W.H, 1972. Planktonic Foraminifera and Biostratigraphy of Some Neogene Formations, Northern Florida and Atlantic Coastal Plain. Tulane Studies in Geology and Paleontology, 9: 139 pages. Akers, W.H., and P.E. Koeppel 1973. Age of Some Neogene Formations, Atlantic Coastal Plains and Mexico. In L.A. Smith and J. Hardenbol, editors. Proceedings of the Symposium on Calcareous Nannofossils, Twenty Third Annual Meeting, Gulf Coast Association of Geological Societies and Society of Economic Paleontologists and Mineralogists, pages 80-84. Andrews, G.W. 1980. Neogene Diatoms from Petersburg, Virginia. Mi- cropaleuntolog}', 26:17 48. Beard, J.H. 1969. Pleistocene Paleotemperature Record Based on Planktonic Foraminifers, Gulf of Mexico. Trans- actions of the Gulf Coast Association of Geological Socie- ties, 19:535-553. Benda, W.K., and H.S, Puri 1962. The Distribution of Foraminifera and Oslracoda off the Gulf Coast of the Cape Romano Area, Florida, Transactions of the Gulf Coast Association of Geological Societies, 12:303-341, 5 plates. Bender, M.L. 1973. Helium-Uranium Dating of Corals. Geochimica et Cosmochimica Acta, 37:1229-1247. Benson, R.H., and G.L. Coleman 1963. Recent Marine Ostracodes from the Eastern Gulf of Mexico. University of Kansas Paleontological Contri- butions, .4rthropoda, 2:1-52, 8 plates, 33 figures. Berggren, W.A. 1972a, A Cenozoic Time-Scale—Some Implications for Regional Geology and Paleobiogeography. Le- thaia, 5(2): 195-215, 9 figures. 1972b. Late Pliocene-Pleistocene Glaciation. Initial Reports of the Deep Sea Dnlling Project, 8:953-963. 1973. The Pliocene Time Scale: Calibration of Plank- tonic Foraminiferal and Calcareous Nannofossil Zones. Nature, 243:391-397. Berggren, W.A., and CD. Hollister 1974. Paleogeography, Paleobiogeography and the His- tory of Circulation in the Altantic Ocean. In W.W. Hay, editor, Studies in Paleo-oceanography. Society of Economic Paleontologists and Mineralogists, Special Publication (Tulsa), 20:126-186. Berggren, W.A., J.D. Phillips, A. Bertels, and D. Wall 1967. Late Pliocene-Pleistocene Stratigraphy in Deep- NUMBER 53 119 Sea Cores from the South-Central North Atlantic. Nature, 216:253-255. Berggren, W.A., and J.A. Van Couvering 1974. The Late Neogene: Biostratigraphy, Geochronol- ogy and Paleoclimatology of the Last 15 Million Years in Marine and Continental Sequences. Pa- laeogeography, Palaeoclimatology, and Palaeoecology, 16(1/2): 1-216, 15 figures, 12 tables. Blackwelder, B.W. 1981a. Late Cenozoic Stages and Molluscan Zones of the U.S. Middle Atlantic Coastal Plain. Paleontological Society Memoir, 12 {fournal of Paleontology, 55(5), Supplement): 1-35, 10 plates, 9 figures, 1 table. 1981b. Stratigraphy of Upper Pliocene and Lower Pleis- tocene Marine and Estuarine Deposits of North- eastern North Carolina and Southeastern Vir- ginia. U.S. Geological Survey Bulletin, 1502-B:1-16, 1 plate, 6 figures. Blow, W.H. 1969. Late Middle Eocene to Recent Planktonic Fora- miniferal Biostratigraphy. In P. Bronnimann and H.H. Renz, editors, Proceedings of the First Interna- tional Conference on Planktonic Microfossils, Geneva, 1967, 1:199-422, 54 plates, 43 figures. Leiden: E.J. Brill. Brady, G.S. 1866. On New or Imperfectly Known Species of Marine Ostracoda. Zoological Society of London Transactions, 5:359-393, plates 57-62. 1869. [Description of Ostracoda from] la Nouvelle-Prov- idence. In A.G.L. de Folin and L. Perier, editors, Les Fonds de la mer, 1(26): 122-126, plate 14. Paris. Brown, P.M. 1958. Well Logs from the Coastal Plain of North Caro- lina. North Carolina Department of Conservation and Development Bulletin, 72: 68 pages, 8 plates. Butler, E.A. 1963. Ostracoda and Correlation of the Upper and Mid- dle Frio from Louisiana to Florida. Louisiana De- partment of Conservation, Geological Survey Bulletin, 39: 100 pages. Casey, R.E., K.H. McMillen, and M.A. Bauer 1975. Evidence for and Paleoceanographic Significance of Relict Radiolarian Populations in the Gulf of Mexico and Caribbean (abstract). Geological Society of America Abstracts with Programs, 7(7): 1022. Cheetham, A.H., and P.B. Deboo 1963. A Numerical Index for Biostratigraphic Zonation in the Mid-Tertiary of the Eastern Gulf Transac- tions of the Gulf Coast Association Geological Societies, 13:139-147. Clark, W.B., B.L. Miller, L.W. Stephenson, and B.L. John- son 1912. The Coastal Plain of North Carolina. North Caro- lina Geological and Economic Survey, 3: 552 pages. Cooke, C.W. 1936. Geology of the Coastal Plain of South Carolina. U.S. Geological Survey Bulletin, 867: 196 pages. Cronin, T.M. 1980. Biostratigraphic Correlation of Pleistocene Marine Deposits and Sea Levels, Atlantic Coastal Plain of the Southeastern United States. Quaternary Re- search, 13:213-229, 6 figures, 3 tables. Cronin, T.M., and J.E. Hazel 1980. Ostracode Biostratigraphy of Pliocene and Pleis- tocene Deposits of the Cape Fear Arch Region, North and South Carolina. U.S. Geological Survey Professional Paper, 1125-B: 1-25, 9 figures, 1 table. Curtis, D.M. 1960. Relation of Environmental Energy Levels and Ostracod Biofacies in East Mississippi Delta Area. American Association of Petroleum Geologists Bulletin, 44:471-494, 3 plates, 17 figures, 1 table. Dall, W.H. 1892. On the Marine Pliocene Beds of the Carolinas. In W.H. Dall, Part II: Introductory o/Contributions to the Tertiary Fauna of Florida: Tertiary Mol- lusks of Florida. Wagner Free Institute of Science Transactions, 3(2):201-217. DuBar, JR. 1958. Age and Stratigraphic Relationship of the Caloos- ahatchee Marl of Florida. Transactions of the Illinois State Academy of Science, 50:187-193. 1974. Summary of the Neogene Stratigraphy of South- ern Florida. In R.Q. Oaks, Jr., and J.R. DuBar, editors, Post-Miocene Stratigraphy, Central and Southern Atlantic Coastal Plain, pages 206-231. Logan: Utah State University Press. DuBar,J.R.,H.S.Johnson, Bruce Thom, and W.O, Hatchell 1974. Neogene Stratigraphy and Morphology, South Flank of the Cape Fear Arch. In R.Q. Oaks, Jr., and J.R. DuBar, editors, Post-Miocene Stratigraphy. Central and Southern Atlantic Coastal Plain, pages 139- 173. Logan: Utah State University Press. DuBar, JR., and J.R. Solliday 1963. Stratigraphy of the Neogene Deposits, Lower Neuse Estuary, North Carolina. Southeastern Geol- ogy, 4{4):2l3-233. DuBar, J.R, J,R. Solliday, and J.R. Howard 1974. Stratigraphy and Morphology of Neogene Depos- its, Neuse River Estuary, North Carolina. In R.Q. Oaks, Jr., and J.R. DuBar, editors, Post-Miocene Stratigraphy, Central and Southern Atlantic Coastal Plain, pages 102-122. Logan: Utah State Univer- sity Press. Edwards, R.A. 1944. Ostracoda from the Duplin Marl (Upper Mio- cene) of North Carolina. Journal of Paleontology, 18(6):505-528, plates 85-88. 120 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Emiliani, C, S. Gartner, and B. Lidz 1972. Neogene Sedimentation on the Blake Plateau and the Emergence of the Central American Isth- mus. Palaeogeography, Palaeoclimatology, Palaeoecology, 11(1): 1-10. Engel, P.C, and F.M. Swain 1967. Environmental Relationships of Recent Ostracoda in Mesquite, Aransas and Copano Bays, Texas Gulf Coast. Transactions of the Gulf Coast Association of Geological Societies, 17:408-427, 2 plates, Gibson, T.G. 1967. Stratigraphy and Paleoenvironment of the Phos- phatic Miocene Strata of North Carolina. Geolog- ical Society of America Bulletin, 78:631-659. 1970. Late Mesozoic-Cenozoic Tectonic Aspects of the Atlantic Coastal Margin. Geological Society of Amer- ica Bulletin, 81:1813-1822. 1971. Miocene of the Middle Atlantic Coastal Plain. In R.E. Gernant, T.G. Gibson, and F.C. Whitmore, Jr., Environmental History of Maryland Miocene. Maryland Geological Survey Guidebook, 3:1-16. Gibson, T.G., and W.M. Walker 1967. Flotation Methods for Obtaining Foraminifera from Sediment Samples. Journal of Paleontology, 41(5): 1294-1297. Grossman, S. 1967. Ecology of Rhizopodea and Ostracoda of South- ern Pamlico Sound Region, North Carolina, Part 1: Living and Subfossil Rhizopod and Ostracode Populations. University of Kansas Paleontological Con- tributions, 44 (ecology, article l):7-82, 21 plates, 17 figures, 13 tables. Hall, D.D. 1965. Paleoecology and Taxonomy of Fossil Ostracoda in the Vicinity of Sapelo Island, Georgia. In R.V. Kesling, D.G. Darby, R.N. Smith, and D.D. Hall, Four Reports of Ostracod Investigations (National Sci- ence Foundation Project GB-26), 79 pages, 20 plates. Hazel, J.E. 1967. Classification and Distribution of the Recent Hemicytheridae and Trachyleberididae (Ostra- coda) off Northeastern North America. U.S. Geo- logical Survey Professional Paper, 564:1-49, 11 plates, 2 figures, 1 table. 1968a. Ostracodes from the Brightseat Formation (Dan- ian) of Maryland. Journal of Paleontology, 42:100- 142, plates 21-26, 17 figures, 19 tables. 1968b. Pleistocene Ostracode Zoogeography in Atlantic Coast Submarine Canyons. Journal of Paleontology, 42(5): 1264-1271, 3 figures, 2 tables. 1970. Binary Coefficients and Clustering in Biostratig- raphy. Geological Society of America Bulletin, 81:3237- 3252, 7 figures, 1 table. 1971a. Ostracode Biostratigraphy of the Yorktown For- mation (Upper Miocene and Lower Pliocene) of Virginia and North Carolina. U.S. Geological Survey Professional Paper, 704:1-13, 6 figures, 1 table. 1971b. Paleoclimatology ofthe Yorktown Formation (Up- per Miocene and Lower Pliocene) of Virginia and North Carolina. In H.J. Oertli, editor, Paleoecol- ogie des Ostracodes. Centre de Recherches Pau-SNPA Bulletin, 5 (suppl.):361-375. 1975a. Ostracode Biofacies in the Cape Hatteras, North Caroline, Area. In F.M. Swain, editor. Biology and Paleobiology of Ostracoda. Bulletins of American Paleontology, 65(282):463-488. 1975b. Patterns of Marine Ostracode Diversity in the Cape Hatteras, North Carolina, Area. Journal of Paleontology, 49(4): 731-734. 1977. Distribution of Some Biostratigraphically Diag- nostic Ostracodes in the Pliocene and Lower Pleis- tocene of Virginia and Northern North Carolina. U.S. Geological Survey Journal of Research, 5(3):373- 388, 10 figures, 1 table. Hazel, J.E., and P.C. Valentine 1969. Three New Ostracodes from off Northeast North A.menca. J ournal of Paleontology, 43(3):741-752. Howe, H.V. 1941. Use of Soap in the Preparation of Samples for Micropaleontologic Study. Journal of Paleontology, 15(6):691. Howe, H.V., and W.A. van den Bold 1975. Mudlump Ostracoda. In F.M, Swain, editor, Bi- ology and Paleobiology of Ostracoda. Bulletins of American Paleontology, 65(282):303-315. Hulings, N.C, 1966, Marine Ostracoda from Western North Atlantic Ocean off the Virginia Coast. Chesapeake Science, 7(l):40-56, 8 figures, 1967, Marine Ostracoda from the Western North Atlan- tic Ocean: Labrador Sea, Gulf of St, Lawrence and off Nova Scotia. Crustaceana, 13(3):629-659, 7 figures. King, CE., and L.S. Kornicker 1970. Ostracoda in Texas Bays and Lagoons: An Ecol- ogic Study. Smithsonian Contributions lo Zoology, 24:1- 92, 21 plates, 15 figures, 19 tables. Krutak, P.R. 1971. The Recent Ostracoda of Laguna Mandinga, Veracruz, Mexico. Micropaleontology, 17:1-30, 4 plates, 10 figures, 6 tables, MacNeil, F,S, 1938. Species and Genera of Tertiary Noetinae. US. Geological Survey Professional Paper, 189-A: 1-49. Malkin, Doris 1953. Biostratigraphic Study of Miocene Ostracoda of New Jersey, Maryland, and Virginia. Journal of Paleontology, 27(6):761-799, 14 figures, plates 78- 81, 2 tables. Mansfield, W.C. 1928. Notes on Pleistocene Fauna from Maryland and NUMBER 53 121 Virginia, and Pliocene and Pleistocene Faunas from North Carolina. U.S. Geological Survey Profes- sional Paper, 150-F: 129-140. 1929. The Chesapeake Miocene Basin of Sedimentation as Expressed in the New Geologic Map of Vir- ginia. Washington Academy of Science Journal, 19(13) :263-268. 1936. Additional Notes on the Molluscan Fauna ofthe Pliocene Croatan Sand of North CaroVina. Journal of Paleontology, 10(7):665-668. 1943. Stratigraphy of the Miocene of Virginia and the Miocene and Pliocene of North Carolina. In Julia Gardner, Mollusca from the Miocene and Lower Pliocene of Virginia and North Carolina, Part 1: Pelecypoda. U. S. Geological Survey Professional Paper, 199-A: 1-19. McLean, J.D., Jr. 1957. The Ostracoda ofthe Yorktown Formation in the York-James Peninsula of Virginia. Bulletins of American Paleontology, 38(167):57-103, 12 plates. 1966. Miocene and Pleistocene Foraminifera and Ostra- coda of Southeastern Virginia. Virginia Division of Mineral Resources Report of Investigations, 9: 123 pages, 7 figures, 5 tables. Morales, G.A. 1966. Ecology, Distribution, and Taxonomy of Recent Ostracoda ofthe Laguna de Terminos, Campeche, Mexico. Universidad Nacional Autonoma de Mexico, Instituto de Geologia, Boletin, 81: 103 pages, 8 plates, 46 figures. Morkhoven, F.P.C.M. van 1963. Post-Paleozoic Ostracoda, Volume II: Generic Descrip- tions. 478 pages. Amsterdam: Elsevier Publishing Company. Murray, G.E. 1961. Geology ofthe Atlantic and Gulf Coastal Province of North America. 692 pages. New York: Harper and Bros. Olsson, A.A. 1964. The Geology and Stratigraphy of South Florida. In A.A. Olsson and R.E. Petit, Some Neogene Mollusca from Florida and the Carolinas. Bulletins of American Paleontology, 47(217):509-526. Parker, F.L. 1973. Late Cenozoic Biostratigraphy (Planktonic Fora- minifera) of Tropical Atlantic Deep-Sea Sections. Revista Espanola de Micropaleontologia, 5:253-289, 3 plates, 4 figures, 10 tables. Plusquellec, P.L., and P.A. Sandberg 1969. Some Genera of the Ostracode Subfamily Cam- pylocytherinae. Micropaleontology, 15:427-480. Poag, C.W. 1972. New Ostracod Species from the Chickasawhay Formation (Oligocene) of Alabama and Missis- sippi. Revista Espafiola de Micropaleontologia, 4(1) :65- 96. 1974. Late Oligocene Ostracodes from the U.S. Gulf Coastal Plain. Revista Espanola de Micropaleontologia, 6(l):39-74. Pokorny, V. 1970. The Genus Caudites Coryell and Fields, 1937 (Os- tracoda, Crust.) in the Galapagos Islands. Acta Universitatis Carolinae, Geologica, 4:267-302. Poore, R.Z. 1979. Oligocene through Quarternary Planktonic Fora- miniferal Biostratigraphy of the North Atlantic (DSDP Leg 49). Initial Reports of the Deep Sea Drilling Project, 49:447-517. Pooser, W.K. 1965. Biostratigraphy of Cenozoic Ostracoda from South Carolina. University of Kansas Paleontological Contribution, Arthropoda, 8:80 pages. Puri, H.S. 1953a. The Ostracode Genus Trachyleberis and its Ally Actinocythereis. American Midland Naturalist, 49:171- 187, 2 plates, 7 figures, 2 tables. 1953b. Contribution to the Study of the Miocene of the Florida Panhandle, Part III: Ostracoda. Florida Geological Survey Bulletin, 36(1954):215-345. 17 plates, 14 figures, 12 tables. 1960. Recent Ostracoda from the West Coast of Florida. Transactions of the Gulf Coast Association of Geological Societies, 10:107-149, plates 1-6. Puri, H.S. and Hulings, N.C. 1957. Recent Ostracode Facies from Panama City to Florida Bay Area. Transactions of the Gulf Coast Association of Geological Societies, 7:167-190. Puri, H.S., and V.V. Vanstrum 1971. Stratigraphy and Paleoecology ofthe Late Ceno- zoic Sediments of South Florida. In H.J. Oertli, editor, Paleoecologie des Ostracodes. Centre de Re- cherches Pau-SNPA Bulletin, 5(suppl.):433-448, 4 figures. Puri, H.S., and R.O. Vernon 1964. Summary of the Geology of Florida and a Guide- book to the Classic Exposures. Florida Geological Survey Special Publication 5, 312 pages. Rodriguez, L.R. 1969. Pliocene Ostracodes from the Playa Grande For- mation, North Central Venezuela, South Amer- ica. Boletin Informativo, 12(6): 155-212. Sandberg, P. 1964a. Notes on Some Tertiary and Recent Brackish- Water Ostracoda. Pubblicazioni della Stazione Zoolo- gica di Napoli, 33(suppl.):496-514, 3 plates, 1 fig- ure. 1964b. Larva-Adult Relationships in Some Species ofthe Ostracode Genus Haplocytheridea. Micropaleontology, 10:357-368, 2 plates, 2 figures. 1970. Scanning Electron Microscopy of Freeze-dried Os- tracoda (Crustacea). Transactions of the American Microscopical Society, 89(1): 113-124, 14 figures. 122 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Stephenson, M.B. 1938. Miocene and Pliocene Ostracoda of the Genus Cytheridea from Florida. Journal of Paleontology, 12(2): 127-148, plates 23-24, 20 figures. 1944. New Ostracoda from Subsurface Middle Tertiary Strata of Texas, yourna/ of Paleontology, 18(2): 156- 161, plate 28. Swain, F.M. 1951. Ostracoda from Wells in North Carolina, Part 1: Cenozoic Ostracoda. U.S. Geological Survey Profes- sional Paper, 234-A:l-58, 7 plates, 3 figures. 1955, Ostracoda of San Antonio Bay, Texas. Journal of Paleontology, 29(4):561-646, plates 59-64, 1968, Ostracoda from the Upper Tertiary Waccamaw Formation of North Carolina and South Carolina. U.S. Geological Survey Professional Paper, 573-D: 1-37. 7 plates, 30 figures, 2 tables. 1974. Some Upper Miocene and Pliocene (?) Ostracoda of Atlantic Coastal Region for Use in Hydrogeo- logic Studies. U.S. Geological Survey Professional Pa- per, 821:1-50. 13 plates, 1 figure, 1 table. Teeter, J.W. 1975. Distribution of Holocene Marine Ostracoda from Belize. American Association of Petroleum Geologists, Studies in Geology, 2:400-499, 23 figures, 3 tables. Tressler, W.L., and E.M. Smith 1948. An Ecological Study of Seasonal Distribution of Ostracoda, Solomons Island, Maryland, Region. Chesapeake Biological Laboratory Publication, 71: 57 pages, 4 plates. Ulrich, E.O., and R.S. Bassler 1904. Systematic Paleontology, Miocene (Ostracoda). Maryland Geological Survey, 2:98-130, plates 35-38. Valentine, P.C. 1971. Climatic Implications of a Late Pleistocene Ostra- code Assemblage from Southeastern Virginia. U.S. Geological Survey Professional Paper, 683-D: 1-28, 4 plates, 11 figures, 2 tables. van den Bold, W.A. 1965. Middle Tertiary Ostracoda from Northwestern Puerto Rico. Micropaleontology, 11:381-414. 1967. Ostracoda ofthe Gatun Formation, Panama. Mi- cropaleontology, 13:306-318. Waller, T.R. 1969. The Evolution of the Argopecten gibbus Stock (Mol- luca: Bivalvia), with Emphasis on the Tertiary and Quaternary Species of Eastern North Amer- ica. Paleontological Society Memoir, 3 (Journal of Paleontology, 43(5, supplement): 1-125. Ward, L.W., and B.W. Blackwelder 1975. Chesapecten, a New Genus of Pectinidae (Mollusca: Bivalvia) from the Miocene and Pliocene of East- ern North America. U.S. Geological Survey Profes- sional Paper, 861:1-24. 1980. Stratigraphic Revision of Upper Miocene and Lower Pliocene Beds of the Chesapeake Group, Middle Atlantic Coastal Plain. U.S. Geological Sur- vey Bulletin, 1482-D: 1-61. Welby, C.W., and C.J. Leith 1969. Miocene Unconformity, Pamlico River Area, North Carolina. Geological Society of America Bulletin, 80:1149-1154, Williams, R,B, 1966, Recent Marine Podocopid Ostracoda of Narra- gansett Bay, Rhode Island. University of Kansas Paleontological Contributions, 11:1-36, 27 figures. Plates 1-38 124 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 1 Cytheridea aff. C. virginiensis (Malkin, 1953) 1. Exterior view female left valve, USNM 172518, Yorktown Formation, sample 6 of this study, USGS 24883, X 81. 2. Interior view female left valve, USNM 313688, Yorktown Formation, James City County, Virginia, sample 42 of Hazel (1971a), USGS 24912, X 82. Cytheridea virginiensis (Malkin, 1953) 3. Exterior view female left valve, USNM 172519, Yorktown Formation, Greenville County, Virginia, sample 27 of Hazel (1971a), USGS 24830, X 93. 4. Interior view female left valve, USNM 172511, Yorktown Formation, James City County, Virginia, sample VA-7 of Malkin (1953, pi. 79: fig. 30), X 93. 3. NUMBER 53 125 126 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 2 Cytheridea campwallacensis, new species 1, Exterior view male left valve, USNM 172506, Yorktown Formation, James City County, Virginia, sample 42 of Hazel (1971a), USGS 24912, X 75. 3. Interior view female left valve, USNM 191351, Yorktown Formation, Greenville County, Virginia, sample 27 of Hazel (1971a), USGS 24830, X 90. 4. Exterior view female right valve, holotype, USNM 172619, Yorktown Formation, James City County, Virginia, sample VA-7 of Malkin (1953, pi. 79: fig. 29), X 83. Cytheridea virginiensis (Malkin, 1953) 2. Interior view female right valve, small form, USNM 191358, Croatan Formation, sample 15 of this study, USGS 25378, X 98. NUMBER 53 127 128 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 3 Cytheridea carolinensis, new species 1. Exterior view female left valve, USNM 191357, Croatan Formation, sample 15 of this study, USGS 25378, x 106. 2. Interior view female left valve, USNM 191337, Croatan Formation, sample 13 of this study, USGS 25377, X 114. 3. Exterior view female right valve, USNM 191359, "Yorktown" Formation near Colerain, North Carolina, sample 15 of Hazel (1971a), USGS 24892, X 106. 4. Exterior view male left valve, holotype, USNM 172509, Croatan Formation, sample 12 of this study, USGS 24886, X 105. 1. NUMBER 53 129 130 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 4 Peratocytheridea sandbergi, new species 1. Exterior view female left valve, holotype, USNM 172663, Yorktown Formation, sample 4 of this study, USGS 24881, X 87. 2. Interior view male right valve, USNM 172662, Yorktown Formation, sample 4 of this study, USGS 24881, X 87, 3. Exterior view male left valve, USNM 172661, Yorktown Formation, sample 6 of this study, USGS 24883, X 87. Peratocytheridea setipunctata (Brady, 1869) 4. Exterior view female right valve, USNM 191504, Croatan Formation, sample 15 of this study, USGS 25378, X 74. NUMBER 53 131 132 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 5 Neocaudites variabilus, new species 1. Exterior view male left valve, USNM 172745, Norfolk Formation (upper Pleistocene) at Virginia Beach, Virginia, locality P2-1 of Valentine (1971), USGS 23787, X 98. 2. Exterior view female right valve, USNM 191387, Yorktown Formation, Hampton City, Virginia, sample 34 of Hazel (1971a), USGS 24810, X 110. 3. Interior view female right valve, USNM 191386, type locality ofthe Duplin Formation near Magnolia, North Carolina, USGS 23639, X 120. Neocaudites subimpressus (Edwards, 1944) 4. Exterior view female left valve, USNM 559432, Duplin Formation, near Lumberton, North Carolina, locality 3 of Edwards (1944, pi. 87: fig. 30), X 109. NUMBER 53 13: 134 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 6 Neocaudites triplistriatus (Edwards, 1944) 1. Left exterior view female carapace, USNM 172753, type locality ofthe Duplin Formation, near Magnolia, North Carolina, locality 1 of Edwards (1944), USGS 23639, X 137. Neocaudites angulatus, new species 2. Interior view female right valve, USNM 191388, type locality of the Duplin Formation, near Magnolia, North Carolina, locality 1 of Edwards (1944), USGS 23639, X 110. 3. Exterior view male? right valve, USNM 172741, Croatan Formation, near James City, North Carolina, locality 80 of Waller (1969), USGS 25443, X 102. 4. Left exterior view female carapace, holotype, USNM 172742, Waccamaw Formation, near Old Dock, North Carolina, locality NC-4 of Swain (1968), USGS 25445, X 110. 2. NUMBER 53 135 136 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 7 Neocaudites vanabilus, new species 1. Exterior view female right valve, holotype, USNM 190494, type locality of the Duplin Formation, near Magnolia, North Carolina, locality 1 of Edwards (1944), USGS 23639, X 87. Pterygocythereis alophia, new species 2. Exterior view male left valve, USNM 191496, "Yorktown" Formation, Colerain Landing, North Carolina, sample 14 of Hazel (1971a), USGS 24891, X 81. 4. Exterior view female right valve, holotype, USNM 172642, Holocene sample 1861 from Raleigh Bay, south of Cape Hatteras, 34°45.6' N lat., 75°44.6' W long., 41 meters, X 78. Pterygocythereis inexpectata (Blake, 1929) 3. Exterior view female left valve, USNM 172748, Yorktown Formation, sample 6 of this study, USGS 24883, X 66. NUMBER 53 137 138 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 8 Actinocythereis captionis, new species 1. Left exterior view female carapace, holotype, USNM 172461, Holocene sample 2251, southwest of Cape Fear, 33°42.7' N lat,, 78°45,0' W long., 12 meters, X 85. 2. Left exterior view male carapace, USNM 172463, Holocene sample 2251, southwest of Cape Fear, 33°42.7' N lat., 78°45.0' W long., 12 meters, X 90. 4. Interior view male right valve, USNM 191348, "Yorktown'' Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 90. Actinocythereis marylandica (Howe and Hough, 1935) 3. Exterior view female left valve, USNM 190459, Yorktown Formation, near Palmyra, North Carolina, sample 11 of Hazel (1971a), USGS 24889, X 87. NUMBER 53 139 140 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 9 Actinocythereis dawsoni (Brady, 1870) 1. Exterior view female left valve, USNM 190458, Yorktown Formation, James City County, Virginia, sample 42 of Hazel (1971a), USGS 24912, X 87, Actinocythereis mundorffi (Swain, 1951) 2. Exterior view female left valve, small form, USNM 190457, Yorktown Formation, James City County, Virginia, sample 42 of Hazel (1971a), USGS 24912, X 85. Cytheropteron?yorktownensis (Malkin, 1953) 3. Exterior view female left valve, USNM 190980, Yorktown Formation at Suffolk, Virginia, sample 29 of Hazel (1971a), X 180. Cytheropteron talquinensis (Puri, 1954) 4. Exterior view female left valve, USNM 190981, Yorktown Formation, Williamsburg, Vir- ginia, sample 44 of Hazel (1971a), USGS 24820, X 137. NUMBER 53 141 142 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 10 Murrayina macleani Swain, 1974 1. Exterior view female right valve, ornate form, USNM 172630, Yorktown Formation, sample 4 of this study, USGS 24881, X 95. 2. Exterior view female right valve, smooth form, USNM 172630, Yorktown Formation, Suffolk, Virginia, sample 28 of Hazel (1971a), USGS 24811, X 95. 3. Exterior view male left valve, ornate form, USNM 191395, Yorktown Formation, Suffolk, Virginia, sample 28 of Hazel (1971a), USGS 24811, X 90. 4. Interior view male right valve, USNM 191394, Yorktown Formation, near Skippers, Virginia, sample 27 of Hazel (1971a), USGS 24830, X 82. 1. NUMBER 53 143 144 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 11 Hermanites ascitus, new species 1. Exterior view female left valve, holotype, USNM 172745, Duplin Formation, left bank of Lumber River near Lumberton, North Carolina, USGS 25751, X 122. 2. Interior view female right valve, USNM 191016, Duplin Formation, Robeson Farm near Tar Heel, North Carolina, USGS 25755, X 122. 3. Exterior view male right valve, USNM 191333, Yorktown Formation, sample 9 of this study, USGS 25358, X 127. Murrayina barclayi McLean, 1957 4. Exterior view female left valve, USNM 172755, Yorktown Formation at Petersburg, Virginia, sample 38 of Hazel (1971a), USGS 24908, X 88. UMBER 53 145 146 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 12 Orionina vaughani (Ulrich and Bassler, 1904) 1. Exterior view male left valve, USNM 172477, Yorktown Formation at Williamsburg, Virginia, sample 43 of Hazel (1971a), USGS 24821, X 80. Caudites paraasymmetricus, new species 2. Exterior view female right valve, USNM 172757, Croatan Formation, sample 15 of this study, USGS 25378, X 112. 3. Interior view male right valve, USNM 172740, Croatan Formation, sample 15 of this study, USGS 25378, X 112. 4. Left exterior view female carapace, holotype, USNM, 172756, Croatan Formation, sample 15 of this study, USGS 25378, X 113. 2. NUMBER 53 147 148 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 13 Palaciosa minuta (Edwards, 1944) Exterior view male left valve, USNM 167401, Croatan Formation, right bank of Neuse River near James City, North Carolina, USGS 25444, X 115. Interior view female right valve, USNM 167403, Croatan Formation, locality 80 of Waller (1969), USGS 25443, X 144. Interior view right valve of disarticulated holotype of Edwards (1944, pi. 87: figs. 1-3), USNM 559425, Duplin Formation near Lumberton, North Carolina. Note that the small pillars in the anterocentral part of the valve in 3 have become centers of calcite deposition in 4, X 144. Radimella confragosa (Edwards, 1944) 2. Left exterior view female carapace, USNM 172675, sample 12 of this study, USGS 24886, X 110. NUMBER 53 149 150 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 14 Malzella conradi (Howe and McGuirt, 1935) 1. Exterior view female left valve, angulate form, USNM 167407, Yorktown Formation at Williamsburg, Virginia, sample 44 of Hazel (1971a), USGS 24820, X 101. 2. Left exterior view female carapace, subquadrate form, USNM 190468, Red Bay Formation of Puri and Vernon (1964), {Area zone), upper Miocene, 1 mile (1,61 km) southeast of Red Bay, Fla„ USGS 24709, X 85, 4, Left exterior view female carapace, intermediate form, USNM 190692, Jackson Bluff Formation {Ecphora zone), Pliocene, Leon County, Florida, USGS 25158, X 93. Malzella evexa, new species 3. Exterior view female left valve, holotype, USNM 172653, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 95, NUMBER 53 151 152 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 15 Malzella evexa, new species 1. Exterior view male left valve, USNM 313689, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 107. 2. Interior view female right valve showing anterior hinge elements, USNM 191340, Duplin Formation, Cedar Bluff Landing on the Savannah River, Georgia, USGS 23863, X 236. 3. Interior view female right valve, same specimen as figure 2 showing posterior hinge elements, X 236. 5. Exterior view female left valve, Edwards' (1944, pi. 86: fig. 18) specimen, USNM 559759, Duplin Formation, near Magnolia, North Carolina, locality 1 of Edwards (1944), X 122. Aurila laevicula (Edwards, 1944) 4. Exterior view female left valve, USNM 172491, Yorktown Formation at Petersburg, Virginia, sample 39 of Hazel (1971a), USGS 24908, X 103. NUMBER 53 153 154 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 16 Muellerina bassiounii, new species 1. Exterior view male left valve, USNM 172618, "Yorktown" Formation near Mt. Gould, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 115. 4. Exterior view female left valve, holotype, USNM 167410, Croatan Formation at its type locality (MacNeil, 1938; DuBar and Solliday, 1963), near James City, North Carolina, X 103. Muellerina canadensis petersburgensis, new subspecies 2. Exterior view female left valve, holotype, USNM 167408, Yorktown Formation at Peters- burg, Virginia, sample 38 of Hazel (1971a), USGS 24908, X 103. Muellerina ohmerti, new species 3. Exterior view female left valve, holotype, USNM 112741, Holocene, Atlantic shelf south of Long Island, 40°20' N lat., 73° 15' W long., 36 meters, sample 1287 of Hazel (1967, 1970), and Valentine (1971), X 95. NUMBER 53 155 156 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 17 Muellerina blowi, new species 1. Exterior view female left valve, holotype, USNM 167411, Yorktown Formation at Suffolk, Virginia, sample 29 of Hazel (1971a), USGS 24814, X 92. 3. Exterior view female right valve, USNM 172621, Yorktown Formation at Suffolk, Virginia, sample 29 of Hazel (1971a), USGS 24814, X 98. Muellerina wardi, new species 2, Exterior view female right valve, USNM 172489, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 90. 4. Exterior view female left valve, holotype, USNM 167409, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 88. NUMBER 53 157 158 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 18 Muellerina canadensis petersburgensis, new subspecies 1. Interior view male right valve, USNM 172625, Yorktown Formation at Petersburg, Virginia, sample 38 of Hazel (1971a), USGS 24908, X 89. 3. Interior view female right valve, USNM 172626, Yorktown Formation at Halifax, North Carolina, sample 24 of Hazel (1971a), USGS 24904, X 91. Muellerina blowi, new species 2. Interior view male right valve, USNM 172720, Yorktown Formation at Suffolk, Virginia, sample 29 of Hazel (1971a), USGS 24814, X 102. Muellerina wardi, new species 4, Exterior view male left valve, USNM 172632, Yorktown Formation on the Piankatank River, Middlesex County, Virginia, sample 46 of Hazel (1971a), USGS 24801, X 104. 5. Interior view male right valve, USNM 172634, Yorktown Formation in Petersburg, Virginia, sample 39 of Hazel (1971a), USGS 24909, X 116. Muellerina bassiounii, new species 6. Interior view female right valve, USNM 191402, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 111. NUMBER 53 159 160 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 19 Thaerocythere carolinensis, new species 1. Exterior view male left valve, USNM 172658, "Yorktown'' Formation, near Mt. Gould Landing, North Carolina, sample 16 of Hazel (1971a), USGS 24893, X 108. 3. Interior view female right valve, USNM 191415, "Yorktown" Formation at Colerain Landing, North Carolina, sample 14 of Hazel (1971a), USGS 24891, X 97. 4. Left exterior view female carapace, holotype, USNM 172657, "Yorktown" Formation at Colerain Landing, North Carolina, sample 14 of Hazel (1971a), USGS 24891, X 90. Thaerocythere schmidtae (Malkin, 1953) 2. Exterior view male left valve, USNM 172657, "Yorktown" Formation at Colerain Landing, North Carolina, sample 14 of Hazel (1971a), USGS 24891, X 90. NUMBER 53 161 162 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 20 Hirschmannia? hespera, new species 1. Exterior view female left valve, holotype, USNM 172575, sample 14 of this study, USGS 24887, X 147. 2. Interior view female right valve, USNM 172576, sample 14 of this study, USGS 24887, X 147. Hirschmannia? quadrata, new species 3. Exterior view female left valve, holotype, USNM 172571, "Yorktown" Formation near Yadkin, Virginia, sample 31 of Hazel (1971a), USGS 24905, X 149. 4. Interior view male right valve, USNM 172573, "Yorktown" Formation near Yadkin, Virginia, sample 31 of Hazel (1971a), USGS 24905, X 166. 3. NUMBER 53 163 164 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 21 Cytheromorpha incisa, new species 1. Exterior view female left valve, holotype, USNM 172560, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 18 of Hazel (1971a), USGS 24895, X 136. 2. Interior view male right valve, USNM 172559, "Yorktown" Formation, Colerain Landing, North Carolina, sample 15 of Hazel (1971a), USGS 24892, X 136. Hirschmannia? hespera, new species 3. Exterior view male left valve, USNM 172574, Croatan Formation, sample 14 of this study, USGS 24887, X 168. 4. Interior view female right valve, USNM 172577, Croatan Formation, sample 14 of this study, USGS 24887, X 154. 3. NUMBER 53 165 166 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 22 Cytheromorpha macroincisa, new species 1. Exterior view female left valve, holotype, USNM 172562, "Yorktown" Formation, near Mt, Gould Landing, North Carolina, sample 18 of Hazel (1971a), USGS 24895, X 116. 2. Interior view female right valve, USNM 172561, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 18 of Hazel (1971a), USGS 24895, X 118. 3. Interior view female right valve, showing central muscle field, same specimen as figure 2, X 356. 4. Interior view female right valve, showing posterior hinge elements, same specimen as figure 2, X 390. 5. Interior view female right valve, showing anterior hinge elements, same specimen as figure 2, X 390. Cytheromorpha warneri Howe and Spurgeon, 1935 6. Exterior view female left valve, USNM 172555, Jackson Bluff Formation {Ecphora zone), Leon County, Florida, USGS 25158, X 105. NUMBER 53 167 168 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 23 Cytheromorpha suffolkensis, new species 1. Exterior view female left valve, holotype, USNM 172551, Yorktown Formation, Suffolk, Virginia, sample 29 of Hazel (1971a), USGS 24814, X 142. 2. Interior view male right valve, USNM 172552, Yorktown Formation, Suffolk, Virginia, sample 29 of Hazel (1971a), USGS 24814, X 142. 3. Interior view male right valve, showing anterior hinge elements, same specimen as figure 2, X 600. 4. Interior view male right valve, showing posterior hinge elements, same specimen as figure 2, X 600. Cytheromorpha incisa, new species 5. Interior view male right valve, showing anterior hinge elements, USNM 172559, "Yorktown" Formation, Colerain Landing, North Carolina, sample 15 of Hazel (1971a), USGS 24892, X 544. 6. Interior view male right valve, showing posterior hinge elements, same specimen as figure 5, X 544. 5. NUMBER 53 169 170 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 24 Pseudocytheretta burnsi (Ulrich and Bassler, 1904) Exterior view male left valve, smooth form, USNM 172638, Yorktown Formation, Hampton City, Virginia, sample 36 of Hazel (1971a), USGS 24907, X 70. Right exterior view female carapace, pitted form, USNM 172637, Yorktown Formation, Hampton City, Virginia, sample 36 of Hazel (1971a), USGS 24907, X 75. Loxoconcha edentonensis Swain, 1951 Exterior view female left valve, USNM 172603, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 19 of Hazel (1971a), USGS 24896, X 120. Exterior view male left valve, USNM 172604, Croatan Formation, sample 12 of this study, USGS 24886, X 120. NUMBER 53 171 172 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 25 Proteoconcha jamesensis, new species 1. Exterior view female left valve, holotype, USNM 167377, Yorktown Formation, Nansemond County, Virginia, sample 32 of Hazel (1971a), USGS 24622, X 90. 2. Interior view female left valve, USNM 191438, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 17 of Hazel (1971a), USGS 24894, X 108. Puriana rugipunctata (Ulrich and Bassler, 1904) 3. Exterior view female left valve, USNM 172669, Yorktown Formation at Petersburg, Virginia, sample 38 of Hazel (1971a), USGS 24908, X 110. Puriana mesacostalis (Edwards, 1944) 4. Exterior view female left valve, USNM 172668, "Yorktown'' Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 112. NUMBER 53 173 174 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 26 Puriana convoluta Teeter, 1975 1. Exterior view male left valve, USNM 191499, Waccamaw Formation, Old Dock, North Carolina, USGS 25445, X 130. 2. Exterior view female left valve, USNM 172651, Holocene sample 1861, Raleigh Bay, south of Cape Hatteras, 34°45.6' N lat., 75°44,6' W long., 41 meters, X 115, 4. Interior view female right valve, USNM 172652, Holocene sample 1861, Raleigh Bay, south of Cape Hatteras, 34°45.6' N lat., 75°44.6' W long., 41 meters, X 113. Puriana rugipunctata (Ulrich and Bassler, 1904) 3. Exterior view female right valve, USNM 172478, Yorktown Formation, Williamsburg, Virginia, sample 44 of Hazel (1971a), USGS 24820, X 135. NUMBER 53 175 176 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 27 Puriana carolinensis, new species Exterior view female right valve, holotype, USNM 172649, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 109. Interior view female left valve, USNM 172647, type locality ofthe Duplin Formation, near Magnolia, North Carolina, USGS 23639, X 150. Exterior view male left valve, USNM 172648, type locality ofthe Duplin Formation, near Magnolia, North Carolina, USGS 23639, X 150. Proteoconcha jamesensis, new species 2. Exterior view male left valve, USNM 167379, Yorktown Formation, Hampton City, Virginia, sample 36 of Hazel (1971a), USGS 24907, X 117, NUMBER 53 177 178 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 28 Paracytheridea cronini, new species 1. Exterior view male left valve, USNM 191428, type locality ofthe Duplin Formation, near Magnolia, North Carolina, USGS 23639, X 115. 2. Interior view female right valve, USNM 191412, type locality of the Duplin Formation, near Magnolia, North Carolina, USGS 23639, X 118. Paracytheridea rugosa Edwards, 1944 3. Exterior view female left valve, USNM 191497, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 122. Paracytheridea altila Edwards, 1944 4. Exterior view female left valve, USNM 191500, Croatan Formation, sample 13 of this study, USGS 25377, X 122. NUMBER 53 179 180 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 29 Paracytheridea cronini, new species 1. Exterior view female left valve, holotype, USNM 172759, type locality of the Duplin Formation, near Magnolia, North Carolina, USGS 23639, X 113. Paracytheridea mucra Edwards, 1944 2. Exterior view female left valve, USNM 172758, Yorktown Formation, Suffolk, Virginia, sample 29 of Hazel (1971a), USGS 24814, X 102. Microcytherura choctawhatcheensis (Puri, 1954) 3, Exterior view male left valve, USNM 191498, Waccamaw Formation, pit near Old Dock, North Carolina, USGS 25445, X 130. Microcytherura similis (Malkin, 1953) 4, Exterior view female left valve, USNM 172751, Yorktown Formation, Nansemond County, Virginia, sample 32 of Hazel (1971a), USGS 24622, X 117. NUMBER 53 181 182 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 30 Microcytherura expanda, new species 1. Exterior view male left valve, USNM 191481, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 123. 2. Exterior view female left valve, holotype, USNM 172752, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 117. 3. Interior view female right valve, USNM 191482, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 20 of Hazel (1971a), USGS 24897, X 117. Microcytherura similis (Malkin, 1953) 4. Exterior view male left valve, USNM 191483, Yorktown Formation, Nansemond County, Virginia, sample 32 of Hazel (1971a), USGS 24622, X 127. NUMBER 53 183 184 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 31 Microcytherura minuta, new species 1. Interior view female right valve, USNM 191484, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 18 of Hazel (1971a), USGS 24895, X 135. 2. Exterior view male left valve, USNM 313690, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 18 of Hazel (1971a), USGS 24895, X 138. 3. Exterior view female left valve, holotype, USNM 172750, Yorktown Formation, Williams- burg, Virginia, sample 45 of Hazel (1971a), X 150, Microcytherura similis (Malkin, 1953) 4. Interior view female right valve, USNM 313691, Yorktown Formation, Nansemond County, Virginia, sample 32 of Hazel (1971a), USGS 24622, X 154, NUMBER 53 185 186 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 32 Bensonocythere trapezoidalis (Swain, 1974) 1, Exterior view male left valve, USNM 167395, Yorktown Formation, near Palmyra, North Carolina, sample 11 of Hazel (1971a), USGS 24889, X 90, 2. Interior view female right valve, USNM 167396, Yorktown Formation, near Palmyra, North Carolina, sample 11 of Hazel (1971a), USGS 24889, X 113. Bensonocythere rugosa, new species 3, Interior view female right valve, USNM 167383, Yorktown Formation at Petersburg, Virginia, sample 38 of Hazel (1971a), USGS 24908, X 90. 4. Exterior view female left valve, holotype, USNM 167382, Yorktown Formation at Peters- burg, Virginia, sample 39 of Hazel (1971a), USGS 24909, X 86. NUMBER 53 187 188 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 33 Bensonocythere ricespitensis, new species 1. Exterior view female left valve, holotype USNM 167394, Yorktown Formation, Hampton City, Virginia, sample 36 of Hazel (1971a), USGS 24907, X 100. 2. Interior view female right valve, USNM 167391, Yorktown Formation, Hampton City, Virginia, sample 36 of Hazel (1971a), USGS 24907, X 98. 3. Exterior view female right valve, same specimen as figure 2, X 98. 4. Interior view male right valve, USNM 191383, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 18 of Hazel (1971a), USGS 24895, X 128. 1. NUMBER 53 189 190 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 34 Bensonocythere bradyi, new species 1. Exterior view male right valve, holotype, USNM 167380, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 17 of Hazel (1971a), USGS 24894, X 78. 2. Interior view female right valve, USNM 167381, Yorktown Formation, Hampton City, Virginia, sample 37 of Hazel (1971a), USGS 24805, X 75. Bensonocythere gouldensis, new species 3. Exterior view male left valve, USNM 167389, "Yorktown'' Formation, near Mt. Gould Landing, North Carolina, sample 18 of Hazel (1971a), USGS 24895, X 90, 4. Interior view female right valve, USNM 167388, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 19 of Hazel (1971a), USGS 24896, X 95. Bensonocythere calverti (Ulrich and Bassler, 1904) 5. Exterior view male left valve, USNM 167387, Yorktown Formation, near Palmyra, North Carolina, sample 11 of Hazel (1971a), USGS 24889, X 68. NUMBER 53 191 192 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 35 Bensonocythere blackwelden, new species 1. Exterior view male left valve, USNM 191505, Yorktown Formation, Hampton City, Virginia, sample 37 of Hazel (1971a), USGS 24805, X 110, 2. Interior view male left valve, USNM 172481, Yorktown Formation, Hampton City, Virginia, sample 37 of Hazel (1971a), USGS 24805, X 110. 4. Exterior view female left valve, holotype, USNM 167398, Yorktown Formation, Isle of Wight County, Virginia, sample 33 of Hazel (1971a), USGS 24823, X 90. Bensonocythere whitei (Swain, 1951) 3. Exterior view female left valve of Swain's holotype, USNM 560640, "Yorktown" Formation, Edenton Naval Air Base, North Carolina, 55 foot (16,8 m) depth in well (downdip stratigraphic equivalent of localities 3 and 4 of Hazel, 1971a), X 109, NUMBER 53 193 194 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 36 Echinocythereis leecreekensis, new species 1, Exterior view female right valve, holotype, USNM 191495, Croatan Formation, sample 10 of this study, USGS 25359, X 70, 2, Exterior view female left valve, USNM 191510, Croatan Formation, sample 10 of this study, USGS 25359, x 70. 3, Interior view female right valve, USNM 191508, Croatan Formation, sample 11 of this study, USGS 25376, X 77, Echinocythereis planibasalis (Ulrich and Bassler, 1904) 4, Exterior view male left valve, USNM 172754, Yorktown Formation near Jamestown, Virginia, sample 41 of Hazel (1971a), USGS 24717, X 70. NUMBER 53 195 196 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 37 Bensonocythere rugosa, new species 1. Exterior view male right valve, USNM 167384, Yorktown Formation, Hampton City, Virginia, sample 36 of Hazel (1971a), USGS 24907, X 118. Bensonocythere gouldensis, new species 2. Exterior view female left valve, holotype, USNM 167385, "Yorktown" Formation, near Mt. Gould Landing, North Carolina, sample 19 of Hazel (1971a), USGS 24896, X 75. 3. Exterior view female left valve, USNM 167390, Yorktown Formation, Hampton City, Virginia, sample 37 of Hazel (1971a), USGS 24805, X 82. Bensonocythere blackwelderi, new species 4. Exterior view female right valve, USNM 167400, Yorktown Formation, Hampton City, Virginia, sample 36 of Hazel (1971a), USGS 24907, X 112. NUMBER 53 197 198 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 38 "Pontocythere" sp. I 1. Exterior view female left vale, USNM 172523, Yorktown Formation, Petersburg, Virginia, sample 38 of Hazel (1971a), USGS 24908, X 90. Bensonocythere bradyi, new species 2. Exterior view female left valve, USNM 191514, Yorktown Formation, Petersburg, Virginia, sample 38 of Hazel (1971a), USGS 24908, X 80. 4. Exterior view male left valve, USNM 167382, Yorktown Formation, Petersburg, Virginia, sample 38 of Hazel (1971a), USGS 24908, X 93. Echinocythereis leecreekensis, new species 3. Exterior view male right valve, USNM 191509, Croatan Formation, sample 11 of this study, USGS 25376, X 70. NUMBER 53 199 The Post-Yorktown Stratigraphy and Geomorphology of the Neuse-Pamhco Area, Eastern North Carohna Walter H. Wheeler, Raymond B. Daniels, and Erling E. Gamble ABSTRACT Regional study ofthe post-Yorktown history of the Neuse-Tar-Pamlico rivers area, based on lim- ited exposures and many auger holes, reveals the buried Aurora paleoscarp on the erosional surface of the Yorktown Formation, extending from the Pamlico River at the Lee Creek Mine southward to the coast near Morehead City. Overlying the Yorktown Formation where present and the Cas- tle Hayne Formation elsewhere in the area is a complex of pre-late Pleistocene deposits herein referred to the informal Small sequence or Cro- atan Formation (including the James City For- mation), which includes both marine and non- marine beds. The late Pleistocene surficial mate- rials overlying the Small sequence are referred to the Talbot and Pamlico morphostratigraphic units (msu). The Talbot msu includes the Flanner Beach and Neuse formations. The Pamlico msu includes most of the surface sediments east of the Suffolk Scarp, locally represented by the Minne- sott Ridge. Introduction The Lee Creek Mine of Texasgulf Inc. is situ- ated in the northern part of our regional study Walter H. Wheeler, Department of Geology, University of North Carolina, Chapel Hill, North Carolina 27514. Raymond B. Dan- iels, 9112 Leesville Road, Raleigh, North Carolina 27612. Erling E. Gamble, Department of Agriculture, Soil Conservation Service, MTSC, Federal Building, U.S. Courthouse, Room 393, Lincoln, Nebraska 68508. area of the Coastal Plain in the Neuse-Tar-Pam- lico rivers region of eastern North Carolina. The mine lies northwest of Aurora in Beaufort County, North Carolina. Most of the data are from auger holes that give the relation between the regional geology and the section exposed at the Lee Creek Mine. The stratigraphic column in the Lee Creek Mine is, in ascending order: Pungo River For- mation, middle Miocene; Yorktown Formation, upper Miocene and/or lower Pliocene; Small se- quence (which may prove to be an enlarged concept of the Croatan Formation or the James City Formation), upper Pliocene or lower Pleis- tocene; and the Pamlico morphostratigraphic unit, upper Pleistocene. The Pungo River Formation is the lowest unit exposed in the mine but is not within the scope of this particular study. The Yorktown Formation in this region consists of fossiiiferous to barren marine sandy loam to silty clay. It lacks the organic zones of the over- lying Small sequence and is noted for the presence of Ecphora and other guide fossils. The top of the Yorktown Formation is an erosional surface that has very little relief in the areas between major streams. However, a gently sloping buried scarp, the Aurora paleoscarp, marks the top of the Yorktown Formation along a north-south trend, which happens to lie in the vicinity of the Lee Creek Mine. Immediately below the unconformity, the Yorktown Forma- 201 202 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY tion is generally enriched in calcium carbonate, which may form a crust difficult to penetrate with a power auger. This is not the case at the Lee Creek Mine, however. The Small sequence (Croatan or James City Formation) consists of sands or sandy loams, fre- quently very rich in marine shells alternating with organic layers. The organic layers may be as peats, as sandy loams with abundant wood chunks or fragments, or merely an organically stained sediment suggesting a paleosol. There is no pebble zone at the base ofthe Small sequence, but the change of lithology is distinct enough to allow identification of the boundary over a wide area. At the Lee Creek Mine the Small sequence consists of incline-bedded sands very rich in large and small marine shells that grade upward into a sparsely fossiiiferous sandy loam with or without organic staining. Our regional work clearly shows that the upper highly fossiiiferous layer in the Lee Creek Mine is not an upper part ofthe Yorktown Formation, but is a distinct and separate unit. The Pamlico morphostratigraphic unit (msu) includes all surface sediments east of the Suffolk Scarp (with the exclusion of some Recent mate- rial). The unit is typically fine textured in the upper 3 to 5 feet (0.9 to 1.5 m) and becomes sandier toward the base. Marine fossil shells are abundant, both as a basal hash layer and as few to common shells dispersed through a sandy ma- trix. The Pamlico never contains intercalated or- ganic layers; the only organic zones are at the top of the unit. At the Lee Creek Mine the Pamlico msu consists of near-shore marine and estuarine sands and silty sands notable for a profusion of burrows and other lebensspuren of marine orga- nisms. A prominent zone of pebbles, cobbles, and a few boulders mark the distinct unconformity at the base. The organic rich silty sands underlying the unconformity are at the upper part of the Small sequence or Croatan Formation. The Minnesott Ridge, a ridge of sand up to a mile wide, is associated with the inner edge ofthe Pamlico terrace and morphostratigraphic unit. The seaward face of the Minnesott Ridge forms part of the Suffolk Scarp. The ridge terminates at the Neuse River to the south and at a point about 4 miles (6.4 km) west of the Lee Creek Mine to the north. ACKNOWLEDGMENTS.—Much of the material presented in this report was modified from a portion ofthe guidebook for the joint 1972 annual meeting of the Carolina Geological Society and the Atlantic Coastal Plain Geological Association (Daniels et al., 1972). The authors have benefited from discussions and field trips with Dr. Edward S. Belt of Am- herst College, Dr. H. Allen Curran of Smith College, Dr. William A. White of the University of North Carolina at Chapel Hill, Dr. Clayton Ray ofthe National Museum of Natural History, Smithsonian Institution, Dr. Frank C. Whitmore, Jr., of the U.S. Geological Survey and Dr. C. Stephen Holzhey of the Soil Conservation Ser- vice, U.S. Department of Agriculture, Beltsville, Maryland. An earlier version of the manuscript was reviewed by Frank C. Whitmore, Jr., Druid Wilson, and the late Louis Ray. Portions of Wheeler's field expenses were de- frayed by grants of the Alumni Research Fund and some of the cost of illustrations was defrayed by the Smith Fund, both of the University of North Carolina at Chapel Hill. Stratigraphic Nomenclature The unconsolidated Neogene deposits of the Atlantic Coastal Plain are very similar to each other, and their interpretation is difficult enough for the uninitiated worker without turning him loose into an unfamiliar terminological jungle. Accordingly, we have chosen to circumvent some of these difficulties by assigning to the three stratigraphic units that we discuss names that do not imply more knowledge of their genesis than we have. The provisional and informal name. Small se- quence (Daniels et al., 1972), is applied to the older Pleistocene (or possibly Pliocene) unit im- mediately overlying the Yorktown Formation in NUMBER 53 203 the study area (Figure 1). The Small sequence is in part the marine James City Formation of DuBar and Solliday (1963) but includes nonma- rine beds. The name, "Croatan Formation," pro- posed by Dall (1892) and redefined by Mansfield (1928) may prove to be most useful (Fallaw and Wheeler, 1969). The surficial materials underlying the Talbot and the Pamlico surfaces but overlying the Small sequence are referred to as the Talbot and the Pamlico morphostratigraphic units respectively. The term "morphostratigraphic unit" is applied to surficial deposits in the Coastal Plain because, as Frye and Willman (1960:7-8) state in their original discussion, "they are identifiable by their form and not their lithology, which in many cases ... is not distinguishable from one . to the next, they are not normal rock-stratigraphic units." To quote further for clarification: "A morphostratigraphic unit is defined as comprising a body of rock that is identified primarily from the surface form it displays; it may or may not be distinctive lithologically from contiguous units; it may or may not transgress time through its ex- tent." Frye and Willman (1962) in a later discus- sion of this idea indicated that the term could be applied to fluvial terraces and other nonmarine stratigraphic and geomorphic elements. We would like to extend the term morphostrati- graphic unit (hereafter abbreviated to "msu") to include marine sediments and associated surfaces, much as Thom (1967:48) has done in South Carolina. The term "msu" is preferred to "terrace" or "formation" (cf. Johnson, 1906; Cooke, 1937), because a terrace is a surface and a formation is a mappable lithologic unit. Many ofthe Coastal Plain surficial sediments do not fit the definition of a formation, but the term "morphostratigraphic unit" can easily be applied to these elements without violating sound strati- graphic principles. The use of morphostrati- graphic unit does not preclude the application of formational names if they are defined under the rules of formal lithostratigraphic nomenclature. Since the body of this paper was written, our subsequent work has made us more inclined to regard the "Croatan Formation" as a valid and available name for the post-Yorktown but pre- late Pleistocene sediments of the Neuse-Pamlico area. Under this interpretation the provisional, informal name of "Small sequence" is a junior synonym of Croatan (Wheeler, Daniels, and Gamble, 1979). The Croatan Formation is distinguishable lith- ologically and stratigraphically in its type area along the south bank ofthe Neuse River between Cherry Point Marine Air Station and Flanner Beach. These outcrops lie approximately 31 miles (50 km) south of the Lee Creek Mine. Lithologic continuity is shown in auger samples from Lee Creek Mine to the north bank ofthe Neuse River. The Croatan Formation consists mainly of silty sands and sands that are very fossiiiferous at several localities. One of the most fossiiiferous of these Croatan localities is the Lee Creek Mine. In addition, there is generally a nonmarine organic zone or zones near the top ofthe formation. These organic zones range from a thin buried soil to layers rich in peaty fine sands or clays. Some of these zones contain large masses of wood, such as the cypress stumps at Flanner Beach. Dall (1892:209) named the Croatan Formation for two outcrops, one by the mouth of Slocum's Creek, on what is now the Cherry Point Marine Air Station, and the other at a beach about 2 miles (3.2 km) to the west. He noted that there was "ferruginous sandy clay, 10 to 12 feet [3 to 3.7 m]" at the top with "bluish clay with fossils (Pliocene), 5 to 6 feet [1.5 to 1.8 m]" under that. He differentiated the lower clay as "Pliocene" and noted its distinct lithology. Dall did not, however, explicitly place the top of the Croatan Formation at the top ofthe "blue clay." On the other hand, Mansfield (1928:135) noted that the fossil collections studied by Dall came from both Croatan and post-Croatan units at this locality. Mansfield proposed to clarify the matter by re- stricting the name "Croatan" to the lower or "Pliocene" part. He also noted an unconformity at the top of the Croatan Formation, as thus defined. This disconformity may be seen at Dall's western locality along the shoreline of the Neuse MILES 0 10 20 30 I I I I i_ I I I I 0 10 20 30 40 50 770 KILOMETERS FIGURE 1—Location of study area in eastern North Carolina. NUMBER 53 205 estuary at or just below the water line. Many fossil colonial corals can be seen there, whereas corals are rare in the overlying Flanner Beach Formation.The name "Croatan Formation" as defined (however poorly) by Dall and redefined clearly by Mansfield is therefore, a valid and available name, and the James City Formation of DuBar and Solliday (1963) must then be con- sidered a junior synonym, if applied to the entire Croatan Formation. Snyder and Hine (pers. comm.) regard the James City as a member of the Croatan Formation. Mixon and Pilkey (1976:9) regard the Croatan, as we have used it, as not a valid geologic unit, in either lithostratigraphic or biostratigraphic sense, inasmuch as it lumps dissimilar stra- tigraphic units such as the marine James City Formation of DuBar and Solliday (1963) (Pliocene?), the cypress stump bed at Flanner Beach and other swamp deposits (Pleisto- cene), and fossiiiferous marine beds in the area of the Tex- asgulf Sulphur phosphate mine which appear to have been assigned to the Yorktown Formation (Miocene and Pliocene) by other workers. The generalized stratigraphic sequence in the study area is shown in Figure 2. Fine sand on Minnesott Ridge; Pamlico Fm, under Pamlico Surface to east of Suffolk Scarp PamlicoFm.isloyered loamy sand and some siltier beds. Generdlly a fossiiiferous zone at base Talbot Fm. Pleistocene Sticky, silty loomwittifine sand, typically sandier near base. Uncorrformityat base in many places. Fossils make a basol conglomerote at Flanner Beocti Western Facies Small sequence (including James City Fm.) Generally non-fossiliferous sands and loams. Central Facies Bedded fine sands and silts Many layers witfi whole and broken fossils.LJp tofive organic layers present; there may be peat with stumps and logs or there may be organic rich sands. Eastern Facies Fossiiiferous loamy sands, with a few silty or clayey beds near top. Yorl(town Fm. Lower Pliocene Fine sands and silty sands with many fossiiiferous loyers Top 2 feet (0.6m.) very calcareous. Other calcareous zones Upper Miocene Pungo River Fm Middle Miocene Castle Hayne Fm. Eocene FIGURE 2.—Generalized and composite stratigraphic column between Swansboro (east of Jacksonville) and the Lee Creek Mine. (Small sequence = Croatan Formation.) YORKTOWN FORMATION The Yorktown Formation is regarded as upper Miocene or lower Pliocene or both. In this region it consists dominantly of fossiiiferous to barren marine fine sand and loam, silty sand and silty clay. There are many layers and lenses with abundant to few marine fossils. There is a distinct erosional surface at the contact of the Yorktown Formation with the overlying Small sequence (= Croatan Forma- tion). The uppermost Yorktown beds immedi- ately below this unconformity were apparently enriched in calcium carbonate, resulting in either a highly calcareous and compact very light green loam or a partially indurated material. This cal- careous zone is widespread through the area, but happens to be missing at the Lee Creek Mine. THE AURORA PALEOSCARP.—Contours on the top of the Yorktown Formation reveal a very gentle, buried scarp that trends north-south across our study area (Figure 3). It goes from the north bank of the Pamlico River at the Lee Creek Mine south to the coast near Morehead City. The Lee Creek Mine is on the trend of this feature, the scarp slopes 40 feet (12.2 m) in 3.5 miles (5.6 km) or about 11 feet (3.4 m) per mile (1.6 km) (Figure 3). This seemingly insignificant slope in the present interfluve areas contrasts strongly with the general slope of the post-Yorktown un- conformity of 1 foot (0.3 m) per mile (1.6 km) to the west of the scarp and 3 feet (0.9 m) per mile (1.6 km) to the east. The Pamlico estuary is cut deeply into the Yorktown Formation (Welby, 1971). Our contours at the top of the Yorktown For- mation along the Aurora paleoscarp match very well with those of Welby along this portion ofthe Pamlico River (Welby, 1971, fig. 2). Welby's data were mainly from a High Resolution Boomer survey made from a boat on the Pamlico and Pungo rivers. As a consequence, his contour lines 206 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 5 0 5 10 15 KILOMETERS for forming the restricted, shallow water marine basin in which the middle Miocene Pungo River Formation was deposited. No cause and effect relationship is claimed for the association of the Suffolk Scarp and Aurora paleoscarp, the crustal segment boundary of Brown, Miller, and Swain (1972), or the hinge near the western boundary of the Pungo River Formation (Miller, 1971). Nevertheless, this strik- ing juxtaposition deserves to be kept in mind as more data are produced. SMALL SEQUENCE (= Croatan Formation) Overlying the Castle Hayne or the Yorktown formations in the area studied (Figures 4, 5) is a FIGURE 3.—Structure contour map on top of Yorktown Formation between the Pamlico River and Morehead City, North Carolina. (Elevations in feet below mean sea level.) barely extend onto land areas. Nevertheless, the Aurora paleoscarp may be seen clearly on his map, with the position of the Lee Creek Mine marked at its crest. The present day Pamlico estuary is cut into sediments that were previously laid down in stream channels and irregular de- pressions, which were previously cut in Pliocene- Pleistocene time into the upper portion of the Yorktown Formation (Welby, 1971:201-203). The smoothness of the unconformity as mapped is somewhat misleading because our auger holes are a mile or more apart. Additional detail will undoubtedly show that the scarp is much more irregular than shown in Figure 3, because the relief at the top of the Yorktown locally may reach 30 feet (9.1 m). However, we feel that the major outlines of the scarp are reasonably well established. The boundary between crustal segment F and crustal segment G of Brown, Miller, and Swain (1972) passes through this area with a north-south trend. The boundary between these two segments is a hinge zone. This hinge zone has been cited by Miller (1971:35-41) as part ofthe mechanism 0 10 20 30 40 50 77° KILOMETERS FIGURE 4.—Location of sections of Figures 5, 6, 8, 11, and 12. NUMBER 53 207 NEUSE-TAR RIVER DIVIDE 50 Mi. W 80- « 40 0 — ** 40- 80 SO Mi. NEUSE- WHITEOAK RIVER DIVIDE W SO- Organic zone Fossils FIGURE 5.—Generalized west to east relations of Small sequence to overlying and underlying sediment. (The broken lines with question marks indicate uncertain boundaries; m-f Sand indicates medium-fine sand.) complex of beds called the Small sequence. This proposed name is derived from the community of Small in Beaufort County where the sequence was first recognized in a series of auger holes. The Small sequence includes all strata between the Yorktown Formation, or the Castle Hayne For- mation where the Yorktown is absent, and the overlying surficial sediments. It does not include the overlying Flanner Beach Formation of DuBar and Solliday (1963) or the Neuse Formation of Fallaw and Wheeler (1969), both of which are part of the Talbot msu. The Small sequence is a complex of interbedded clays to sands with one or more organic horizons. (An organic horizon may be an A-1 horizon of a buried soil, a peaty or mucky sand, or even a pure peat or muck; Daniels, Gamble, and Wheeler, 1972.) The Small sequence includes the fossiiiferous James City Formation, a nonfossiliferous facies north of New Bern, which seems to occupy the same strati- graphic position as the James City Formation and a fossiiiferous sand and nonfossiliferous sand 208 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY North South 18 Km. lilJFosslls I Sands ISilt ICIay iMori iOrganic Zone with wood etc. FIGURE 6.—North-south traverse from Lee Creek Mine to west-east traverse of Neuse-Tar divide. (Small sequence = Croatan Formation.) to clay unit east of the Suffolk Scarp. Further work is needed to clarify whether or not the Small and the James City are or are not synonymous. Oaks and Coch (1973:56) regard the James City Formation as a "possible equivalent ofthe Sedley Formation of southeastern Virginia." A 45-mile (72-km) traverse across the Talbot and Pamlico surfaces north of the Neuse-Tar divide (Figure 5) shows the division of the Small sequence into three facies. The Small sequence is largely nonfossiliferous sands in the western fa- cies. These nonfossiliferous sands interfinger to- ward the east in the vicinity of the Suffolk Scarp with sands to clays that have one or more organic horizons. Farther east the multiple organic hori- zons are replaced by the eastern facies composed of fossiiiferous sands, although the upper part of the section may contain silts to clays and occa- sional thin organic units. The sequence is sharply separated from the underlying Yorktown by the distinct changes in lithology and a distinct dis- conformity that is traceable over large areas of eastern North Carolina (Welby, 1971). The con- tact with the overlying Talbot msu is question- able in the western part of the traverse, but near the Minnesott Ridge the contact is distinct. The sediments exposed at the Lee Creek Mine have been traced south to connect with the 45-mile (72-km) traverse just east ofthe Minnesott Ridge (Figures 4, 6). In a traverse along the Neuse-White Oak river divide, the three facies of the Small sequence are similar to those found in the Neuse-Tar divide traverse (Figure 5). The fossiiiferous facies occur- ring west of the multiple horizons is the James City Formation. In the Neuse-Tar traverse the multiple organics occupy an area a few miles wide, but they apparently are limited to a very narrow band between the Neuse and White Oak rivers. In both cross-sections (Figure 5), the or- ganics occur in the immediate vicinity of the Suffolk Scarp, but this probably is a coincidence because organics occur over an area about 20 to 25 miles (32 to 40 km) wide between the Neuse and Pamlico estuaries (Figure 7). In both trav- erses (Figure 5) the Small sequence thickens con- siderably toward the east. The following section with multiple organic horizons in the Small sequence was described from an auger hole located on Beaufort County Road 1931, 0.5 mi (0.8 km) west of its junction with county road 1927; altitude 34.2 feet (10.4 m) (Figure 7: locality 1). Munsell color designations (indicated in parentheses) are used in the descrip- tions. The boundary with the underlying bed is described as abrupt (<1 inch; 2.5 cm), clear (1-2 inches; 2.5-5.1 cm), or gradual (2-5 inches; 5.1- 12.7 cm). NUMBER 53 209 V ) Ox, °'n//. INSTON I NEW BERN 1 X — *%«'—'«.* ■^ / :^' \ .JACKSONVILLE 'V/*.^ ^^ o o MINNE,SOTT 'A.—-O /FilDGE-9^ • organics in Small sequence O no orgonics in Small sequence PAMLICO SOUND MILES 5 0 5 I 1 I i_ 20 I 1 L J I I I 5 0 5 20 30 KILOMETERS FIGURE 7.—Location of bore holes. (Localities 1-3 are discussed in detail in the text.) Depth infect (meters) 0 to 0.5 (0 to 0.2) 0.5 to 5 (0.2 to 1.5) 5 to 7 (1.5 to 2.1) 7 to 11.5 (2.1 to 3.5) 11.5 to 13.5 (3.5 to 4.1) Description Road fill TALBOT MORPHOSTRATIGRAPHIC UNIT Sandy loam to loamy sand. Soil profile in the Talbot msu Pale yellow (5Y 7/3) to light gray (2.5Y 7/2) fine sand; abrupt transition to Greenish gray (5GY 5/1) sticky silt loam grading downward to dark greenish gray (5GY 4/1) loamy fine sand at 8.5 ft (2.6 m); clear to Yellow (lOYR 7/6) medium fine to fine sand; gradual to 13.5 to 21.5 Greenish gray (5GY 6/1 to 5/1) medium (4.1 to 6.6) fine to fine sand to loamy sand; lower 2 ft (0.6 m) are sticky loam; base of Talbot abrupt to Small sequence. SMALL SEQUENCE (= Croatan Formation) 21.5 to 24 Darker than very dark brown (lOYR 2/2) (6.6 to 7.3) organic clay loam; contains wood frag- ments up to 2 in (5.1 cm) long; gradual to 24 to 26 Darker than very dark brown (lOYR 2/2) (7.3 to 7.9) fine loamy sand to sandy loam; gradual to 26 to 31.5 Gray (5Y 5/1) medium fine loamy sand with (7.9 to 9.6) bodies of very dark grayish brown (10 YR 5/3); abrupt to 210 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 31.5 to 38.5 Greenish gray (GY 5/1) medium fine loamy (9.6 to 11.7) sand becoming greener than dark greenish gray (5GY 4/1) at 36 ft (11m); abrupt to 38.5 to 49 Greenish gray (5GY 5/1) medium fine sand (11.7 to 14.9) grading to medium coarse sand at 49 ft (14.9 m); abrupt to 49 to 58 Dark greenish gray (5GY 4/1) stiff sticky (14.9 to 17.7) clay to fine clay loam; clear to 58 to 63.5 Dark gray (10 YR 4/1) medium sandy clay (17.7 to 19.4) loam; common to many wood fragments 0.5 in (1.3 cm) in diameter or less; clear to 63.5 to 66.5 Dark gray (lOYR 4/1) sticky silty clay with (19.4 to 20.3) few to common wood fragments; abrupt to 66.5 to 74 Black (5YR 2/1) peaty sticky silty clay loam, (20.3 to 22.5) common small (less than 0.5 in; 1.3 cm) wood fragments, abrupt to 74 to 78.5 Dark gray (lOYR 4/1) sticky silty clay; grad- (22.5 to 23.9) ual to 78.5 to 89 Dark greenish gray (5GY 4/1) silty clay loam (23.9 to 27.1) grading to medium fine sandy clay loam at 81 ft (24.7 m); sands become fine to very fine at 86 ft (26.2 m); clear to abrupt to 89lo91 Dark gray (5Y 4/1) fine sandy clay loam (27.1 to 27.7) with few to common darker bodies of dis- seminated organic matter; base of Small sequence; abrupt to YORKTOWN FORMATION 91 to 98.5 Dark greenish gray (lOGY 4/1) medium (27.7 to 30.0) coarse sandy clay loam grading to light olive gray (5Y 6/2) "marl" with some material partially cemented by carbonate; Yorktown Formation; base of hole at 98.5 ft (30.0 m). Some ofthe variation ofthe Small sequence in its eastern distribution is given in the following section, which was described from an auger hole located approximately 2 miles (3.2 km) east of Hobucken on Pamlico County Road 1228, 1.75 miles (2.8 km) beyond end of pavement and 0.1 mile (0.2 km) from end of road (Figure 7: locality 2). Altitude 3 feet (0.9 m). Description Depth in feel (meters) Road nil 0 to 1 (0 to 0.3) PAMLICO MORPHO.STRATIORAPHIC UNIT Sandy clay loam soil profile in Pamlico msu; gradual to 1 to 5.5 (0.3 to 1.7) 5.5 to 10 Pale olive to olive (5Y 6/3 to 5/3) medium (1.7 to 3.0) sand interbedded with minor strata of fine sand; grades to gray (5Y 5/1) medium sand with minor strata of fine sand to loamy sand at 8.5 ft (2.6 m); clear to 10 to 18 Olive gray (5Y 4/2) fine sandy clay loam (3.0 to 5.5) interbedded with fine sandy loam and medium sand; base of Pamlico; abrupt to SMALL SEQUENCE (= Croatan Formation) 18 to 27 Greenish gray (5GY 5/1) sandy clay loam (5.5 to 8.2) interbedded with medium loamy sand to sand; abrupt to 27 to 28 Dark olive gray (5Y 3/2) micaceous organic- (8.2 to 8.5) rich silty clay that darkens on exposure; clear to 28 to 31.5 Greenish gray (5GY 5/1) loam sand to me- (8.5 to 9.6) dium sand interbedded with thin strata of more clayey material; abrupt to 31.5 to 57 Bluish gray (5B 6/1) "marl"; coarse and (9.6 to 17.4) medium sand with abundant shells; highly calcareous; grades to greenish gray (5GY 67/1) at 40 ft (12.2 m); clear to 57 to 66 Greenish gray (5GY 5/1) sticky silty clay (17.4 to 20.1) loam; calcareous; base of Small sequence; abrupt to YORKTOWN FORMATION 66 to 68.5 Bluish gray (5B 6/1) slightly sticky hard (20.1 lo 20.9) drilling clay loam marl; probably York- town Formation; base of hole 68.5 ft (20.9 A series of 12 drill holes in a 4-mile (6.4-km) traverse across the Suffolk Scarp clearly shows the discontinuous nature of the organic horizons and the associated beds (Figures 4, 8). Nodular or slightly indurated limestone at the top of the Yorktown is removed in places and the relief of the erosion surface is about 30 feet (9 m). The base ofthe Small sequence in the western part of the traverse is fossiiiferous sands (James City?) grading upward into fossiiiferous and nonfossili- ferous silty beds. Near the Suffolk Scarp is a complex sequence of fossiiiferous and nonfossili- ferous silty beds with intercalated but horizon- tally discontinuous organic-rich beds, which con- tain bald cypress wood (A.C. Barefoot, pers. comm.). Many of these organic horizons with cypress wood are at 20 to 50 feet (6.1 to 15.2 meters) below sea level. Fossiiiferous marine sands and nonfossiliferous sands of probable marine NUMBER 53 211 0 Km. -20 :il Sand 3 Ed Silt I " 2 Calcareous [rri zone luJ Fossils S Wood 0 Mi. Organic horizons FIGURE 8.—West to east cross-section across Suffolk Scarp near Harlow Creek south of the Neuse estuary near the Craven-Carteret county line. (Location of auger holes is indicated by short vertical lines across top of figure, and their bottoms by short horizonal lines over inverted Vs.) .3 6-1-20 origin also are found interbedded with the silts to silty clays. The organic horizons and associated silty beds pinch out to the east and interfinger with fine to medium sands. A detailed section shows the complex vertical changes in sediments of the Small sequence (Fig- ure 9). Some of the silty beds are extremely soft and semifluid and are very similar to the sedi- ments in the modern salt marshes at the mouth of the Cape Fear River. This suggests that these soft, semifluid beds have never been dried. Yet within the same section there can be extremely tough dense clays separated by organic horizons. The tough clays, if deposited in a salt-water environment, must have been dried or de-watered sometime during their history. The organic zones and the cypress wood indicate at least short pe- riods during which vegetation was growing in a swamp or brackish water marsh. The relation between the buried organic hori- zons and the surficial Talbot msu is well illus- trated at Flanner Beach on the south side of the Neuse estuary. The base of the section is a gray 4-, -10 2- 0--0 WPZ Loom, Bh (Humate) Medium fine sandy loom Medium fine sand Fossil bed Base of Pomlico novi -10 Silty clay Organic zone-silty clay loam, SYR 2/2 Medium loamy sand -30 Organic zone-silty clay, I0YR3/2 Clay-silty clay with wood, Taxodium distichum (Bald Cypress) 12-^40 14 -50 ^ Oi Tough cloy Organic zone-silt loam, lOYR 2/2 Clay Clay Silt Sand ^^ EZilFossils r»l Wood Organic horizons FIGURE 9.—Small sequence near Harlow Creek showing complex vertical changes in sediments. This hole is at 0 miles in Figure 8. 212 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY to olive brown clay exposed about 2 feet (0.6 m) above the estuary. The organic horizon at the top of the clay (Figures 4, 10) is truncated by a Dinocardium layer at the base of the overlying Talbot. At places logs or roots are imbedded in the clay but buried by the Dinocardium layer. The clay is without marine fossils in the exposed sec- tion, but fossils do occur in four nearby drill holes. There apparently was some weathering of the clay, as shown by the organic carbon and oxida- tion of iron, before the overlying Talbot msu was deposited. DEPOSITIONAL ENVIRONMENT.—Our studies of fossils contained within the Small sequence are far from complete, but some speculation about depositional environments can be made. There seems to be a distinct possibility that the relatively clean fossiiiferous sands in the eastern part of the Small sequence represent fully marine conditions. There are no or few muds and organic horizons within the sands that can be interpreted as estu- ary or lagoon facies. An upper clayey part occurs in about one-half of the bore holes, and these may represent localized lagoonal conditions. The organic horizons of the Small sequence indicate that the sediments had to be exposed to subaerial conditions for a few hundred years to develop relatively thick organic layers and allow cypress stumps 6 feet (1.8 m) or more in diameter to develop (Flanner Beach section. Figure 10). The presence of marine shells intercalated with organic debris in other sections suggests that these organic layers formed very close to sea level. The relatively restricted horizontal area occupied by multiple organic layers suggests that the fluctuat- ing marine, lagoon-like, and subaerial conditions remained in about the same geographic location throughout the deposition of 20 to 40 feet (6 to 12 m) of sediment. There appears to be little evidence that these conditions shifted westward as sea level rose. The origin of the nonfossiliferous sands of the western facies of the Small sequence is open to considerable question. We are not sure whether these sands interfinger with the James City For- mation farther to the south or whether they rep- resent a leached James City Formation. A third possibility is that they are lagoonal to fluvial sands that may or may not be related to the James City Formation. There is a problem in determining the behavior of sea level during the deposition of the Small sequence. If each organic horizon represents a separate transgression and regression, then in Depth (Meters) 1400 Ft. Organic horizons FIGURE 10.—Sediments expo.sed at Flanner Beach, North Carolina. This section is a composite of auger holes (shown by the vertical columns) and the exposed section in the bluff on the south bank of the Neuse estuary. The portion of the bank covered by slump is shown. The section below the beach line is known from auger hole information only, (msu = morphostratigraphic unit; Small .sequence = Croatan Formation.) NUMBER 53 213 some areas there have been at least five of these. However, it can be suggested also that sea level may have risen in minor increments during dep- osition of the Small sequence, and sediment ac- cumulation may more or less have kept pace with this rise in the manner known to be taking place in North Carolina lagoons today (Ingram, 1968). The first hypothesis requires repeated rise and fall of sea level with transgression and regression, but the second one requires only that the shore- line remain somewhere within a limited geo- graphic area as sea level rises. The Small sequence is well exposed in the Lee Creek Mine. The unit exhibits many of the as- pects of variability that characterize the Small sequence elsewhere. The most spectacular bed in the mine is a lens of extremely fossiiiferous incline- bedded sands on the north face of the pit. Mol- lusks and colonial corals are very abundant. The steepness ofthe inclined bedding, about 15°, with large shells as clasts is quite spectacular. It is analogous to the over-wash of a sandbar cited at Yorktown, Virginia, in the Yorktown Formation (G. H. Johnson, 1972:22, 39). The shell unit at Lee Creek Mine grades vertically and laterally into beds of sand with fossils scattered singly or as stringers or into unfossiliferous silty sand. The Small sequence in its upper portion is fine silty sand with some organic stains, so contrasting in appearance to the incline-bedded fossiiiferous sand that one might easily place it in the overly- ing Pamlico msu. However, the layer of scattered pebbles, cobbles, and boulders that marks the base ofthe Pamlico msu lies above the organically stained Small sequence. RELATION OF SMALL SEQUENCE TO SURFICIAL SEDIMENTS.—The relations between the Talbot and Pamlico msu and the Small sequence are important in dating and developing a history of the lower Coastal Plain. Locally, such as at the Flanner Beach section, there is a distinct litho- logic disconformity separating the Talbot msu and the Small sequence. Yet, if one considers the multiple organic layers ofthe Small sequence and their relations to one another, there are discon- formities within the Small sequence that are just as distinct and may represent as much change in depositional environment. Are we, then, justified in placing a formational break at the base of the Dinocardium layer in the Flanner Beach section? We think we are justified because this disconform- ity has been traced over a wide area, whereas the disconformities within the Small sequence are of very limited geographic extent. The relation between the Pamlico msu and the underlying Small sequence seems to be much more distinct than between the Talbot msu and the underlying Small sequence. The Pamlico nearly everywhere is sharply separated from the Small sequence by a basal fossil "hash" layer that can be traced for miles, but is not present at the Lee Creek Mine. About 8 out of 10 holes have this hash layer and seldom does it vary more than 3 to 5 feet (0.9 to 1.5 m) in altitude, usually less within a local area. TALBOT AND PAMLICO MORPHOSTRATIGRAPHIC UNITS The major morphostratigraphic units (msu) in the area are the Talbot and Pamlico of Stephen- son (1912) and other authors. These surface units are lithologically similar to, and difficult to dis- tinguish from, each other. TALBOT MORPHOSTRATIGRAPHIC UNIT.—By our definition, the Talbot msu is the surface unit that occurs between the toe of the Walterboro Scarp, altitude 45 feet (13.7 m) and the top ofthe Suffolk Scarp (Figure 6). The Minnesott ridge sand at the top of the Suffolk Scarp in Beaufort County and Pamlico County is probably associated with the Pamlico msu. We have not mapped the exact areal distribution of the Talbot throughout the Neuse Basin, although its eastern limit, the Suf- folk Scarp, has been mapped (Figures 1, 4, 7). The Talbot msu has almost any texture from sand to silt to clay. It is coarsest at the base and becomes finer toward the top in about half our drill holes. However, as at the Flanner Beach section (Figure 10), there are vertical and hori- zontal changes in texture over short distances that can range from sands to clays with any one lithology occurring at any level within a vertical section. 214 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Fossils occur near the base of the Talbot only in its eastern part (Figures 5, 8), and then usually in a somewhat clayey matrix. Only 20 of 59 bore holes in the Talbot have marine fossils somewhere within the section, and the largest number of holes with fossils are south of the Neuse River. The fossils may form a very concentrated layer such as the basal Dinocardium zone at Flanner Beach (Figures 4, 10), or they may be sparsely scattered throughout a silty or clayey matrix. The fossils at Flanner Beach (DuBar and Sol- liday, 1963) and at sections near Bear Creek (Fallaw and Wheeler, 1969; Fallaw, 1973) place the Talbot definitely within the Pleistocene. PAMLICO MORPHOSTRATIGRAPHIC UNIT.—The Pamlico msu includes all surface sediments east of the Suffolk Scarp with the exclusion of Recent eolian sands and swamp or marsh deposits in river valleys and lagoons. The Minnesott Ridge (Figure 7) is a unit that is associated with the Pamlico. It is a ridge of sand 1 mile (1.6 km) or less wide, whose seaward face forms part of the Suffolk Scarp between the Pamlico and Neuse estuaries. The northern terminus ofthe Minnesott Ridge is about 4 miles (6.4 km) west of the Lee Creek Mine, and the southern terminus is at Minnesott Beach on the Neuse River. The ridge has a maximum altitude of 65 feet (19.8 m). It rises 25 to 30 feet (7.6 to 9.1 m) above the Talbot surface to the west and 50 feet (15.2 m) above the Pamlico surface to the east. The topography of the ridge crest is extremely variable; it may be flat or dune-like with small irregular depressions. Texturally the ridge is sand to loamy sand. We have not found clay or silt lenses in the more than 20 holes we have drilled in it. The ridge sand, along its western edge buries organic layers at the top of the Talbot (Figures 4, 11). Near the center and eastern part of the ridge, the organic layers and the ridge sands overlie sands and other lithologies of the Talbot msu. The east side ofthe ridge is the Suffolk Scarp and the ridge sands merge laterally to the east with the clayey and sandy Pamlico which everywhere lies below 20 feet (6.1 m). The sand texture of the ridge plus the perched Minnesott Ridge Sand I I I i I I I i ; I 1 w i I ParDlico msu- lo «- 9 Small sequence Yorktown Fm. WEST 60-* * Talbot msu 40- .^ > ' 'i » • I I >/ ( » c , ) 1 j r- ; Ll:.ll^^-^^-^^-^'^.-:''','.'ll^L:..'.-'.:'.'-'-'TALBdT?'. ■, T- SMALL SEQUENCE 3 Km. J —I— 0.5 I 2.0 Mi. LO LS I SILT. CLAY SAND i-J FOSSILS FIGURE 12.—Cross-section of Minnesott Ridge near Arapahoe, North Carolina. (Small sequence = Croatan Formation.) Discussion We now believe that our studies show (1) a persistent unconformity between the Yorktown Formation and Croatan Formation (Small se- quence), (2) the presence of both marine and nonmarine aspects of the Croatan Formation, and (3) an unconformity between the Croatan and later Pleistocene units. Typically, an organic- rich nonmarine layer in the Croatan Formation lies just beneath this unconformity. These char- acteristics ofthe Croatan Formation may be seen both regionally and in the walls of the Lee Creek Mine. During the initial work on the Atlantic Coastal Plain, Stephenson (1912) separated the Talbot and Pamlico msu largely on the supposition that a scarp meant a new cycle of transgression and regression. He mapped a different formation be- tween each pair of scarps, although he had little direct evidence of formational changes. The scarps and surfaces that Stephenson mapped in North Carolina generally hold; he did an excel- lent job in a short time with very poor topo- graphic coverage. Whether or not a new forma- tion starts at the toe of each scarp, however, may be debated for years because even intensive work with drill rigs and modern laboratory techniques frequently leaves us only the numbers and few facts, but not unequivocal proof We have three detailed traverses across the Suffolk Scarp that will allow us to speculate somewhat on the possible stratigraphic changes across it (Figures 4, 8, 11, 12). South ofthe Neuse estuary, the Suffolk Scarp is a somewhat indis- tinct feature with a toe altitude between +15 and +20 feet (+4.6 and +6.1 m). Several drill holes across the scarp suggest that the Pamlico to the east is inset into and slightly below the Talbot that lies to the west (Figure 8). This relation is based upon a fossil hash assumed to be at the base of the Pamlico that occurs at a reasonably uniform level over wide areas east of the scarp. The traverse across the Suffolk Scarp in Beaufort County can be interpreted in at least two ways. We presently argue that the Minnesott Ridge sands grade into the silty and clayey upper part of the Pamlico at the toe of the scarp. The fossiiiferous bed near the base of the Talbot west ofthe scarp is truncated by the overlying Pamlico farther east. The change in lithology across this contact between the Pamlico and Talbot east of the scarp may be minor. A second possibility is that the Pamlico and Talbot are one sedimentary unit. This would make the Minnesott Ridge a post-depositional feature probably associated with a high stand of sea level at about +20 feet (+6 m), and it would make the Pamlico surface an erosion surface. This was the view of Fallaw and Wheeler (1969) who regarded the Neuse Formation (the type section of which is in the Talbot msu on the North bank of the Neuse estuary) as continuous across the scarp. NUMBER 53 217 Literature Cited Austin, J.A., B.F. Molnia, and H.A. Curran 1973. Stratigraphy of Marine and Estuarine Pleistocene Beds Exposed along the Pamlico River, North Carolina. Geological Society of America, Abstracts with Programs, 4:375. Brown, P.M., J.A. Miller, and F.M. Swain 1972. Structural and Stratigraphic Framework, and Spatial Distribution of the Permeability of the Atlantic Coastal Plain, North Carolina to New York. United States Geological Survey Professional Pa- per, 796: 79 pages, 59 plates. Cooke, C. Wythe 1937. The Pleistocene Horry Clay and Pamlico Forma- tion near Myrtle Beach, South Carolina.yoM^-na/ of the Washington Academy of Sciences, 27:1-5. Dall, W.H. 1892. On the Marine Pliocene Beds of the Carolinas. In W.H. Dall, Part II: Introductory o/Contributions to the Tertiary Fauna of Florida: Tertiary Mol- lusks of Florida. Wagner Free Institute of Science Transactions, 3(2):201-217. Daniels, R.B., E.E. Gamble, and C.S. Holzhey 1972. Thick "h" Horizons in the North Carolina Coastal Plain, I: Morphology and Relation to Texture and Soil-ground Water. Agronomy Abstracts, Ameri- can Society of Agronomy Annual Meeting, 64:124. Daniels, R.B., E.E. Gamble, and W.H. Wheeler 1972. Buried Pre-Talbot Organic Horizons Related to Changing Sea Level in Eastern North Carolina: Geological Society of America, Abstracts with Programs, 4:68-69. Daniels, R.B., E.E. Gamble, W.H. Wheeler, and C.S. Hol- zhey 1972. [Some Details of the Surficial Stratigraphy and Geomorphology ofthe Coastal Plain between New Bern and Coats, North Carolina.] In Annual Meet- ings and Field Trip Guidebook of Carolina Geological Society and Atlantic Coastal Plain Geological Association, October 7-8, 1972. 68 pages. Raleigh: Department of Natural and Economic Resources, North Car- olina. DuBar, J.R., and J.R. Solliday 1963. Stratigraphy of the Neogene Deposits, Lower Neuse Estuary, North Carolina. Southeastern Geol- ogy: 4:213^233. Fallaw, W.C. 1973. Depositional Environments of Marine Pleistocene Deposits in Southeastern North Carolina. Bulletin of the Geological Society of America, 84:257-268. Fallaw, W.C, and W.H. Wheeler 1969. Marine Fossiiiferous Pleistocene Deposits in Southeastern North Carolina. Southeastern Geology, 10:35-54. Frye, J.C, and H.B. Willman 1960. Classification ofthe Wisconsin Stage in the Lake Michigan Glacial Lobe. Illinois State Geological Sur- vey Circular, 285: 16 pages. 1962. Note 27—Morphostratigraphic Units in Pleisto- cene Stratigraphy. Bulletin ofthe American Association of Petroleum Geologists, 46:112-113. Holzhey, C.S., R.B. Daniels, and E.E, Gamble 1972. Thick "h" Horizons in the North Carolina Coastal Plain, II: Physical and Chemical Properties, and Rates of Organic Additions from Surface Sources. Agronomy Abstracts, American Society of Agronomy An- nual Meeting, 64; 125. Ingram, R.L. 1968. Vertical Profiles of Modern Sediments along the North Carolina Coast. Southeastern Geology, 9:237- 244. Johnson, B.L. 1906. Pleistocene Terracing in the North Carolina Coastal Plain. Science, 26:640-642. Johnson, G.H. 1972. Geology of the Yorktown, Poquoson West, and Poquoson East Quadrangles, Virginia. Virginia Division of Mineral Resources Report of Investigations, 30: 57 pages. Mansfield, W.C. 1928. Notes on Pleistocene Fauna from Maryland and Virginia, and Pliocene and Pleistocene Faunas from North Carolina. United States Geological Survey Professional Paper, 150-F: 129-140. Miller, J.A. 1971. Stratigraphic and Structural Setting of the Middle Miocene Pungo River Formation of North Caro- lina. 82 pages. Ph.D. dissertation, University of North Carolina at Chapel Hill. Mixon, R.B., and O.H. Pilkey 1976. Reconnaissance Geology of the Submerged and Emerged Coastal Plain Province, Cape Lookout Area, North Carolina. United States Geological Survey Professional Paper, 859: 45 pages. Oaks, R.Q., Jr., and N.K. Coch 1973. Post-Miocene Stratigraphy and Morphology, Southeastern Virginia. Virginia Division of Mineral Resources Bulletin, 82: 135 pages. Stephenson, L.W. 1912. The Quaternary Formations. In W.B. Clark et al., The Coastal Plain of North Carolina. North Caro- lina Geological and Economic Survey, 3:266-290. 218 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Swanson, V.E., and J.G. Palacas 1965. Humate in Coastal Sands of Northwest Florida. United States Geological Survey Bulletin, 1214-B: 29 pages. Thom, B.C. 1967. Coastal and Fluvial Landforms: Horry and Mar- ion Counties, South Carolina. Louisiana State University Studies, Coastal Plain Series, 19: 75 pages. Welby, C.W., 1971. Post-Yorktown Erosional Surface, Pamlico River and Sound, North Carolina. Southeastern Geology, 13:199-205. Welch, J.S., R.W. Frey, and E.S. Belt 1972. Physical and Biogenic Sedimentary Structures as Depositional Indicators in the Pleistocene of North Carolina. Geological Society of America, Abstracts with Programs, 4:68-69. Wheeler, W.H., R.B. Daniels, and E.E. Gamble 1979. Some Stratigraphic Problems of the Pleistocene Strata in the Area from Neuse River Estuary to Hofmann Forest, North Carolina. In G.R. Baum, W.B. Harris, and V.A. Zullo, editors. Structural and Stratigraphic Framework for the Coastal Plain of North Carolina. Field Trip Guidebook, Car- olina Geological Society and Atlantic Coastal Plain Geo- logical Association, pages 41-50. Raleigh: Depart- ment of Natural Resources and Community De- velopment, Geological Survey Section (North Car- olina). Observations on the Paleoecology and Formation of the "Upper Shell" Unit, Lee Creek Mine H. Allen Curran and Patricia L. Parker ABSTRACT The "Upper Shell" unit at the Lee Creek Mine (Pliocene age, maximum thickness 3 m) is re- markable for its concentration of well-preserved mollusk shells in a sparse quartz sand matrix, and it is dominated by several species of bivalves, with many shells articulated. The unit can be subdi- vided into three bivalve assemblage zones char- acterized by associations of dominant species. Zone 1 is dominated by Mercenaria mercenaria, an infaunal, shallow- to medium-burrowing, siphon- ate clam. Zone 2 is characterized by an epifaunal bivalve assemblage that includes Glycymeris amer- icana, Argopecten eboreus, Anomia simplex, and Ostrea meridionalis. Thin but highly concentrated accu- mulations of Argopecten and Anomia form distinct layers within zone 2. Zone 3 is marked by a return o^Mercenaria mercenaria accompanied by specimens of Geukensia sp. and an increase in oyster shells. The characteristics of the zones of the "Upper Shell" unit strongly suggest that these shell beds were formed by a series of localized catastrophic events that produced mass mortality of the mol- luscan assemblages, rather than by processes of gradual shell accumulation. The disappearance of Mercenaria mercenaria from the sequence may have been due largely to the inability of juveniles of this species to penetrate a shell pavement formed immediately after a mass mortality event. Return of Mercenaria mercenaria in zone 3 marks a change in bottom environmental conditions in the area. The overlying "Shell Hash" unit con- tains the bivalve Corbicula densata, representative of lower salinity conditions. This unit consists H. Allen Curran and Patricia L. Parker, Department of Geology, Smith College, Northampton, Massachusetts 01063. primarily of shell material reworked from the underlying "Upper Shell" unit and probably rep- resents an accumulation formed in an estuarine tidal channel. Introduction One of the most prominent units revealed by strip mining operations at the Lee Creek Mine of Texasgulf Inc. is a shell bed with a maximum thickness of 3 meters located toward the top of the exposed stratigraphic sequence. Known lo- cally as the "Upper Shell" unit, these beds are remarkable for their abundant, well-preserved megafossils, consisting largely of bivalves, gastro- pods, and large coral heads. Many ofthe mollusk shells are whole and unworn, bivalves are fre- quently articulated, and some of the shells retain faint coloration patterns. During the summers of 1971 and 1972, the actively worked and advancing western face of the mine (approximately 1100 m in length) was cut by a series of five equidistantly spaced, par- allel drainage trenches of up to 240 meters in length, which ran normal to the mine's west wall (Figure 1). Each of these trenches cut to the base or slightly below the base of the "Upper Shell" unit and provided continuous and easily accessi- ble exposures. At first glance the "Upper Shell" unit appeared to be a homogenous sequence, conspicuously dominated by large, flat-lying bi- valve shells. Closer inspection revealed that the unit consisted of a sequence of three major shell 219 220 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 3.5 3 "SHELL HASH" UNIT uncTU SCALLOP "*■ LAYERS "♦' 2 3 4 r "■' 3 -A,. 1 \ PIT 2 N 2 / // // // / / // // // /■ / II II II / 3 L , SOUTH "* WALL 2 A A // 1 /I l\ II 1 2 ^ ^ ^ ^ 2 2 1 2 2 1 2 1 1 1 FIGURE 1.—Correlation of bivalve assemblage zones 1-3 of the "Upper Shell" unit between sections established in the five drainage trenches that ran normal to the west wall of the Lee Creek Mine in 1971-1972. Arrows mark positions of layers with concentrations of Argopecten and Anomia shells that cut zone 2. The "Shell Hash" unit also is exposed prominently on the south wall of the mine. Inset diagram shows the spatial distribution of the drainage trenches in 1971. layers, each characterized by a distinctive assem- blage of bivalve species. The varied nature and good exposures of the "Upper Shell" unit presented an excellent oppor- tunity for detailed sampling of the shell layers. Paleoecological data derived from the samples include the relative abundances of dominant spe- cies in each layer, changes in species abundances through the section, and size-frequency analyses of selected species. These data have enabled a paleoecological interpretation of the varying bi- valve associations. The question of how extensive shell layers like this one are formed is intriguing, and our work has enabled us to speculate con- cerning mechanisms of origin of these shell beds. The full extent of the "Upper Shell" unit was examined in each of the drainage trenches and the thickness of each faunal zone measured and recorded. Bulk samples were taken from each shell layer and all trenches in the 1971 trench positions (Figure 1). The trench network permit- ted close correlation of faunal zones between the five trenches. Samples of 1 cubic foot (0.028 m) volume (dimensions IXIXI ft where possible) were dug from the trench faces and placed in a large wooden wash box similar to that described by McKenna (1965). Samples were then field washed in the drainage stream at the base of each trench and boxed for laboratory study. One sam- ple was taken from each layer thinner than 0.3 meter and two samples from layers thicker than 0.3 meter. ACKNOWLEDGMENTS.—We wish to thank the personnel of Texasgulf Inc. for permission to work in the mine. June Crawford and Jack Hird of Texasgulf Inc. were particularly helpful in mak- ing arrangements for us. Edward Belt and John Welch of Amherst College and Jeremy Reiskind of Mt. Holyoke College assisted in the field work and provided much helpful discussion. Useful criticism ofthe manuscript was provided by Rob- ert Gernant of the University of Wisconsin-Mil- waukee, Druid Wilson of the National Museum of Natural History, Smithsonian Institution, and Blake Blackwelder and Lauck Ward of the U.S. Geological Survey. The project was supported by National Science Foundation COSIP Grant GY- 7657 to Amherst, Mt. Holyoke, Smith, and Wil- liams Colleges. The Stratigraphic Setting The "Upper Shell" unit consists dominantly of flat-lying, well-preserved bivalve shells in a poorly consolidated matrix of light gray, fine to coarse quartz sand and fragmented shell. At most points of exposure the shells and shell fragments are tightly packed with little surrounding matrix (Figure 2A, B), although the amount of fine sand, silt, and clay increases greatly at some places. Designated by Gibson (1967:639, fig. 4) as unit 8 of the Yorktown Formation, the "Upper Shell" unit directly overlies a partly indurated, medium to coarse, carbonate cemented quartz sandstone known locally as the "Boulder Bed" (unit 7 of Gibson). The detailed stratigraphy ofthe entire section above the "Boulder Bed" ofthe Yorktown Formation, including all units recognized in this study, is presented in Figures 3 and 4 of Belt, Frey, and Welch (this volume). NUMBER 53 221 Questions concerning the age and formational assignment of the "Upper Shell" unit have not been entirely resolved. Although the shell bed previously has been assigned to the Yorktown Formation (Gibson, 1967), the contact between the "Upper Shell" unit and the underlying "Boulder" bed is easily recognized and may rep- resent a significant break in the local stratigraphic sequence. Hazel (p. 84, herein) refers Gibson's units 6-9 of the Yorktown Formation to the Croatan Formation and considers these beds to be late Pliocene-early Pleistocene in age. The "Upper Shell" unit has been traced regionally to the south through samples from auger drillings reported on by Wheeler, Daniels, and Gamble (this volume). They use the name James City Formation (or Croatan Formation) for these shell deposits and consider them to be of Pliocene age. This correlation is supported by DuBar, Solliday and Howard (1974:109). Description of the Shell Bed At the time of this study (1971-1972), weath- ering and ground water percolation had caused considerable slumping and iron oxide discolora- tion of the trench walls. On fresh surfaces, the bivalve shells formed distinct horizontal layers dominated by one or two species. We were able to recognize three distinct bivalve faunal assem- blages in the "Upper Shell" unit in the area of the trenches. The faunal zones were dominated by abundant, largely mutually exclusive occur- rences of three species: Mercenaria mercenaria (Lin- naeus), Glycymeris americana (DeFrance), and the scallop Argopecten eboreus (Conrad). We measured a maximum thickness of 3 meters for the "Upper Shell" unit at the 1971 location of trench 2 (Fig- ure 1). At this location and in trench 3, three faunal assemblage zones could be recognized and established. Zone 2 is cut by several thin (0.15 m) layers dominated by Argopecten eboreus. Trenches I, 4, and 5 had incomplete and thinner sections ofthe "Upper Shell" unit, but one or more zones ofthe unit were present in each of these trenches. Characteristics of the faunal zones are illustrated in Figure 2. The dominant faunal constituents and maximum thickness of each zone are as follows: ZONE 1 (maximum thickness 0.6 m): Mer- cenaria mercenaria (Linnaeus) dominant, ZONE 2 (maximum thickness 2.15 m): Glycymeris americana (DeFrance) dominant, zone cut by thin layers (0.15 m) composed primarily of shells oi Argopecten eboreus (Conrad) and Anomia simplex d'Orbigny. ZONE 3 (maximum thickness 0.25 m): Mercenaria mercenaria (Linnaeus), dominant; Ostrea meridion- alis Heilprin and Geukensia sp., common. The horizontal bedding pattern ofthe "Upper Shell" unit is broken by steeply dipping (up to 30°), shell-dominated cross-beds along much of the north wall of the mine. Shells in these beds are disarticulated and often broken. The large valves and valve fragments that make up the bulk of these beds typically exhibit worn surfaces re- sulting from transport under high energy condi- tions. The anomalous bedding pattern and worn condition of shells forming these cross-beds indi- cates that these beds were deposited after forma- tion of the "Upper Shell" unit. As one moves laterally away from the cross-beds into the area of trench 1 (west), beds of zone 1, characterized by flat-lying, often articulated Mercenaria merce- naria, become visible. We interpret these cross-beds as representative of a localized event of reworking and redeposition of shells derived primarily from the "Upper Shell" unit. This reworking event possibly oc- curred during the formation and filling of a tidal channel as part of the general reworking of shells associated with the formation of the overlying "Shell Hash" unit. A common component of the cross-bedded shell accumulation is large valves of the oyster Crassostrea virginica (Gmelin). This spe- cies does not occur in the beds of the "Upper Shell" unit, but the species may have flourished in the vicinity of or along the banks of the now- filled tidal channel, thus explaining its occurrence with the reworked shells. Modern analogs sup- porting this interpretation have been reported by Howard and Frey (1973:1177, fig. 8) and Wie- demann (1972) from estuaries of the Georgia coast. Here Holocene shells and shells reworked 222 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 2.—A, Contact between the "Boulder Bed" and Mercenaria dominated zone 1 of the "Upper Shell" unit, a. Close-up of zone 1 showing large Mercenaria valves (1) which form a hard shell pavement at top ofthe zone, c, Concentration of epifaunal bivalves, Glycymeris (2), Anomia (3), and Argopecten (4) characterize zone 2. D, Thin layers of concentrated valves of Argopecten (5) and Anomia (6) form a subassemblage within zone 2. E. South wall of mine showing sharp contact (arrows) between "Upper Shell" unit and overlying "Shell Hash" unit, F. Close-up of zone 2 showing articulated Glycymeris specimens. NUMBER 53 223 from pre-Holocene deposits occur in tidal channel accumulations, with valves of C. virginica fre- quently dominating the accumulation. Overlying the "Upper Shell" unit is a shell hash bed of variable thickness (maximum thick- ness 0.4 m in trench 2; 1.2 m along south wall of mine, see Figure 2E). This zone is referred to as the "Shell Hash" unit by Belt, Frey, and Welch (this volume). It consists of poorly sorted, fine to coarse quartz sand with variable amounts of silt and clay, woody debris, and abundant shell frag- ments, including whole, articulated valves of the clam Corbicula densata (Conrad). As will be dis- cussed subsequently, the presence of C. densata indicates a significant change in paleodeposi- tional conditions to an estuarine environment of lowered salinity. We interpret the "Shell Hash" unit to represent an episode of shell reworking with the major part of the fragmented shell de- rived from the underlying "Upper Shell" unit. As Howard and Frey (1973:1177) have pointed out, stratigraphic mixing of this type probably has occurred repeatedly in the Cenozoic history ofthe Atlantic Coastal Plain. Belt, Frey, and Welch relate the "Shell Hash" unit to the depositional cycle that includes their overlying "Mud and Sand" unit. At many locations in the pit, we found that the upper part of the shell hash in- creased markedly in silt and clay content and formed a gradational contact with the overlying "Mud and Sand" unit, whereas at other points, the "Shell Hash" unit is cut by channels now filled with mud and sand. It is, therefore, difficult to establish precise timing for the reworking event, but the evidence for reworking as repre- sented by the "Shell Hash" unit is strong. Paleoecology and Paleodepositional Environments Based on its foraminiferal assemblage, Gibson (1967:646) postulated that the "Upper Shell" unit was deposited in warm-temperate subtidal waters of 15 meters or less. Blackwelder and Ward (in prep.) have completed a detailed taxonomic study of the molluscan fauna of the "Upper SS8 Scallop Layers —.—Ostrea Glycymeris N=188 Mercenaria N=469 0 10 20 30^ FIGURE 3.—Plot of the relative abundances of Glycymeris americana, Mercenaria mercenaria, and Ostrea meridionalis based on the number of specimens of each species in bulk samples from trench 3. (N = total number of identifiable bivalve specimens in each sample.) Shell" unit and adjacent beds, and they suggest that these beds, including the overlying "Shell Hash" unit, originated during the development of an offshore bar system. We agree that shallow subtidal conditions existed during the time of formation of the "Upper Shell" unit, but we propose that the overlying "Shell Hash" unit represents an accumulation of primarily reworked shells that formed in an estuarine tidal channel. The pronounced transition that occurs in the composition of the molluscan assemblages of the "Upper Shell" unit (Figure 3) indicates that sig- 224 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY nificant changes occurred in the molluscan bot- tom communities during formation of this shell bed. The molluscan assemblage zones can be grouped into several different types based on the life mode of the dominant bivalve species. The first of these assemblages, represented by zone 1, consists dominantly of infaunal suspension feed- ers characterized by a population of large Merce- naria mercenaria (Figure 2A, B). These shallow- to moderately deep-burrowing, siphonate clams were sediment-surface suspension feeders. Large numbers of specimens of the shallow-infaunal, nonsiphonate suspension feeders Astarte concentrica Conrad and Venericardia granulata Say also were present. Other shallow-infaunal suspension feed- ers include species of the bivalve genera Abra, Anadara, Corbula, Diplodonta, Macrocallista, Noetia, and Spisula, all of which occurred in small num- bers. Numerous juveniles of Glycymeris americana, small oysters, and a few scallops were the major epifaunal suspension feeding bivalves. Carnivo- rous and herbivorous gastropods in low numbers also formed part of the epifauna. Zone 2 represents a second bivalve assemblage dominated by abundant epifaunal/semi-infaunal suspension feeders, particularly Glycymeris amen- cana (Figure 2c, F). Most species oi Glycymeris have had or have a life mode similar to that of the modern species G. pectinata (Gmelin). Stanley (1970:127-128) found that in the Florida Keys living individuals of C pectinata commonly lie free on coarse bottom sediments or buried under a thin layer of sediment with posterior current openings and sometimes the shell margin itself exposed. Indeed, Thomas (1975) has postulated that glycymerids, with their evolutionarily con- servative and functionally generalist traits, have occupied coarse bottom and current-swept shal- low marine environments throughout their his- tory. Thomas' studies (summarized 1975:223- 225) of the occurrence of G. americana in Neogene beds of the Atlantic Coastal Plain show that G. americana favored an unstable shell gravel-sandy substrate. Although G. americana may have been covered by a thin layer of sediment, it lived essentially unprotected from strong current or wave action sufficient to cause burial and mass mortality. Other epifaunal suspension feeders present in zone 2 include numerous specimens of Ostrea meridionalis which are larger (10-14 cm) than those of zone 1, and concentrations of large numbers of Anomia simplex d'Orbigny, Argopecten eboreus, and the gastropod Crepidula aculeata (Gme- lin). Mercenaria mercenaria is completely absent from this zone, although specimens of the shal- low-infaunal, nonsiphonate bivalves Astarte concen- trica and Venericardia granulata remained common. Within zone 2 are thin but persistent shell layers of a bivalve subassemblage, composed al- most exclusively of flat-lying, frequently articu- lated valves of the scallop Argopecten eboreus and single valves of Anomia simplex (Figures 2D, 4), both epifaunal suspension feeders. The origin of these layers will be discussed in the following section. 50% An intriguing paleoecological question pre- sented by this sequence within the "Upper Shell" FIGURE 4.—Plot of the relative abundances of Argopecten eboreus and Anomia simplex indicating the concentrations of these epifaunal bivalves in thin layers within zone 2. (N = total number of identifiable bivalve specimens in each sam- ple; dashed line = Anomia; solid line = Argopecten.) NUMBER 53 225 unit is, "What caused the disappearance of the Mercenaria mercenaria-dovci\n2iL\.ed infaunal assem- blage and its replacement by the Glycymeris amer- zVa/jQ-dominated epifaunal assemblage?" The large number of articulated valves and the essen- tially unworn condition of the fossils strongly suggest that these clams were killed by a cata- strophic event resulting in local mass mortality^ without subsequent transport of the shells. Mass mortality of shallow-infaunal clams like M. mer- cenaria would likely result in the formation of a shell pavement on the substrate surface. As sug- gested by Gernant (1970:54), the formation of a shell pavement (Figure 2B) would make it difficult or impossible for shallow- to medium-depth bur- rowers, such as M. mercenaria, to resettle until a soft bottom substrate of sufficient depth is re- formed. Thus a likely explanation for the disap- pearance of M. mercenaria above zone 1 would be that, following the formation of a shell pavement, juveniles of M. mercenaria were prevented from recolonizing because they could not penetrate the shell pavement. This would open the way for G. americana, an essentially epifaunal clam capable of colonizing the newly formed shell pavement, to establish itself and become the dominant bivalve of the overlying zone. Juvenile specimens of some infaunal bivalves which reach large adult sizes, such as Eucrassatella, Macrocallista, and Barbatia, do occur with Glycy- meris americana in small numbers. If lack of suffi- cient depth of sediment were the only deterrent to Mercenaria mercenaria survival, it would seem likely that some juveniles of M. mercenaria would ^ Kranz (1974:237) has used the term "anastrophe" to refer to catastrophes of limited scope and area that generally produce mass mortality in the affected area. Kranz attri- buted this definition of "anastrophe" to Brongersma-Sanders (1957). However, Brongersma-Sanders (1957:941) did not consider "anastrophe" to be the proper word to use in referring to localized catastrophes affecting marine orga- nisms, although she noted that the word had been used previously in this sense. In this paper, we consider localized catastrophic events and catastrophic burial events to be "anastrophes" and "anastrophic burial" events as defined by Kranz, but we follow the recommendation of Bron- gersma-Sanders that the word "anastrophe" not be used. be present. The disappearance of M. mercenaria may represent only local displacement of this clam to a nearby geographic area that presented more favorable bottom conditions for burrowing. On the other hand, this disappearance might also be related to more fundamental environmental changes, such as increased water depth and/or salinity that proved detrimental to M. mercenaria. The transition from zone 1 to zone 2 in the "Upper Shell" unit can be recognized clearly in the area of the Lee Creek Mine, but further work on the regional characteristics of the "Upper Shell" unit will be necessary before the true mag- nitude of this transition can be gauged. In zone 3, Mercenaria mercenaria reappears (Fig- ure 3), and this reappearance is coupled with a sharp decline in the population oi Glycymeris amer- icana. Specimens of the shallow-infaunal, suspen- sion-feeding bivalve Eucrassatella sp. also increase in number, and specimens of the mussel Geukensia sp. and Ostrea meridionalis are abundant and pos- sibly indicate a change to conditions of lowered salinity. This assemblage represents the final stage in the development ofthe "Upper Shell" unit. The overlying "Shell Hash" unit contains abundant specimens of Corbicula densata, some of which are articulated. Corbicula has wide distri- bution in fresh- to brackish-water deposits of Cenozoic age (Gibson, 1967:646). The presence of this bivalve, the absence of planktonic Fora- minifera, and the increased influx of fine-grained sediment formed the basis for Gibson's conclusion that this unit (top of Gibson's unit 8) was formed in a bay or sound of lowered salinity, less than 15 meters deep. We think that the "Shell Hash" unit represents a reworked deposit with the bulk ofthe shell derived from the underlying "Upper Shell" unit. If this reworking occurred in an estuarine tidal channel, as suggested earlier, the presence of species favoring lowered salinity conditions, such as C. densata, mixed with fully marine species would be expected. Formation of the Shell Bed A combination of several factors indicates to us that the zones of the "Upper Shell" unit were 226 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY formed by localized catastrophic events that pro- duced mass mortality. These following factors are included. (1) The excellent preservational condi- tion ofthe fossils, being largely unworn and some with traces of original coloration. Many bivalve and gastropod shells are complete, with the bi- valves frequently articulated. The presence of delicate, thin-shelled clams such as Anomia simplex and articulated Argopecten eboreus, and echinoid plates, which are especially vulnerable to me- chanical destruction, indicates little or no trans- portation and/or exposure to wave action follow- ing death. (2) A broad range of size classes is present for the abundant species in our collec- tions. Figure 5 is a size-frequency plot for Glycy- meris americana that is typical of those plotted for this species and several other abundant bivalves {Astarte concentrica, Venericardia granulata). The pres- ence of many specimens of different sizes suggests that size-selective winnowing processes were not operative following death of the clams. (3) The majority of shells are generally free of encrusting organisms, such as bryozoans, barnacles, and worm tubes, and do not show the strong effects of marine borers, which normally attack molluscan shells exposed on the substrate following death of the animal. Mass mortality of marine invertebrates may result from a variety of causes, including vulcan- ism, rapid temperature and/or salinity changes, toxic water conditions, and the action of severe S 35 • E 30 o %. 25 N=143 ^ 20 o 9 ^h^jlT^^ n E 10 3 4 7 8 9 10 11cm FIGURE 5.—Size-frequency distribution of specimens oi Gly- cymeris americana from zone 2 of trench 2. Size measurement is of valve height; right and left valves are included with only one measurement from articulated specimens. Storms (Brongersma-Sanders, 1957). As indicated by Brongersma-Sanders (1957:942), catastrophes capable of killing benthic organisms may occur repeatedly in a given area. In the fossil record the result of repeated kills would be a sequence of fossiiiferous strata, probably much like the se- quence of beds of the "Upper Shell" unit. Recently Kranz (1974) has shown that rapid burial ("anastrophic burial") can be lethal to marine bivalves, leading Kranz to conclude that catastrophic burial events probably have been much more important in the formation of ancient shell deposits than previously realized. Severe winter storms and hurricanes occur frequently off the North Carolina coast today, and undoubtedly they also occurred during the time of formation of the "Upper Shell" unit. Such storms can be highly disruptive to shallow-bottom environ- ments and are known to be causes of mass mor- tality (Brongersma-Sanders, 1957). The "Upper Shell" unit, characterized by its abundance of shell material with a minimal amount of sediment matrix, seems indicative of conditions of forma- tion other than by gradual accumulation of shells in a quiet, subtidal environment. Gernant (1970) described zones of well-pre- served, highly concentrated fossils from the Chop- tank Formation (Miocene) of Maryland. He de- veloped an explanation for these shell concentra- tions based on a swell-traction mechanism first suggested by Powers and Kinsman (1953) to ex- plain the formation of concentrated shell layers found off the mouth of Chesapeake Bay today at depths of 15 to 45 meters. Powers and Kinsman postulated that large swells produced by offshore storms establish a pressure gradient capable of disrupting the substrate to the extent that epi- faunal and shallow- to medium-depth infaunal bivalves would be exposed, relocated on their sides, and smothered by rapid sedimentation fol- lowing passage of the storm. Experiments by Kranz (1974:260-261) on living bivalves indicate that epifaunal suspension feeders (such as, Glycy- meris and Argopecten) and bivalves that cement to the substrate (such as Ostrea and Anomia) are particularly vulnerable to rapid burial and are generally unable to escape more than 1 cm of NUMBER 53 227 burial. Shallow-burrowing, siphonate suspension feeders (such as Mercenaria) are more adept at escape, being able to cope with and escape from 10-50 cm of rapidly deposited sediment. The presence of the abundant flat-lying, often articulated valves of Mercenaria mercenaria that characterize zone 1 of the "Upper Shell" unit suggests that these clams were killed by a local- ized catastrophic event, possibly by relocation and rapid burial similar to the mechanism pro- posed by Powers and Kinsman (1953). As sug- gested earlier, the formation of a hard shell pave- ment probably inhibited the recruitment of ju- veniles of Mercenaria and favored the resultant resettlement ofthe area by epifaunal assemblages dominated by Glycymeris and Argopecten. Bivalves of the epifaunal assemblages of zone 2 would have been highly vulnerable to rapid burial. Brenner and Davies (1973) described lay- ers of whole shells ofthe pecten-like bivalve Camp- tonectes from the Jurassic of Wyoming and Mon- tana. These layers are similar to the concentrated Argopecten eboreus layers within zone 2 (Figure 2D) of the "Upper Shell" unit. Brenner and Davies (1973:1694, fig. 12) suggest that "swell lag" de- posits such as these were formed by the Powers- Kinsman mechanism. Kranz (1974:263) noted that fossil assemblages can be biased by catastrophic burial. Thus, it may be possible to produce different fossil assem- blages from a single life assemblage. A reasonable example might be the subassemblages of zone 2 ofthe "Upper Shell" unit dominated by Glycymeris americana and Argopecten eboreus. The catastrophic burial process may have resulted in the segrega- tion ofthe epifaunal assemblage into the subzones dominated by Glycymeris and by Argopecten and Anomia (Figures 3, 4). This alternation of layers within zone 2 could well result from multiple anastrophic burial events capable of segregating the epifaunal bivalves of the "Upper Shell" unit. Zone 3, as discussed earlier, represents a change in environmental conditions suitable for the re- turn of Mercenaria mercenaria. However, the con- centration of shells and the presence of many articulated valves of Mercenaria, Guekensia, and Ostrea suggest that localized catastrophic kill events continued to be important in the formation of this zone. Conclusions The "Upper Shell" unit at the Lee Creek Mine can be subdivided into three zones based on distinctive bivalve assemblages. Zone 1 is char- acterized by an infaunal bivalve assemblage dom- inated by Mercenaria mercenaria. Zone 2 consists of an epifaunal/semi-infaunal assemblage charac- terized by Glycymeris americana and is cut repeat- edly by thin, concentrated shell layers oi Argopec- ten eboreus and Anomia simplex. A reappearance of M. mercenaria in the sequence along with the occurrence of Geukensia sp. and an increase of shells of the oyster Ostrea meridionalis define zone 3. The disappearance of M. mercenaria from the sequence probably resulted from the formation of a shell pavement, which prevented successful re- colonization of infaunal bivalves and facilitated the establishment of an epifaunal bivalve assem- blage. Later, environmental conditions changed sufficiently to permit the return of M. mercenaria. The bivalve assemblage zones can be recognized in the exposures of the mine area, but further regional work will be necessary before the full geographic extent ofthe zones can be established. The preservation of fossils in excellent condi- tion, the large number of articulated bivalves in a broad range of size classes, and the close packing of shells with generally little matrix suggest that zones of the "Upper Shell" unit were formed by localized catastrophic events causing mass mor- tality ofthe benthic communities. This mass mor- tality may have resulted from localized cata- strophic burial events due to the action of severe winter storms or hurricanes, with resultant large swells causing disruption of the substrate and subsequent burial of the bottom dwellers. A mechanism of this type has been described by Gernant (1970). However, the possibility that sudden changes in other environmental condi- tions were responsible for mass mortality of the benthic fauna cannot be ruled out. The "Shell Hash" unit is composed primarily of fragmented and worn shell material reworked 228 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY from the underlying "Upper Shell" unit. The presence of the bivalve Corbicula densata, many specimens of which are articulated, suggests that this unit was deposited under conditions of low- ered salinity; possibly in an estuarine tidal chan- nel. Analogous reworked shell accumulations have been reported from the modern Georgia estuarine system (Howard and Frey, 1973). The timing of this episode of reworking cannot be established with certainty; it may represent a regression event following deposition of the "Upper Shell" unit, or it may be associated with the deposition of the overlying Pleistocene units as postulated by Belt, Frey, and Welch (this volume). Literature Cited Blackwelder, B.W., and L.W. Ward In prep. Late Pliocene and Early Pleistocene Mollusca from the James City and Chowan River Forma- tions at the Lee Creek Mine. Brenner, R.L., and D.K. Davies 1973. Storm-generated Coquinoid Sandstone: Genesis of High-Energy Marine Sediments from the Upper Jurassic of Wyoming and Montana. Geological So- ciety of America Bulletin, 84(5): 1685-1697. Brongersma-Sanders, M. 1957. Mass Mortality in the Sea. Geological Society of America Memoir, 67(1):941-1010. DuBar, JR., J.R. Solliday, and J.F. Howard 1974. Stratigraphy and Morphology of Neogene Depos- its, Neuse River Estuary, North Carolina. In R.Q. Oaks, Jr., and J.R. DuBar, editors, Post-Miocene Stratigraphy, Central and Southern Atlantic Coastal Plain, pages 102-122. Logan: Utah State Univer- sity Press. Gernant, R.E. 1970. Paleoecology of the Choptank Formation (Mio- cene) of Maryland and Virginia. Maryland Geolog- ical Survey Report of Investigations, 12: 90 pages. Gibson, T.G. 1967. Stratigraphy and Paleoenvironment of the Phos- phatic Miocene Strata of North Carolina. Geolog- ical Society of America Bulletin, 78(5) :631-650. Howard, J.D., and R.W. Frey 1973. Characteristic Physical and Biogenic Sedimentary Structures in Georgia Estuaries. American Associa- tion of Petroleum Geologists Bulletin, 57(7): 1169-1184. Kranz, P.M. 1974. Anastrophic Burial of Bivalves and its Paleoeco- logical Significance, yourna/ of Geology, 82(2):237- 265. McKenna, M.C. 1965. Collecting Microvertebrate Fossils by Washing and Screening. In Bernhard Kummel, and David Raup, editors, Handbook of Paleontological Techniques, pages 193-203. San Francisco: W. H. Freeman and Co. Powers, M.C, and B. Kinsman 1953. Shell Accumulations in Underwater Sediments and Their Relation to the Thickness of the Traction Zone. Journal of Sedimentary Petrology, 23(4):229-234. Stanley, S.M. 1970. Relation of Shell Form to the Life Habits of the Bivalvia (Mollusca). Geological Society of America Memoir, 125: 296 pages. Thomas, R.D.K. 1975. Functional Morphology, Ecology, and Evolution- ary Conservatism in the Glycymerididae (Bival- via). Palaeontology, 18(2) :217-254. Wiedemann, H.U. 1972. Shell Deposits and Shell Preservation in Quater- nary and Tertiary Estuarine Sediments in Geor- gia, U.S.A. Sedimentary Geology, 41(2): 103-125. Pleistocene Coastal Marine and Estuarine Sequences, Lee Creek Mine Edward S. Belt, Robert W. Frey, and John S. Welch ABSTRACT Pleistocene and uppermost Tertiary sediments in the Lee Creek Mine exhibit remarkable asso- ciations of physical and biogenic sedimentary structures. These, and close modern analogs, per- mit detailed interpretations of depositional con- ditions. Environments recognized include: supra- tidal, intertidal, subtidal estuarine, fluvial chan- nel, freshwater swamp, ebb-tidal sand bar, and shallow nearshore shelf or open estuary deposits. Four sequences of deposition, the uppermost of Sangamon age, were found. These sequences, separated by unconformities, are related to sea- level change by means of transgressive and re- gressive sedimentary phases. Analysis of the pro- portion of the respective phases within cycles suggests that the region underwent only moderate tectonic subsidence during the Pleistocene. Introduction Excellent exposures of a wide variety of Pleis- tocene and older sediments were created by min- ing operations in the Texasgulf phosphate mine near Aurora, North Carolina (Figure 1). The upper 50 feet (15 m) of section in the pit were studied by John Welch, Edward Belt, Allen Cur- ran, James Austin, and Robert Frey during the summers of 1971 and 1972, and were rechecked Edward S. Belt, Department of Geology, Amherst College, Amherst, Massachusetts 01002. Robert W. Frey, Department of Geology, University of Georgia, Athens, Georgia 30602. John S. Welch, 621 Jersey Avenue, Winston-Salem, North Carolina 27101. by Curran in 1973. Preliminary results were soon announced (Welch, Frey, and Belt, 1972; Curran and Frey, 1972; Curran, lannicelli, and Frey, 1973). In the present report, essentially completed in 1974, we emphasize sedimentologic aspects of the various facies studied and define cycles of sedimentation that we believe will be of use to students ofthe Pleistocene in North Carolina and elsewhere. Originally, we expected this volume to appear prior to publication of ichnologic and paleoecologie aspects of the project, emphasized by Curran and Frey. However, these faunal stud- ies, together with analyses of problematical sedi- mentary structures, have now been published (Bromley et al., 1975; Curran, 1976; Curran and Frey, 1977), and thus form corroborative evidence for interpretation of the depositional environ- ments described herein. Unfortunately, the sections upon which this report was based were long ago removed by min- ing operations, and the mined-out area has been reclaimed; of those outcrops, all that remains are numerous photographs, detailed logs, and a few sand samples, reposited at Amherst College. However, the basic stratigraphic section as estab- lished herein was still recognized in the mine in 1978. Reconnaissance of the surrounding region by Belt and Austin in 1972 (Austin, Molnia, and Curran, 1973) failed to reveal comparable facies elsewhere. Perhaps exposures in the Lee Creek Mine were unique remnants of Pleistocene facies not found elsewhere; abrupt lateral facies changes are characteristic of many modern estuarine se- 229 230 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 0 10 50 Miles 60 Ki lometers • LEE CREEK MINE PR.- PAMLICO RIVER FIGURE 1.—Location of Lee Creek Mine and its relationship to Suffolk Scarp, Cozistal Plain of eastern North Carolina. quences (Howard and Frey, 1980b). On the other hand, perhaps deep leaching of other exposures destroyed much of the evidence. Characteristic sediments in the Lee Creek Mine permitted re- markable preservation of physical and biogenic sedimentary structures, and we found little indi- cation of chemical leaching of heavy minerals (Scott, 1976); the opposite is true of many other Pleistocene deposits of the southeastern Atlantic Coastal Plain. The section consisted of sand, mud, peaty sand, and coquina facies containing distinctive physical and biogenic sedimentary structures, made all the more notable by their striking similarity to recent analogs. These facies, analyzed with respect to local depositional environments and related to intervening unconformities, suggest four cycles of transgression and regression in the section be- tween the Pliocene Yorktown Formation and Holocene soils. Essentially a three-dimensional correlation was possible within a rectangular area 3600 by 1500 feet (1200 by 500 m) in the mine. In addition to this report, various other aspects of Pleistocene facies and fossils at the Lee Creek locality are reported in this volume (Whitehead; Wheeler, Daniels, and Gamble; and Curran and Parker), as well as by Blackwelder and Ward (in prep.). ACKNOWLEDGMENTS.—This project was made possible through financial support from the Na- tional Science Foundation COSIP grant GY-7657 to undergraduate students at Williams, Amherst, Mt. Holyoke, and Smith colleges, Massachusetts. Students from Amherst and Smith wrote honors theses on topics related to Lee Creek Pleistocene deposits and fossils. John Welch did an honors thesis on physical stratigraphy and sedimentology of the section described herein, in 1971-1972. Belt was his advisor. Robert Frey was drawn into the research as a result of his visits with Belt and Welch in the field during 1971. Welch's work was followed in 1972-1973 by James Austin's honors thesis. He measured an additional 11 sections in the pit, because the face had by then moved several hundred yards (meters) to the west (Figure 2); his results confirmed those of Welch. We also acknowledge Austin's attempt to correlate the pit succession with sections along the south shore of the Pamlico River west ofthe mine (Figure 1). We are especially grateful to Allen Curran for introducing us to the Texasgulf phosphate mine, for making logistical and other arrangements in Morehead City, and for directing the overall field program during the two summers. We also ap- preciate the cooperation of Texasgulf Inc., in particular June Crawford, Jack Hird, and Jim MacDonald, for allowing us access to the mine and for several pertinent maps showing the posi- tion of trenches at the west end of the mine during the two years of study. H. Allen Curran, Roy L. Ingram, Jack W. Pierce, James D. Howard, Donald R. Whitehead, Jules R. Dubar, and Nicholas K. Coch read various versions of the manuscript and offered valuable suggestions. John E. Sanders suggested several references inadvertently omitted from an NUMBER 53 231 earlier draft. Jeremy Reiskind and Walter H. Wheeler made helpful suggestions in the field. Curran's students, Patricia L. Parker and Sandra K. lannicelli, supplied useful information on par- ticular stratigraphic horizons. Clayton E. Ray implemented C dating of our peat samples through the Smithsonian Institution. F.M. Hue- ber of that institution identified some wood frag- ments. Stratigraphy and Depositional Environments To our knowledge, no prior sedimentologic study of Pleistocene sediments from the Lee Creek Mine had been made. Previous workers in the general vicinity of the mine concentrated on thin roadside exposures, thicker and rather well-ex- posed sections along the Neuse and Pamlico River estuaries, or samples from auger holes as deep as 100 feet (30 m). Recent concepts and controver- sies concerning the Pleistocene of eastern North Carolina were outlined by Richards (1950, 1962, 1969), Fallaw and Wheeler (1969), Daniels et al. (1972), Fallaw (1973, 1975), and DuBar, Solliday, and Howard (1974). The regional correlation and geologic setting for strata reported herein are summarized by Wheeler, Daniels, and Gamble (this volume), to whom the reader is referred for regional perspective. For broader regional rela- tionships, see the summaries by Oaks and DuBar (1974), Blackwelder and Ward (1976), and DuBar and DuBar (1980). Our field methods consisted of carefully locat- ing the position and elevation of key horizons in sections, logging the strata in detail, noting tex- tures, primary physical and biogenic sedimentary structures, paleocurrent directions, body fossils, and defining lithogenetic units. Grain-size analy- ses were made on all units (Folk, 1974) in order to define the range of textures present. Detailed logs of measured sections, not reproduced here, may be found in theses by Welch (1972) and Austin (1973). Selected bedding or burrow fea- tures were sampled in the field by means of epoxy peels, and numerous photographs were taken in black and white, as well as in color. General details ofthe section emerged in 1971 (Welch, Frey, and Belt, 1972) by careful correla- tion of units throughout the network of trenches at the west end of the mine (Figure 2). A more detailed stratigraphic column (Figure 3) was con- structed from these data; subsequent interpreta- tions, and a schematic cross-section (Figure 4) led to our reconstruction of cycles of Pleistocene coastal marine sedimentation (Table 1). Grain- size analyses (Figure 8; cf Hails and Hoyt, 1969) proved to be much less valuable in facies inter- pretations than did unit geometry and physical and biogenic sedimentary structures. Because salinities are such integral parts of our interpretations of depositional environments, we offer the following qualifications of descriptive terms used subsequently: Full marine salinity is 36 %o; inshore and nearshore-shelf salinities rarely attain that level, however; thus we consider 32 to 36 %o as "normal" salinity, 28 to 32 %o as "nearly normal" salinity, and 24 to 28 %o as "slightly brackish" salinity. Many "normal" estuarine an- imals can tolerate salinities as low as 10 %o for short intervals of time. The four depositional sequences identified and described here may be characterized briefly as follows: Cycle I—Upper Shell unit: mostly molluscan shells (many articulated bivalves). Lower Pleis- tocene or uppermost Tertiary; overlies Pliocene Yorktown Formation. Cycle II—Two units. (A) Shell Hash unit: coarse quartz sand and broken shells in trough- cross-bedded sets, a "basal conglomeratic sand," overlain by, and in places laterally equivalent to, (B) Mud and Sand unit: sandy member contain- ing flasers, cross-beds, Ophiomorpha, and Skolithos, and muddy member containing bioturbate tex- tures and Thalassinoides. Cycle III—Current unit: lower member of tab- ular and trough cross-beds having east-northeast mean trend, containing Ophiomorpha and Skolithos. Overlain by flaser and wavy bedded mud and sand member containing above trace fossils, Plan- olites, "sitz marks," and several other traces. Cycle IV—Several units: progression from (A) 232 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TRENCH-1 °^-2 EXCAVATED BELOW SEALEVEL West Wall, July, 1971 TRENCH-Y —r ^ r-- \ 400METERS .^^ L EAST WALL 10 FIGURE 2.—West end of Lee Creek Mine, showing location of sections measured. The 1971 and 1972 positions of trenches are indicated. Open circles represent sections measured by Welch (1972); solid circles represent sections measured by Austin (1973); section 10, east wall, is off the scale ofthe map. Channel unit (fluvial channel): trough cross-bed- ded sand, cut into deposits of cycle III and II, through (B) Peat and Clay unit: clay member having no apparent fossils, into nonmarine peaty member containing a Sangamon temperate flora, ultimately into (C) Mottled unit (marine unit): intensely bloturbated mud and sand containing mollusks. Comments on Trace Fossils Biogenic sedimentary structures were invalua- ble in our environmental interpretations. Because of this, as well as their possible unfamiliarity to the reader and common misconceptions of trace fossils such as Ophiomorpha and Thalassinoides (Frey, 1975:33; Frey, Howard, and Pryor, 1978; Frey and Seilacher, 1980), a word about their usefulness and degree of reliability is in order. Primarily, we stress that our interpretations are based not merely upon identification of trace fossil genera, but rather, upon the closeness-of-fit between geometry, configuration, and composi- tion of trace fossils and close analogs known along the present coast of Georgia (Howard and Frey, 1975, 1980a, 1980c; Frey and Howard, 1980) and the Carolinas (Frey, 1970; Allen and Curran, 1974). In most cases, the fit was remarkably close (Curran and Frey, 1977), and we suspect that many of these extant tracemakers are in fact the same species that made the Pleistocene trace fossils. NUMBER 53 233 KEY Crab Burrows ' +5- 0- -5- -10 -15- -zo-" METERS FIGURE 3.—General post-Yorktown stratigraphic column, Lee Creek Mine. Heavy wavy lines represent major unconformities, MB. = member, u. = unit. W.B. = wavy bedded member of Current unit. Diagram emphasizes associations of important burrow types and marine shells. The ghost shrimp Callianassa major is not re- stricted to beaches or beach-related habitats and is not the only species that constructs Ophiomorpha- type burrows (Weimer and Hoyt, 1964); yet most Pleistocene Ophiomorpha observed at Lee Creek are virtually identical to modern burrows made by C. major on the southeastern U.S. coast (Frey, Howard, and Pryor, 1978). This shrimp inhabits low-intertidal to relatively shallow subtidal sands in waters of normal or near-normal salinity and intermediate to moderately high energy. Such conditions are found along beaches, the nearshore shelf, and in ocean-influenced tidal flats, point bars, and shoals within estuaries and lagoons. Callianassa biformis also constructs Ophiomorpha- type burrows, but these systems are smaller and more irregularly inclined, and the shrimp toler- ates a slightly greater range in salinity, bathy- metry, and current velocities. Therefore, it is found in somewhat muddier sediments than is C. major (Howard and Frey, 1975, 1980c, fig. 20), and also ranges into deeper and (or) less saline waters. Similarly, the trace fossil Thalassinoides can re- sult from burrowing by various shrimp-like ani- mals and may intergrade with Ophiomorpha (Bromley and Frey, 1974; Frey and Seilacher, 1980); but specimens studied by us are remark- ably similar to burrows constructed today by the mud shrimp Upogebia affinis (Frey and Howard, 234 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY SCHEMATIC CROSS SECTION OF UNITS=I97I SOUTH meters 400feet 2- feet r8 NORTH lOOmeters I- -0 TRENCH I i fi: Pamlico: CD O TRENCH U L TRENCH TT TRENCHm TRENCH Ur k 4 >^' ^ .i T ^—^ UNIT '' ^ ROOTY '' MM I, ^^^^y^^^^ MOTTL^ED_^^^^J_^^ ^ ® UPPER SHELL UNIT rr wavy_bed mb\\~ " PE*AT^a^^LAYf-<^'~»~-'"r>^ wavy - j,^ -^ ^g-;^ Q 0 C2 O ^''BOULDER BED" '-^ F'iGURE 4.—Stratigraphic units. Solid circles represent C samples (all exceeded 40,000 yr BP). Heavy lines represent unconformities; heavy line with slashes represents possible unconformity. # represents peaty sand. For burrow types, see Figure 3. 1975). On the Georgia coast, U. affinis burrows in muddier sediments than does Callianassa and pre- fers slightly lower salinities and less current en- ergy, although their bathymetric ranges are com- parable. These conditions are found commonly along salt marsh tidal stream or estuary banks and in muddy point bars and tidal flats, as well as in relatively shallow lagoonal or estuarine mud bottoms. Some Lee Creek specimens of Thalassinoides possibly were constructed by the snapping shrimp Alpheus heterochaelis, although the characteristic anastomosing-downward pattern of its burrows (Basan and Frey, 1977) was not observed. The environmental range of A. heterochaelis broadly overlaps that of Upogebia affinis (Howard and Frey, 1975). Another trace fossil that merits special atten- tion is Skolithos. Several worms living along the present southeastern Atlantic coast construct Sko- lithos-Vik^ dwelling tubes (e.g., Hertweck, 1972), and some minor variations were noted among our Pleistocene specimens. But most Skolithos ob- served by Curran and Frey (1977) were extremely similar to dwelling tubes of the polychaete Onu- phis (Howard and Frey, 1975, 1980a, 1980c). Environments inhabited by Onuphis are generally similar to those preferred by Callianassa biformis, overlapping those of C major; yet the polychaete avoids higher energy levels. It is found commonly in estuarine point bars, tidal flats, relatively shal- low channel bottoms, and low-energy beach and nearshore environments (Howard and Dorjes, 1972), but not in purer muds, such as those typically occupied by Upogebia. Occurrence of these trace fossils in the Lee Creek section is compatible with the kinds of sediments and sedimentary structures associated NUMBER 53 235 today with their extant counterparts (Howard and Frey, 1980c), yielding a powerful additional dimension to our overall study. Furthermore, bio- genic sedimentary structures, unlike tests and shells, are enhanced by diagenesis and ordinarily cannot be reworked or transported into other depositional environments (Frey, 1975; Frey and Seilacher, 1980); if reworked, they are easily rec- ognized as such (Figure 13). Descriptions of Stratigraphic Units YoRKTow^N FORMATION The Yorktown (probably lower or middle Pli- ocene; DuBar and DuBar, 1980) lies beneath the Upper Shell unit (lowest Pleistocene, or possibly upper Pliocene; Blackwelder and Ward, in prep.). Both the Yorktown and the Upper Shell unit, although not studied in detail, are included here for the sake of continuity. Average thickness of the Yorktown, as measured by us, is 45 feet (14 m); the upper 10 feet (3 m) or less is referred to locally as the "Boulder" bed. This bed, possibly a result of post-Yorktown weathering, together with striking faunal changes in the overlying Upper Shell unit, constitutes major evidence within the pit for an unconformity between the Yorktown and overlying units, although not rec- ognized by Gibson (1967). Evidence presented by Welby (1971) and Wheeler, Daniels, and Gamble (this volume) indicates that the upper part of the Yorktown is partially cemented by carbonate to form irregular concretionary masses up to 3 feet (90 cm) across, a feature well expressed through- out eastern North Carolina at this stratigraphic position. Apart from these concretionary masses, the Upper Shell unit typically overlies the York- town with sharp, although in some places grada- tional, contact. This contact apparently was planar where we observed it (i.e., no irregulari- ties) . The upper 10 feet (3 m) of the Yorktown consists of fine- to coarse-grained, muddy, sub- angular to subrounded quartz sand and glaucon- ite. The color of the unit ranges from medium and light gray to dark brown and orange, where stained with iron oxide. No shells or bones were found, but the entire sequence is intensely blotur- bated. The trace fossil Ophiomorpha nodosa is pres- ent, although rare; these small, somewhat irreg- ular specimens may correspond to burrows of Callianassa biformis (Hertweck, 1972; Howard and Dorjes, 1972; Howard and Frey, 1975; Frey, Howard, and Pryor, 1978). Bedding in the unit is obscured by bloturbation and selective lithifica- tion. The environment of deposition of the upper Yorktown presumably was one of nearshore, shal- low marine water, although the quantity of mud matrix suggests a moderate rather than a high- energy regime. A more refined analysis depends on studies of Yorktown facies elsewhere, and their environmental interpretation. UPPER SHELL UNIT The Upper Shell unit is now generally regarded as post-Yorktown, although authors differ as to formational assignment (Gibson; Hazel; Curran and Parker, all this volume; DuBar, Solliday, and Howard, 1974:109; Bailey, 1977; and Black- welder and Ward, in prep.). The unit is charac- terized by large molluscan shells. Faunal lists (Parker, 1972; Blackwelder and Ward, in prep.) consist of 95 species of bivalves, 97 species of gastropods, one species of scaphopod, 3 species of corals, an undetermined number of bryozoans, echinoids, barnacles, crabs, and a hydrozoan. The matrix around shells, many of which are tightly packed, is fine- to medium-grained quartz sand, generally containing minor amounts of silt and clay; the percentage of silt and clay increases greatly in some places. The unit ranges from 2 to 10 feet (0.6 to 3 m) in thickness, although in one area it was entirely removed by erosion prior to deposition of the overlying unit (Figure 4, trench IV). Upper and lower contacts appear to be unconformable, more pronounced relief being visible on the upper con- tact; the Shell Hash unit overlies the Upper Shell unit with obvious relief and a sharp contact in most places. Distinct horizontal bedding (in most places), 236 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY densely packed shells, many of them articulated, and successive changes in faunal assemblages from the lower to the upper part of the unit (Curran and Parker, this volume) suggest re- peated catastrophic "kills" and normal salinity at the base, shifting to lowered salinity at the top. Sandy clay plugs in the outermost whorl of large Busycon snail shells commonly entrap clean- washed shell hash and quartz sand within inner whorls. This sediment contrast suggests that either (1) such shells were transported from a high-energy environment containing little mud into one in which mud was a common constitu- ent, or (2) by a change in depositional regimes, mud was brought into the environment and set- tled between large shells. The high percentage of unabraded, articulated bivalves (Curran and Par- ker, this volume) indicates general absence of shell transport. Thus, periodic influx of mud, possibly delivered by riverine estuaries following torrential rains in the hinterland, together with correspondingly abrupt decreases in salinity, may account for the catastrophic death assemblages. Variations in wave energy, such as those induced by small-scale storms, perhaps enhanced horizon- tal bedding and shell concentrations. Sediments through the entire thickness of the Upper Shell unit along the northern edge of the pit exposure are cross-bedded. Here, shells from lower and upper assemblages are mixed thor- oughly. The cross-beds include both tabular (up to 6 ft, 1.8 m, thick) and trough sets (less than 2 ft, 60 cm, thick). We interpret this occurrence as local reworking and redeposition of sediments of the entire unit, possibly by means of an estuarine channel or estuary-entrance scour hole (Frey, Voorhies, and Howard, 1975; Howard and Frey, 1980c, fig. 14). The reworking may be related to erosion just prior to deposition of sediments ofthe Shell Hash unit, in which case the cross-bedded debris would probably be placed in the Shell Hash unit. The change in faunal assemblages from the base to the top of the unit suggests a change in salinity and/or a change to predominantly inter- tidal exposure; lower zones contain a high-diver- sity fauna of mollusks, corals, echinoids, etc.. whereas subsequent assemblages grade into an upper zone characterized by Geukensia species and Ostrea meridionalis. Geukensia is indicative of slightly lower salinities and shallow subtidal to intertidal environments (Basan and Frey, 1977:55, 64, 65); the same seems to be true of 0. meridionalis (Cur- ran and Parker, this volume). Thus, the Upper Shell unit probably shows normal salinities in its lower part and somewhat decreased salinities and/or increased subaerial exposure in its upper part. Postulated periodic "kills" by lowered salin- ities and mud suggest proximity to a terrigenous sediment supply. These combinations of effects suggest an open estuary mouth or adjacent near- shore-shelf environment for the base of the unit, with perhaps moderate water depths (cf. Frey and Pinet, 1978), grading upward to shallower, less saline waters of the middle reaches of an estuary. Ultimately, the sea regressed and deposits of the Upper Shell unit were channeled by ero- sion. SHELL HASH UNIT The Shell Hash unit consists of poorly sorted, coarse- to fine-grained quartz sand, broken and current-worn invertebrate (mostly molluscan) shells of diverse sizes, mud clasts, and woody debris. It varies in thickness from several inches (centimeters) to 4 feet (1.2 m). The shells (only a few of which are whole, single valves) were de- rived from the underlying Upper Shell unit when, following erosion of the latter, the sea again en- tered the area. Shelly debris generally forms less than 50 percent ofthe deposit. The Shell Hash unit commonly is trough cross- bedded, although gently inclined bedding also is found (Figure 5). Trough sets generally are less than 1 foot (30 cm) thick. Skolithos linearis is abundant locally; Ophiomorpha nodosa is present, although not common. Sediment sorting (Folk, 1974:3-7, 15-48; definition and methodology) in bulk samples is poor (1.25 to 1.86 (j)); mean grain sizes range from coarse to fine sand (0.90 to 2.15 (^). If the quartz matrix alone is sampled (neglecting shells and mud clasts), sorting is very good to moderate. NUMBER 53 237 This unit overlies the irregular surface of the Upper Shell unit and interfingers laterally with, and grades vertically into, the Mud and Sand unit. Molds of broken-shell debris in the base of the latter (shells are leached out in more acidic mud) apparently support this conclusion, al- though proper taxonomic designations were not established. For these reasons, the Shell Hash unit tentatively is considered to be related genetically to the Mud and Sand unit (see Figure 4 for lateral relationships). A high-energy depositional environment is in- dicated by ripped-up mud clasts, trough cross- beds, moderate sorting, and the common occur- FiGURE 5.—Epoxy relief peel of cross-beds and ripple laminae in Shell Hash unit. Mud chips occur in lower half, where shell debris is abundant. Trench II, section 1 of Welch (1972). Scale in millimeters. 238 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY rence of coarse-sand and fine-pebble grain sizes together with the worn and broken molluscan shells. Such facies are developed today within and adjacent to estuarine channels and estuary- inlet shoals on the nearshore shelf of Georgia (Oertel, 1973; Howard and Frey, 1980c, figs. 10, 13, 14), and can include tubes of Onuphis micro- cephala (Skolithos analog) and burrows of Calli- anassa major (Ophiomorpha analog). Faunal evidence for the depositional environ- ment of the Shell Hash unit consists of what initially may appear to be an anomalous associ- ation of well-preserved valves of Corbicula densata, a brackish water bivalve, with Skolithos linearis and Ophiomorpha nodosa, trace fossils most often associated with conditions of near-normal to nor- mal salinity. However, species of the genus Cor- bicula are noted for their ability to tolerate a range of lowered salinity to freshwater conditions and apparently can exist for extended periods under either brackish or freshwater conditions (Evans et al., 1979:194, 212). Howard and Frey (1975:53, 55) have shown that the organisms that form S. linearis and 0. nodosa burrows (especially Calli- anassa biformis) commonly inhabit the shallower, sandy parts of Georgia estuaries and can tolerate salinities as low as 10 %o, at least for short periods. Thus, this faunal association of the Shell Hash unit, rather than being anomalous, seems to be indicative of estuarine conditions during deposi- tion. MUD AND SAND UNIT The Mud and Sand unit in places contains diverse sediments, including boulder-sized rock clasts, but most lithologies fall within two distinc- tive members: (1) a dark gray to dark greenish gray, sandy clay member, and (2) a clay-mottled, tan sand member. The two members interfinger irregularly, some sections consisting entirely of one or the other member and other sections con- taining various combinations of the two. Details of the complex interrelationships between mem- bers have not been determined, although these would have considerable environmental signifi- cance. In general, the sand member has diffuse boundaries but forms an irregular tongue within the mud member (Figure 4). Thickness of the entire Mud and Sand unit ranges from 1 to 16 feet (0.3 to 5 m), due mainly to the irregular topography upon which it and sediments of the Shell Hash unit were deposited; another factor is higher elevation, with respect to present sea level, of the upper contact on the north side (trenches I, II, Figure 4) than on the south side of the pit (trenches III, IV, V). This difference (approximately 5 ft; 1.5 m) is believed to have resulted from erosion occurring after dep- osition of sediments of the Mud and Sand unit but prior to deposition of sediments in the over- lying Current unit. In the region where the Mud and Sand unit has been eroded most deeply (trenches IV, V, Figure 4), pebbles, cobbles, and boulders are most abundant and lie on top ofthe entire unit. These clasts are rounded and show a variety of shapes, from equant to discoid and rod-like. Almost all clasts have a gneissic and quartzitic composition; one gneissic boulder measured more than 1 foot (30 cm) in its longest dimension, and weighed more than 50 pounds (22.5 kg). The nearest outcrop of these lithologies is in the Piedmont, 100 mi (160 km) to the west. None of these clasts was seen to penetrate into the top of the Mud and Sand unit; hence, they are considered to have lain on a hardened mud surface after the mud was compacted and, in our opinion, after erosion of that unit. Only a few small pebbles were found on top of the Mud and Sand unit outside the region of maximum erosion (trenches I, II, Figure 4); we conclude that maximum scour, maximum clast size, and highest concentration of clasts oc- cur in the same area. In addition to clasts, Black- welder and Ward (in prep.) reported rooted tree stumps at the top of what we consider to be the Mud and Sand unit. The clay-mottled sand member of the Mud and Sand unit forms an irregular tongue. Clay mottling results from intense bloturbation of mud-draped ripples and cross-beds. Thorough bloturbation of interbedded sand and mud layers, or of wavy, lenticular, and flaser bedding, in fact commonly produces a gross mixture of biotur- NUMBER 53 239 bated muddy sand (Howard and Frey, 1980a: 174, 1980c: 121). Clay-lined Ophiomorpha nodosa is abun- dant (Figure 6); Skolithos linearis is less so. Other bloturbation structures are common but are less distinctive (Curran and Frey, 1977). "Sitz marks" (possibly bivalve and anemone resting and escape traces; Frey and Howard, 1972) and physical collapse structures (Frey, Howard, and Pryor, figs. 7e, 9) also are common. The top of the unit is everywhere the sandy mud member, and in trench II (Figure 4) it is riddled with Skolithos linearis, possibly indicating a fairly coherent sub- strate (cf. Frey and Howard, 1969, pi. 2: fig. 2; Howard, Frey, and Reineck, 1973:40) related to conditions of the overlying Current unit (Figure 7). The clay-mottled sand member of the Mud and Sand unit contains ripples and cross-beds in tan quartz sand, and these are commonly draped with dark gray mud. Mud-draped ripples, termed "flasers" (Reineck and Wunderlich, 1968) are considered by many authors to be strictly inter- tidal in origin (Klein, 1971, table 1), although they are abundant subtidally within estuaries, the nearshore shelf, and inlet-shoal complexes in Georgia waters (Oertel, 1973; Howard and Frey, 1973, 1975, 1980a, 1980c). Flasers suggest tidal oscillations in energy of transportation and dep- osition of sediments, whether in intertidal or sub- tidal environments. In addition, many clay wisps and flasers are the result of fecal-pellet deposition, the pellets behaving hydrodynamically more as FIGURE 6.—Clay-mottled sand member of Mud and Sand unit. Rippled sand in both wavy bed and flaser structure is modified by bioturbate textures and post-depositional fault. Arrows indicate Ophiomorpha nodosa, some sections being diagonal to face of exposure. Trench III, 1971. No faults observed in beds younger than Current unit. 240 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 7.—Contact of Mud and Sand unit (lower one-third) with Current unit (upper two-thirds). Vertical tubes {Skoli- thos linearis) just below pen penetrate top of Mud and Sand unit and also occur in lower part of Current unit, here containing scattered clay wisps. Trench \\, near section 1 of Welch (1972). sand than as discrete or flocculated particles of clay during transport and deposition; thus, con- trary to previous views of clayey sediments as the result of deposition of suspended sediments under quiet-water conditions, fecal mud is easily accu- mulated under conditions of moderate, persistent wave and current energy (Pryor, 1975). When pellets are compacted after burial, they become indistinguishable from other muds deposited physically or electrolytically in shallow nearshore or back-barrier environments (Frey and Basan, 1978:139-141). Linsen beds (sand ripples, one ripple-layer thick, surrounded by mud) also are reported from tidal flats (Reineck and Wunder- lich, 1968) and subtidally from Georgia estuaries (Oertel, 1973; Howard and Frey, 1973, 1975, 1980a, 1980c). Size analyses of seven samples from the clay- mottled sand member indicate that it generally is poorly sorted, fine- to medium-grained sand, although moderately well-sorted coarse sand also was found (Figure 8). Mean grain sizes of the dominant group ranged from 1.75 to 2.70 ^ and the standard deviation from 0.98 to 1.41 ^. The sandy mud member consists of very stiff, dark gray to dark greenish gray clay having thin coarse sand layers, ripple (linsen) sand layers, and small isolated pebbles. The clay in places includes lam- inae of sand less than 0.06 inches (2 mm) thick (Figure 9) that alternate with thicker mud layers. In general, however, the clay has an unstructured appearance which, upon closer inspection, shows profuse bioturbate textures. The most common SORTING very well well moderately well moderate o / O o .20- .40- .SO- SO ■'i I 00-- 1.20- 1.40- 1.60- I 80 2.00— very fine sand .50 1.00 150 2.00 2.50 3.00 3,50 MEAN0 FIGURE 8.—Grain-size analysis of sediments. (35 samples taken in 1971 and 1972 from sections measured.) Black square is tabular cross-bed member of Current unit; circle is Channel unit; cross is Mottled unit; star is sand member of Mud and Sand unit. Except for sand member of Mud and Sand unit, all sediments lie within compact area (graphical mean (Mz) in <|> units plotted against graphical standard deviation (ai); Folk, 1974). NUMBER 53 241 FIGURE 9.—Laminated upper part of Mud and Sand unit. South wall, Trench II, 1971. Machete blade approximately 1.2 feet (35 cm) long. (Photo by James Austin.) burrow structure is Thalassinoides species; it con- sists of a clay-lined dichotomous tunnel that be- came filled with clay-rich to clean-washed sand and organic matter once the burrow was aban- doned. Other types of biogenic structures (Curran and Frey, 1977:151) resemble bioturbate textures produced by capitellid polychaetes in Georgia estuaries (Howard and Frey, 1973:1182; 1975:51- 62; 1980a, fig. 6.11). Physical and biogenic sedimentary structures ofthe Mud and Sand unit suggest complex inter- fingering of a normal or nearly normal salinity, relatively high-energy depositional environment (clay-mottled sand member containing Ophiomor- pha nodosa and Skolithos linearis) with a fairly low- energy, perhaps less saline, muddy environment (sandy mud member having Thalassinoides spe- cies). We thus consider the sand member to rep- resent a subtidal channel or muddy point bar deposit in an estuary that had free access to seawater of normal salinity, and the mud member to represent a subtidal and/or tidal flat environ- ment within an estuary possibly somewhat farther removed from constant access to normal salinities. The entire Mud and Sand unit might represent part of a subtidal estuary complex having more freshwater influences near the margins, the deeper channel carrying greater quantities of sand and possessing more nearly normal salinities (Hayes, 1975:11; Meade, 1972). Equally plausi- ble, however, is a salt marsh estuary (without an active river at its head) such as Doboy Sound, Georgia (Mayou and Howard, 1975:212); there, mud is abundant in the deeper part ofthe channel but becomes decreasingly important toward the shallower lateral margins ofthe estuary (Howard 242 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY and Frey, 1980a). Salinities are comparable throughout, but differences in energy levels and substrate characteristics greatly modify distribu- tions of infaunal organisms. Ideally, estuaries hav- ing active rivers at their head would accumulate much more sand; in most Georgia estuaries, how- ever, the facies pattern of riverine and salt marsh estuaries is so modified by local sand sources that the two might not be distinguishable in the rock record (Howard and Frey, 1980c: 103). Detailed interpretation of these, and the two members of the Mud and Sand unit, requires excellent stra- tigraphic control and thorough documentation of lateral relationships among sediment bodies. Upright tree stumps at the top ofthe Mud and Sand unit indicate a shoaling ofthe sea just prior to development of an erosional surface that in- cluded a channel cut into the top of the unit. A log recovered from the base of the unit in trench IV (Welch, 1972:2, appendix), contained the hole of a boring bivalve (possibly a pholad), and yielded a ^"^C date (Teledyne Isotopes, Inc., no. 1- 6621) greater than 40,000 years BP. CURRENT UNIT The Current unit is divided into two members: (1) the tabular cross-bed member (lower), and (2) the wavy bedded member (upper). Total thick- ness of the unit ranges from 4 feet (1.2 m) in trench I (Figure 4) to 15 feet (4.7 m) in trench V. It rests unconformably on the Mud and Sand unit, and fills low places of irregular relief devel- oped on it. The upper contact, as near as we can measure it with a transit, is nearly horizontal and almost at present mean sea level. At one place or another, the Current unit is overlain unconformably by a variety of units. In trenches IV and III, it is almost completely re- moved by erosion and is overlain by the Channel unit (Figure 4). In trenches V, II, and I, it is overlain either by discontinuous peaty lenses of the Peat and Clay unit, or by the Mottled unit. This unconformity shows more relief than any other observed in the section, although its relief does not necessarily mean that it is correspond- ingly more profound; the time interval repre- sented by any one ofthe unconformities described herein remains unknown (Table 1). The tabular cross-bed member of the Current unit is the thicker member, ranging from 3 to 10 feet (0.9 to 3 m). It consists of tan, subangular to subrounded, fine- to medium-grained (1.25 to 2.31 <», well sorted (0.30 to 0.65 4>) quartz sand. Minor amounts of feldspar, mica, and heavy minerals are found. All samples plot in a surpris- ingly close cluster (Figure 8), a distinguishing characteristic. The member contains well-developed tabular (planar) cross-beds (Figure 10), in sets that range in thickness from 1 to 5 feet (0.3 to 1.5 m), in exposures on the west end of the pit. Each set, as well as each lamina of foresets, typically is sepa- rated or delineated by heavy minerals or by light brown silty sand, some by muddy sand. The sets appear to be graded. This feature was noted by Terwindt (1971:517), Greer (1975:117), Howard and Frey (1975:41, 1980a: 162, 1980c), Frey and Howard (1980:192-193), and Visser (1980) among modern subtidal and intertidal estuarine sands deposited by ebb-dominant currents that produced tabular to trough cross-bedded sets. Flood-oriented cross-beds are rare along the Geor- gia coast; they are not so rare within relict flood tidal deltas (Belt, 1970:15; Boothroyd, 1978:329- 336). Although some authors consider unidirec- tional features unlikely (Klein, 1971) or at least remarkable in a bidirectional tidal system (Ter- windt, 1971:518), ebb-current dominance in es- tuaries is typical of the southeastern U.S. coast. Here, pronounced inequities of tidal flow result from frictional drag by marsh grasses during early ebb, followed by dramatically accelerated ebb flow down the estuarine water slope once the marshes have been drained (Howard and Frey, 1980a: 156-161, 1980c:93). Such retardation, then acceleration along a steep pressure of hydraulic gradient, does not occur during tidal flood. Tabular cross-beds in the western exposure of the Current unit present a remarkably uniform dispersion of paleocurrent directions (Figure 11); a mode trending to the east-northeast is consistent with an ebb orientation. A few trough cross-beds and current-rippled laminae are found in the NUMBER 53 243 FIGURE 10.—Tabular cross-bed member of Current unit. Cross-bed sets trend northeast, some separated by ripple-laminated sand. Arrows indicate Ophiomorpha nodosa, abundant in upper part of unit. Faint vertical lines near field notebook are clay-filled Skolithos linearis. U- and V- shaped irregularities in certain laminae are ''sitz marks." Heavy minerals and organic debris or silty clay distinguish bedding and laminae. Trench V, south wall, 1971. exposures on the west end of the pit, but the entire Current unit is trough cross-bedded on the east end of the pit (Figure 14). This feature, tabular cross-beds having silt layers at the top of each lamina, grading into trough cross-beds down an estuary, was observed in Lower Cretaceous estuarine sands by Campbell and Oaks (1973) and in modern estuaries (Terwindt, 1971; Greer, 1975; Howard and Frey 1975, 1980a, 1980c; Frey and Howard, 1980; Visser, 1980). The most spectacular sedimentary structure in the tabular cross-bed member consists of V- to U- shaped nested laminae in a cylindrical column up to 10 inches (24 cm) in diameter; the largest one extended through the entire thickness of the Current unit (10 ft or 3 m), in a near-vertical position, penetrating the top of the Mud and Sand unit. The structure apparently was not affected either by sand avalanches of tabular cross-beds or by the change in depositional envi- ronment from the tabular cross-bed member to the wavy bed member of the Current unit. Its origin remains problematical (Bromley et al., 1975:366-369; Curran and Frey, 1977:142). Biogenic sedimentary structures are common in the tabular cross-bed member (although they are even more abundant in the overlying wavy bedded member). The three most common le- bensspuren are Ophiomorpha nodosa, Skolithos li- nearis, and "sitz marks" (Figure 10). However, some of these deep V- to U-shaped laminated structures (Figure 12) are the result of collapse of 244 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY overlying sand into open burrows {Ophiomorpha), as described by Frey, Howard, and Pryor (1978). The smaller nested structures are interpreted as escape burrows of animals having anemone-like behavior (Curran and Frey, 1977:142). Another structure, a sediment-filled depression about 2 feet (60 cm) in width, is most likely a feeding trace of a large sting ray (cf. Frey and Howard, 1969, pi. 3: fig. 5; Howard and Frey, 1975, 1980c, fig. 20). Finally, small (less than 1 in, 2 cm, typically) to very small (a few millimeters in size) V-shaped "kinks" are common in tabular foresets (Figure 10). These may be escape traces of small arthropods (mole crabs, etc.) and/or bivalves, such as Donax or Mulinia (Frey and Howard, 1972). The assemblage of physical and biogenic struc- tures, taken as a whole, indicates a moderate to high-energy environment of rapidly shifting sands (and perhaps a faint reverse tidal flow, suggested by the graded thin silt and clay layers between sets and foresets), representing avalanche lobes or megaripples migrating with the dominant current direction, presumably tidal ebb, and normal or nearly normal salinity. No structures that could be assigned to reactivation surfaces (microerosion produced by tides; Klein, 1970:1118) were seen. The environments of the cross-bed member most probably were shallow subtidal to intertidal, evi- dently a point bar sequence. Recent estuarine point bars of coastal Georgia and South Carolina are elongated parallel with the axis of tidal streams (Barwis, 1978:140; Frey and Howard, 1980, fig. 7.1; Howard and Frey, 1980a, fig. 6.4, 1980c: 111-114). The sand bodies are divided longitudinally into two subfacies: the exposed, channelward side consists of relatively clean, graded or ungraded, megarippled to ripple- laminated sand, deposited under moderately high energy, whereas the more protected, marshward side consists of cross- or wavy-laminated muddy sand and mud deposited under lower energy conditions (Howard, Frey, and Reineck, 1973). Migrations of bars produce large curving foresets having amplitudes of 28 inches (70 cm) to more than 3 feet (1 m). Closely adjacent, shallow sub- tidal deposits are similar, essentially being sub- NORTH t + 10 +6 N = 43 ^ FIGURE 11.—Azimuthal histo- gram of foreset dip directions in tabular cross-bed member of Current unit. Scale (2-10) in- dicates number of readings within each 10-degree class in- terval. N equals total number of readings. ST aqueous extensions of the bars. In broader estu- aries, given an ample sediment supply, one would expect correspondingly broader sediment bodies, grading into high-energy sand flats or shoals and low-energy "mud flats," where the tidal range is high (Hayes, 1975), or into zigzag shoals within a place of moderate tidal range (Ludwick, 1974, fig. 7). A barrier island inlet system (Hubbard, Oertel, and Nummedal, 1979) is not postulated for this member, chiefly because evidence of the barrier itself is lacking. The tabular member is thinner to the north than to the south (Figure 4) and, as mentioned previously, is trough cross-bedded and thinner to the east. Perhaps the mine trenches had not uncovered the lateral equivalent of a muddy subfacies at the level of the tabular member. More likely, a muddy subfacies appears in the sequence as the overlying flaser beds, when the depositional centers shifted gradually with time; nevertheless, the geometry ofthe sand body itself seems to be one of a large equant lens, possibly reflecting either a large point bar or migrating small point bar. The wavy bedded member of the Current unit (Figure 12) is conformable above the tabular cross-bed member and consists of horizontal and ripple-laminated sand layers having mud drapes (wavy and flaser beds; Reineck and Wunderlich, 1968). The sand is fine to medium grained (1.60 to 2.03 (j)), poorly sorted (1.12 to 1.15 (jy), and would plot within or close to the cluster repre- sented by the Mud and Sand unit (Figure 8). The proportion of mud to sand changes markedly from one trench to the next, however. This mem- ber attains a thickness of 1 to 5 feet (0.3 to 1.5 m), thinning gradually to the north. NUMBER 53 245 w-s^Py.r%^^^^^^^^^^^^^^^'^^^^'-"f-«^ FIGURE 12.—Wavy bedded member of Current unit (upper two-thirds), showing flasers and mud layers. Box-shaped depressions (arrows) in heavy mineral layer in underlying tabular cross-bed member probably represent sediment collapse into Ophiomorpha nodosa below (horizontal clay-lined tube). Deep furrow above left arrow also may be a collapse structure, although small anemones make similar looking escape structures. Trench V, east end of north wall, 1971. 246 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Bloturbation was moderate to intense, and a wide variety of biogenic structures was found. The structures include easily recognizable Ophiomorpha nodosa, Skolithos linearis, and "sitz marks"; many others, less easily recognized, also are present. Most Ophiomorpha in the upper part of the tabular cross-bed member originated from the horizon of the wavy bedded member. Calli- anassa major is known to penetrate downward 10 to 12 or more feet (3 to 4 m) into beaches along the coast of Georgia today (Frey, Howard, and Pryor, 1978). Mud drapes (wavy and flaser beds) over cur- rent-rippled sand suggest a tidal influence and marked fluctuation in energy levels within the depositional environment (Howard and Frey, 1980a). Some mud wisps in ripple troughs prob- ably represent accumulations of fecal pellets; the polychaete Onuphis and the shrimp Callianassa (analogs for the Pleistocene animals constructing Skolithos and Ophiomorpha, respectively) contribute large quantities of fecal pellets today that are washed into and preserved within ripple troughs (Oertel, 1973; Pryor, 1975; Frey and Howard, 1980; Howard and Frey, 1980a, 1980c). The mod- erate to intense degree of bloturbation suggests moderately low depositional rates (in contrast to relatively rapid rates for the underlying tabular cross-bed member). Salinity was normal or nearly normal, based upon modern analogs of the bio- genic structures. Hence, either the water became deeper, in which case the wavy and flaser-bedded sand might correspond to the shallow subtidal part ofthe "estuarine accretionary beds" of How- ard and Frey (1980b); or, the water became shallower or remained at essentially the same depth, the deposit corresponding to an interme- diate-energy upper point bar facies or tidal flat adjacent to or overlapping the depocenter for the tabular member. We suggest a fining-upward point bar sequence, possibly spillover lobes filling a multilobed bar margin (Barwis, 1978). Thus, during deposition of sediments of the Current unit, avalanche deposits of clean-washed sand first were laid down in waters of moderate to shallow depth (a few lO's of feet, ~10 m, at most; probably less) in an estuary dominated by ebb currents. The water then shoaled, by up- building of point bars and related features, and/ or by a eustatic drop in sea level. If we accept the eustatic hypothesis, it may have been the first stage in the general eustatic drop that culminated in erosion into the Current, and Mud and Sand units (Figure 4, trenches IV, III). CHANNEL UNIT The Channel unit varies in thickness from zero to 11 feet (3.3 m) and consists of light-gray to white and tan, coarse- to fine-grained, predomi- nantly subangular to subrounded, quartz sand. Minor amounts of heavy minerals, where present, define ripple laminae and trough cross-beds, the latter being the dominant sedimentary structure in the unit (Figure 13). Trough cross-bed sets are 2 to 3 feet (60 to 90 cm) thick; wave lengths are estimated at 6 to 8 feet (1.8 to 2.4 m). Mud clasts up to 6 inches (15 cm) in size indicate (1) the erosion of compacted mud, possibly from the Mud and Sand unit and (2) excavation and transport of pieces of clay-lined Ophiomorpha. Woody debris, twigs, and a large log were found. Streaks of black organic matter commonly accen- tuate cross-bed laminae. The range of grain size and sorting for the Channel unit is greatest of any sand unit studied. Although only a few samples were analyzed, they were selected in order to accentuate the differences. Hence, poorly sorted fine sand, moderately sorted coarse sand, and well sorted fine- to medium-grained sand were found; the field of mean sizes (Figure 8) overlaps those of the Current, Mottled, and Mud and Sand units. Channel unit sands fill a wedge-shaped cut that pinches out to the north, south, and west. The cut penetrated the upper part of the Mud and Sand unit in trench IV (Figure 4), and originally may have cut down even farther; but mining had (by 1971) removed much of the evidence to the east. The clay plug seen 1 mile (1.6 km) to the east (east wall of pit; Figure 14) may have been part ofthe same channel network; it seems to be located in the same stratigraphic position as the Channel unit, between the Current NUMBER 53 247 FIGURE 13.—Channel unit, showing large-scale trough cross-beds. Mud clasts (some of which are fragmented, reworked Ophiomorpha nodosa) accumulated along single horizon (arrow). Dark layer at base, near field notebook, is part of Mud and Sand unit, clasts of which are incorporated into lower part of Channel unit. West end of Trench IV, near section 4 of Welch (1972). FIGURE 14.—Clay plug, probable remnant of Peat and Clay unit, and relationships of Mud and Sand unit and Current unit. Mottled unit, not recognized here, may have been removed by mining or obscured within highly oxidized upper few feet (meters) of section. East wall, Lee Creek Mine, 1972. (Photo by James Austin.) 248 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY unit and a Peat and Clay unit. Channel unit sands wedge out to the west; less and less of the unit was found as mining uncovered the section in that direction during 1972 and 1973. The Channel unit is overlain by the clay member of the Peat and Clay unit; the peat member lies on top of the clay member and the Current unit. Thus, in trenches V and II, no evidence of the Channel unit exists beneath the peat or above the Current unit. A chance roadcut or river bank exposure of Pleistocene sediments may not show the Channel unit phase of deposition at all; large scale mining operations such as those in the Tex- asgulf pit are perhaps necessary to unravel the complexities of the Pleistocene section of North Carolina. No known biogenic structures were found in the Channel unit, with the possible exception of one or two small V-shaped "kinks" ("sitz marks"?) in heavy mineral layers of trough cross-beds. For this reason, and because the overlying Peat and Clay unit apparently is freshwater in origin (p. 266), we conclude that the Channel unit consists of fluvial fill. Primary physical structures in at least one section suggest a fining-up sequence (Allen, 1965:142); trough cross-beds in coarse to medium sand are dominant near the base (Figure 13), and ripple laminae in medium to fine sand at the top. In other sections, trough cross-beds having minor climbing ripples comprise the entire Channel unit. The sequence of events that led to deposition ofthe Channel unit is believed to be (1) erosion (presumably during lowered sea level that accom- panied a glacial event) and cutting ofthe channel down to the depth of the Mud and Sand unit; (2) deposition of cross-bedded sand in a fining-up sequence (presumably as the stream's base level rose with rising sea level); and (3) abandonment of the channel and subsequent filling with sedi- ments of the clay member of the Peat and Clay unit. We deduce that no marine water, or even a salt wedge of estuarine circulation, entered this chan- nel until after deposition of the peat member of the Peat and Clay unit. For this reason, trough cross-bedded sands bearing Ophiomorpha and Sko- lithos found in sections elsewhere (Austin, 1973) must not be correlated with the Channel unit unless they clearly are (1) confined to an elongate channel and (2) overlain by freshwater peaty deposits which, in turn, are (3) overlain by the marine Mottled unit. A radiocarbon date on wood from the top of the Channel unit in trench III (as of 1971) ex- ceeded 42,000 years BP (Smithsonian Institution, Radiocarbon Laboratory, sample SI-1827). Fran- cis M. Hueber identified the wood as the red cedar, funiperus species aii.J. silicicola (Small); the genus is common along the Coastal Plain of North Carolina today. Thus, the channel was situated in a coastal setting. PEAT AND CLAY UNIT The Peat and Clay unit ranges in thickness from 6 inches (15 cm) to 11 feet (3.4 m). It consists of two members: (1) the peat member and (2) the clay member. Although these two lithologies in- tergrade with one another, the clay member gen- erally overlies the Channel unit, the peat member overlies both the Channel unit and the Current unit, and the clay member grades upward into the peat member where the former is present (Figure 4). Both members are found in the west and east wall of the mine. The clay member, where rich in organic debris, is a light chocolate brown when wet (it loses approximately one-third of its volume when dried) and is brownish gray where less rich in organic debris. It has no apparent internal lami- nae, although a crude bedding is seen (Figure 15). The only fossils found were burrows and rare plant remains. In one locality (trench III, Locality W-2, Figure 2) the peat apparently had been stripped away, and the clay member is riddled with Thalassinoides species down to a depth of 3 feet (1 m). Elsewhere, these burrows, which ex- tend down from the Mottled unit, penetrate both the peat and clay. Burrow density increases up- ward to the top ofthe clay member. Each burrow is filled with pure white quartz sand, believed to be related genetically to the overlying marine Mottled unit. NUMBER 53 249 FIGURE 15.—Clay member of Peat and Clay unit. Contact with underlying channel unit is 1 foot (30 cm) below shovel. Abundant Thalassinoides sp. filled with sand gives clay a "riddled" appearance (tip of shovel handle to top of exposure, 2 ft; 60 cm). Mottled unit was removed by mining. No peaty sand is preserved at this site. Trench III, near section 2 of Welch (1972). The peat member is a very granular, sandy, organic deposit containing roots and an occa- sional limb or twig of a woody plant. Its sandy, rubbly fabric is more like a soil than a true peat, and it may have been partially reworked by the sea during initial deposition of the Mottled unit. Where the clay member of the Peat and Clay unit is not present, the peat member lies strati- graphically between the Current unit and the Mottled unit (Figure 4), although in places the peat member, too, is missing. We believe that the clay member of the Peat and Clay unit resulted from clay deposition after abandonment of the fluvial channel that previ- ously had filled with sand (Channel unit). At the east end of the mine, only the clay plug and overlying peat were found (Figure 14). Clay plugs are common in fluvial deposits, and also are found in abandoned delta distributary channels. The overlying peat member is a freshwater de- posit (based upon pollen analyses; Whitehead, p. 266 herein). Because it is almost completely con- fined to the region of the Channel unit, it prob- ably represents a local coastal swamp that devel- oped over the old channel and adjacent lowland surface. The entire depositional history, from Channel unit sands through peat development, is one of progressive decrease in (1) sediment grain size, (2) energy of water movement, and ulti- mately in (3) supply of terrigenous sediment. The peaty deposit represents this last stage. Because the entire sequence can best be explained in terms of a rising sea level (including the marine Mottled unit above the peat), we believe that the fresh- water peaty sediment was developed in a coastal region near sea level, similar to much of eastern Georgia and the Carolinas today. A sample of peat collected in 1971 from the top of the peat member in trench III was dated by the Smithsonian Institution's Radiocarbon Lab- oratory (sample SI-1829). The date exceeded 42,000 years BP. Thus, the peat member is not a late Wisconsinan interstadial deposit. This date, coupled with Whitehead's assessment of the pol- len, suggests a probable Sangamon age. It com- pares favorably with a Sangamon age derived 250 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY from similar pollen from the Bull Creek peat, Horry Co., South Carolina (Whitehead and Campbell, 1976), but less favorably with intersta- dial pollen (Sirkin, Denny, and Rubin, 1977). MOTTLED UNIT The Mottled unit ranges in thickness from 4 to 9 feet (1.2 to 2.7 m) and occurs throughout the western part of the mine. It was not recognized in the cut at the eastern side ofthe mine, although all other units below it, excepting the Channel unit, were found. Its apparent absence may be explained by (1) its elimination by a facies change, (2) intense leaching, which destroyed the color contrast so that biogenic structures no longer are visible, or (3) its removal during early stages of mining. Nevertheless, we believe this unit to be widespread, and because it locally overlies a freshwater peat, it is of great signifi- cance in environmental interpretation. The Mottled unit consists of moderately to poorly sorted fine sand mottled with mud. It varies in color from light gray to tan and dark gray, depending on the percentage of mud mot- tles. The mottling originated through intense blo- turbation (Figure 16), which characterizes this unit. Physical sedimentary structures are found only where bloturbation was less intense; these undis- turbed areas occur as small patches only inches (cm) across. The original primary structures seem to have been interlayered sand and mud, in about equal proportions. Some suggestion of mud- draped ripples was found. The deposition rate was slow, however, and because of the active infauna, the sediments were intensely churned. The Mottled unit generally is fine to very fine sand, moderately well to poorly sorted (Figure 8); sample means range from 2.2 to 3.16 <^ and standard deviations from 0.75 to 1.49 (j). The Mottled unit and sandy member ofthe Mud and Sand unit have similar textural characteristics (Figure 5). The biogenic structures recognized include Ophiomorpha nodosa, Skolithos linearis, and Planolites beverleyensis. Other traces include two types of crab burrows and various escape structures, the latter possibly by anemones, although some ofthe more irregular ones possibly are bivalve resting traces. No fossil shells were found, but molds of several genera occur where concretions formed around one type of crab burrow. The other type of crab burrow is related to the top of the Mottled unit, perhaps developed high intertidally or supra- tidally (they are similar to Ocypode or perhaps Cardisoma burrows; Shinn, 1968, pi. Ill: fig. 1; Frey and Mayou, 1971), and thus may not have been genetically involved with the marine Mot- tled unit. The concretionary burrows are found within the Mottled Unit. These became hardened with limonite that developed as a crust over a clay and sand wall (Figure 17) possibly formed by the stone crab Menippe mercenaria (Curran and Frey, 1977:158). The shells were in the Mottled unit and incorporated in the wall of the burrow. Shells identified are the gastropods Oliva and Terebra and the bivalves Tellina, Mulinia, Ensis, and a scallop. Mulinia occurs in great numbers. Most shells are broken. No other trace of shell material was found in the normal mud and sand lithology ofthe Mottled unit. Waters and bottom sediments during deposition of the Mottled unit must have been rich in organic nutrients to sup- port so many suspension and deposit feeders; and perhaps the organic matter formed acids that later dissolved most shells away. Such sediments today typically are highly reduced, having low pH; except for the crab-burrow concretions, no evidence would remain of marine shells once present in abundance. This led us to wonder whether local conditions within other marine units of the Pleistocene section in the pit might not preserve some shells; but they remained elu- sive, perhaps having been dissolved away without a trace (cf. Stephens, Eason, and Pedlow, 1973; Frey, 1975, fig. 2.2). Crab burrows that developed concretions easily could be reconstructed as branched and un- branched, downward trending tunnels. Other crab burrows, however, did not develop concre- tions. The latter are 2 inches (5 cm) in diameter, penetrate more than 1 foot (30 cm) into the substrate, and are filled with clean quartz sand, NUMBER 53 251 FIGURE 16.—Mottled unit. Bloturbation largely destroyed original physical structures. Vertical tubes (white) are Skolithos linearis; dark clay rings are Ophiomorpha nodosa; large white diagonal burrows having very thin wall (arrows) possibly are Planolites or Palaeophycus. Trench V, southwest side, 1971. FIGURE 17.—Crab burrow exhibiting Tellina sp. mold in limonite crust, Mottled unit. Pits on surface of burrow (here merely a segment of a much larger structure, the upper part being entirely unlithified) are molds of broken bits of other shells, many of which are Mulinia sp. Trench III, northwest corner, 1971. 252 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY evidently derived from the overlying Rooty unit. These 1-shaped burrows possibly were made by the ghost crab Ocypode quadrata (Curran and Frey, 1977:159). Uniform mottling and general textures of the Mottled unit persist throughout the region stud- ied in the west end of the pit, whereas other muddy sand units show extreme variability even within the 0.65 mile (1.2 km) north-south expo- sure. Therefore, the Mottled unit must have had a more uniform environment of deposition; sedi- ments might have been deposited in a lagoon or estuary of nearly normal salinity. A shallow sub- tidal to intertidal flat, landward of a bay-mouth barrier or in the lower reaches of an estuary (as in Georgia today; Howard and Frey, 1975, 1980c), seems probable. Bloturbation by diverse organisms is intense at such sites, and most phys- ical sedimentary structures are modified or oblit- erated; the result is a mixture of highly blotur- bated muddy sand or sandy mud. We see no evidence for a protective barrier to the east, al- though such evidence might be forthcoming in further studies. However, tidal flats also may form in estuary mouths adjacent to the ocean- ward side of barriers where energy levels are reduced, as on protected parts of the Georgia coast (Howard and Dorjes, 1972). Shallow subti- dal mud and sand facies are found immediately offshore of Sapelo Island (Howard and Reineck, 1972:93, 100). The geometry ofthe unit may be crucial in determining which of the depositional models is most appropriate. After deposition ceased, intertidal to supratidal crab burrows {Ocypode, or possibly Cardisoma) ev- idently were excavated in the unit, either from the overlying Rooty unit (which was not then rooty) or during the time between deposition of sediments ofthe Mottled unit and Rooty unit (if the latter is a terrestrial deposit). These conclu- sions further suggest retreat of the sea following deposition ofthe Mottled unit. This possibility is not easily documented because (1) the contact between the Mottled unit and overlying Rooty unit is obscured by root mottling and may be gradational, and (2) we are not sure of the depo- sitional environment ofthe Rooty unit. ROOTY UNIT The Rooty unit, the uppermost unit in the section, consists of yellow to tan and rusty orange, fine- to medium-grained sand and ranges in thick- ness from 3 to 6 feet (1 to 1.8 m). The lower contact in most places is gradational with the underlying Mottled unit, but locally it is a sharp surface. Older Holocene and modern roots have penetrated this unit; a swamp overlying it prior to opening ofthe mine leached sediments consid- erably. These sediments apparently either lacked primary physical structures or, if present, were destroyed by ubiquitous subvertical root struc- tures and leaching. Limonitic concretions are common around rootlets. Two sediment samples from the unit show it to be medium- to fine- grained (1.31 to 2.68 42,000 BP (SI-1829). Given the sequence of sediments associated with this depositional cycle, it would be of interest to determine whether the peat represents a fresh- water deposit (a freshwater marsh deposit asso- ciated, at least initially, with the abandoned river channel) or whether it was formed in a brackish environment (related to the continuing trans- gression that produced the overlying marine sed- iments). Given the infinite radiocarbon date, it would also be of interest to determine whether this depositional unit dates from the Sangamon or from a mid-Wisconsinan interstadial. With these general questions in mind, I undertook a pollen analytical investigation ofthe peat. Methods Four samples of the peat and/or clay were analyzed. Two samples derived from trench III, one from trench IV, and one from the east wall (Figure 2 of Belt, Frey and Welch, herein). The samples were prepared by a standard procedure (boiling in KOH, demineralizing with HCl, boil- ing in cone. HF, and acetolysis) and mounted in silicone oil. Counting was carried out at a mag- nification of 400 diameters using a Leitz Ortho- plan microscope with apochromatic objectives. 265 266 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Results The samples from trenches III and IV con- tained too little pollen for detailed analysis. The slides consisted of finely comminuted organic de- bris, mostly too fine and badly preserved to iden- tify. Only five or six pollen grains were present on each of the slides and they were often badly corroded. The peat sample from the east wall differed in that the organic fragments were larger (mostly vascular tissue and leaf cuticular remains) and pollen was both more abundant and better pre- served. Three entire slides were counted to obtain sufficient pollen for reasonable interpretations. The results are presented in Tables 1 and 2. Note that two sets of percentages are given. "Percent of total pollen" involves a pollen sum including all tree, shrub and herb pollen, but excluding obvious aquatics. "Percent of arboreal pollen" are percentages for three types based on a pollen sum including only arboreal pollen (AP). The latter calculation was necessary so that the spec- trum could be compared more directly with other fossil and modern spectra from the southeast (few published modern and fossil spectra have such high percentages of non-arboreal pollen). Interpretations The high percentages of non-arboreal pollen (NAP) (mostly grass and sedge), and the pollen of a number of different aquatic plants suggest that much of the pollen was locally derived and that the depositional environment was character- ized by standing water for at least a portion of the year. This is further substantiated by the presence of fossil algae {Botryococcus, Pediastrum boryanum, and Tetraedron). A number ofthe aquat- ics represented occur in a wide range of coastal plain environments, sometimes including brack- ish waters (these include Typha, Potamogeton, Sag- ittaria, and Nuphar). However, others apparently occur only in freshwater habitats (both species of Myriophyllum, Brasenia, Nymphaea, Pontederia, and Isoetes) (Radford, Ahles, and Bell 1964). Further- more, the algae mentioned above are character- istic freshwater taxa (Smith, 1950:243, 269, 404). TABLE 1.—Arboreal, shrub, and herb pollen data from the Peat Member, Lee Creek Mine Type of pollen No. of grains % of total (arboreal) pollen ARBOREAL POLLEN Pinus 147 16.44(54.85) Picea 6 0.67 (2.24) Abies 1 0.11 (0.37) Cupressaceae 21 2.35 (7.84) (^ercus 60 6.71(22.39) Gary a 5 0.56 (1.87) Betula 8 0.89 (2.99) Fraxinus 3 0.34 (1.12) Corylus 5 0.56 (1.87) Nyssa 4 0.45 (1.49) Liquidambar 2 0.22 (0.75) Ostrya - Carpinus 2 0.22 (0.75) Castanea 3 0.34 (1.12) Populus 1 0.11 (0.37) Subtotal 268 29.98 SHRUB POLLEN Alnus 140 15.66 Lonicera 1 0.11 Ericaceae 5 0.56 Viburnum 1 0.11 Subtotal 147 16.44 HERB POLLEN Gramineae 215 24.05 Cyperaceae 197 22.04 Compositae 35 3.91 Ligulifiorae 2 0.22 Ambrosia 5 0.56 Artemisia 1 0.11 Rosaceae 1 0.11 Ranunculus 1 0.11 Thalictrum 5 0.56 Umbelliferae 5 0.56 Sanguisorba canadensis 1 0.11 chenopod-amaranth 11 1.23 Subtotal 479 53.58 Total 894 Thus, microfossil evidence is consistent with peat deposition in a freshwater rather than brack- ish environment. Furthermore, in brackish water sediments I would expect far higher percentages of chenopod-amaranth type pollen, low percent- ages of sedge pollen, pollen of typical saltmarsh shrubs (such as Ivafrutescens), and pollen oi Ruppia NUMBER 53 267 TABLE 2.—Aquatic, algal, and miscellaneous pollen data from the Peat Member, Lee Creek Mine Type of pollen No. of % of total grains pollen AQUATICS Polygonum {Persicaria type) 2 0.22 Potamogeton 2 0.22 Brasenia 6 0.67 Nuphar 3 0.34 Myriophyllum scabratum 3 0.34 Myriophyllum heterophyllum 1 0.11 Pontederia 2 0.22 Isoetes 3 0.34 Sagittaria 6 0.67 Nymphaea 1 0.11 Typha-Sparganium type 5 0.56 ALGAE Botryococcus 27 3.02 Pediastrum boryanum 2 0.22 Tetraedron 1 0.11 MISCELLANEOUS Sphagnum 11 1.23 Botrychium cf. dissectum 1 0.11 Osmunda regalis 2 0.22 Monolete fern 62 6.94 Trilete fern 1 0.11 Unknown 15 1.68 Unidentifiable 51 5.70 (e.g., Butler, 1959; Heusser, 1963). Consequently it is quite reasonable to assume that the peat is of freshwater origin. Much of the herb and shrub pollen (compos- ites, Thalictrum, Sanguisorba canadensis, alder) prob- ably derived from wet shores immediately sur- rounding the marsh habitat itself. The question ofthe age relationship ofthe peat deposit can also be resolved, but not quite as definitively. The virtual absence of pollen of "boreal" taxa certainly suggests a temperate en- vironment. Of the many such taxa known from full-glacial and late-glacial deposits in the south- east (e.g.. Whitehead, 1963, 1964, 1965, 1967, 1973, 1981; Frey, 1951, 1953, 1955; Craig 1969; Watts 1970), only Picea, Abies (a single grain), and Sanguisorba (a single grain) are represented in the peat. If one ignores the spruce (0.67% total pollen, 2.24% AP), then the spectrum is remarkably like modern pollen assemblages from both northeast- ern and southeastern North Carolina (e.g.. White- head, 1967, 1981; Whitehead and Tan, 1969; Frey 1951, 1953). The dominant tree types are pine and oak with a number of other temperate deciduous and coniferous taxa represented. The overall impression is of a vegetation reasonably comparable to the present, with climate perhaps a trifle cooler. It is thus apparent that the peat was deposited either during an interglacial or during a rather warm interstadial such as the Mid-Wisconsinan Plum Point or Port Talbot (Whitehead, 1973: 630; Dreimanis, 1973:377). The obvious ap- proach is to compare the pollen spectrum from the peat with known interglacial and interstadial spectra from the same general area. Mid-Wiscon- sinan interstadial spectra are known from the Bay Lakes of Bladen County, southeastern North Car- olina (Frey, 1951, 1953, 1955; Whitehead, 1965, 1967), from Rockyhock Bay in Chowan County, northeastern North Carolina (Whitehead, 1973, 1981), and from an exposure along the intracoas- tal waterway near Long Beach in southeastern North Carolina (Whitehead and Doyle, 1969). Interglacial spectra are known from the exposure at Flanner Beach on the Neuse River (only 50 km from the Lee Creek site) (Whitehead and Davis, 1969) and from the Kempsville formation in southeastern Virginia (Whitehead, unpublished data; Oaks and Coch, 1973). Basically, the spec- trum from the peat horizon is more similar to the interglacial spectra. The interstadial spectra dif- fer consistently in having higher percentages of spruce and more frequent occurrence of boreal taxa, such as Sanguisorba canadensis, Arceuthobium, Abies, Schizaea pusilla, Lycopodium annotinum, L. lu- cidulum, L. clavatum, and L. obscurum. However, the spruce percentage from the peat is higher than that recorded at either the Neuse River site or in the Kempsville Formation. Although the spruce data make assignment of the peat to an intergla- cial a little less certain, it should be emphasized that our knowledge of vegetational changes in the southeast during the Sangamon interglacial is much less nearly complete than our understand- 268 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY ing of conditions during the Plum Point and Port Talbot interstadials. It is obvious that conditions would be appropriate for the survival of some boreal taxa at either end of any interglacial se- quence. In summary, the pollen data from the peat horizon exposed in the sections at the Lee Creek Mine suggest clearly that the depositional envi- ronment was fresh water rather than brackish and that the peat was probably deposited during a portion of an interglacial rather than during an interstadial. Literature Cited Butler, P. 1959. Palynological Studies of the Barnstable Marsh, Cape Cod, Massachusetts. Ecology, 40(4):735-737, 1 figure, 1 table. Craig, A.J. 1969. Vegetational History of the Shenandoah Valley, Virginia. In S.A. Schumm and W.C. Bradley, editors. United States Contributions to Quater- nary Research. Geological Society of America, Special Paper, 123:283-296, 1 figure, 1 plate, 2 tables. Dreimanis, A. 1973. Mid-Wisconsin ofthe Eastern Great Lakes and St. Lawrence Region, North America. Eiszeitalter und Gegenwart, 23/24:377-379. Frey, D.G. 1951. Pollen Succession in the Sediments of Singletary Lake, North Carolina. Ecology, 32(3):518-533, 6 figures, 1 table. 1953. Regional Aspects of the Late-glacial and Post- glacial Pollen Succession of Southeastern North Carolina. Ecological Monographs, 23(3):289-313, 11 figures, 4 tables. 1955. A Time Revision of the Pleistocene Pollen Chro- nology of Southeastern North Carolina. Ecology, 36(4): 762-763. Heusser, C.J. 1963. Pollen Diagrams from Three Former Cedar Bogs in the Hackensack Tidal Marsh, Northeastern New Jersey. Bulletin of the Torrey Botanical Club, 90(1): 16-28, 2 figures. Oaks, R.Q. and N.K. Coch 1973. Post-Miocene Stratigraphy and Morphology, Southeastern Virginia. Virginia Division of Mineral Resources Bulletin, 82: [viii] + 135 pages, 33 figures, 2 plates, 6 tables, 1 appendix. Radford, A.E., H.E. Ahles, and CR. Bell 1964. Guide to the Vascular Flora of the Carolinas, with Dis- tribution in the Southeastern States. 383 pages, map. Chapel Hill, North Carolina: The Book Exchange, University of North Carolina. Smith, CM. 1950. The Fresh-Water Algae of the United States. Second edition, viii -1-719 pages, 559 figures, 3 tables. New York: McGraw-Hill. Watts, W.A. 1970. The Full-Glacial Vegetation of Northwestern Georgia. Ecology, 51(1): 17-33, 9 figures, 2 tables, 2 appendices. Whitehead, D.R. 1963. "Northern" Elements in the Pleistocene Flora of the Southeast. Ecology, 44(2):403-406, 2 figures. 1964. Fossil Pine Pollen and Full-Glacial Vegetation in Southeastern North Carolina. Ecology, 45(4): 767- 777, 7 figures, 3 tables. 1965. Palynology and Pleistocene Phytogeography of Unglaciated Eastern North America. In H.E. Wright, Jr., and D.G. Frey, editors, The (Quaternary ofthe United States, pages 417-432, 5 figures. Prince- ton, New Jersey: Princeton University Press. 1967. Studies of Full-Glacial Vegetation and Climate in Southeastern United States. In E.J. Gushing and H.E. Wright, Jr., editors, (^atemary Paleoecology, pages 237-248, 6 figures, 2 tables. New Haven, Connecticut: Yale University Press. 1973. Late-Wisconsin Vegetational Changes in Ungla- ciated Eastern North America. (Quaternary Research, 3(4):621-631, 5 figures. 1981. Late-Pleistocene Vegetational Changes in North- eastern North Carolina. Ecological Monographs, 51(4):451-471, 9 figures, 6 tables. Whitehead, D.R., and J.T. Davis 1969. Pollen Analysis of an Organic Clay from the In- terglacial Flanner Beach Formation, Craven County, North Carolina. Southeastern Geology, 10(3): 149-164, 3 figures, 2 tables. Whitehead, D.R., and M.V. Doyle 1969. Late Pleistocene Peats from Long Beach, North Carolina. Southeastern Geology, 10(1):1-16, 3 figures, 4 tables. Whitehead, D.R., and K.W. Tan 1969. Modern Vegetation and Pollen Rain in Bladen County, North Carolina. Ecology, 50(2):235-248, 3 figures, 9 tables. Fossil Woods and Resin-like Substances from the Lee Creek Mine Francis M. Hueber ABSTRACT The spoil heaps at the Lee Creek Mine, near Aurora, Beaufort County, North Carolina, are a source of limited quantities of fossil woods and resin-like substances. The woods are permineral- ized (Schopf, 1975:29) with quartz in crystalline or opaline form. Eight specimens of the woods were examined for this report. The stratigraphic sources of only three ofthe eight were established on the bases of foraminiferal assemblages and matrix composition. The specimens came from the lower 4 to 10 feet (1.2 to 3 m) of the early Pliocene Yorktown Formation. The gymnosper- mous genera Pinus hinnaeus, Juniperus Linnaeus, and Taxodium Richards are identified and the remaining five specimens belong to the angio- spermous family Caesalpiniaceae. Their generic identity is tentatively resolved and Gleditsia Lin- naeus is suggested. The biological source of the resin-like substances has not been determined, nor has the stratigraphic source been established for any but one ofthe 42 specimens examined. Its occurrence is in the upper lower or lower middle part ofthe Yorktown Formation at a point higher in the section than that of the woods. Fragments of bark, liquid-filled cavities, quartz-like crystals, and pyrite are the inclusions found in three ofthe specimens; the others are barren. The identity of the fossil woods with modern genera and species suggests their contemporaniety with the deposi- tion of the sediments in which they are found as opposed to being reworked from older sediments. A better stratigraphic control on the occurrences Francis M. Hueber, Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. of the woods in the section, as well as their age relationship to the resin-like substances, remain as the most important objectives in any meaning- ful, additional studies of these entities found at the Lee Creek Mine. Introduction Fossil woods can be found in nearly all of the states along the east coast of the United States but never in the quantity nor of the quality of preservation so typical of localities in several of our western states. Most commonly the fossil woods in the east occur as lignitized logs, twigs or drifted fragments and least commonly as permi- neralized remains. If permineralization is the mode of preservation, quartz is the primary embedding mineral while pyrite, marcasite, li- monite, or hematite are secondary in order of occurrence. Calcite is rarely encountered as the preserving agent, except in Pleistocene or Recent tuffs derived from spring deposits. Fossilized resins, broadly referred to as amber, are found in the Atlantic Coastal Plain. They are restricted in occurrence to the various formations of Cretaceous age ranging geographically from Marthas Vineyard, Massachusetts, southward to near Goldsboro, North Carolina. Most of the specimens are quite small, droplet-like pellets, while larger ones are generally found adhering to or embedded in fragments of lignitized bark or wood. This latter occurrence suggests association of the resin with wounds or wound areas on the 269 270 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY trunks of the source trees. The resin source of the amber has been attributed to gymnosperms, with the exception of one occurrence where analysis indicated an angiosperm source (Langenheim and Beck, 1968:86). The amber at all of the localities is too limited in quantity and quality to be of commercial significance. For a number of years permineralized wood, with quartz as the embedding mineral, and lumps of thoroughly solidified resin-like substances have been collected from the spoil heaps at the Lee Creek Mine. The number of specimens has never been very great; however, many ofthe specimens weigh several kilograms. This brief note is based on the study of eight specimens of wood and 42 specimens of resin-like material. ACKNOWLEDGMENTS.—I wish to express my gratitude to the following collectors who have generously donated to the paleobotanical collec- tions ofthe National Museum of Natural History specimens of fossil wood and resin-like material from the Lee Creek Mine: Gerard Case, Peter J. Harmatuk, James Kaltenbach, Earl Mason, Jack H. McLellan, Royal Mapes, James G. Mead, Franklin Pearce, Robert Purdy, Clayton E. Ray, Donna Ray, Clyde Swindell, and James West- gate. I am particularly appreciative of Thomas G. Gibson's help in establishing the horizons from which four specimens (3 fossil wood, 1 resin-like specimen) were obtained. James P. Ferrigno produced the excellent pho- tographs and photomicrographs. The scanning electron microscope photographs were made pos- sible through the very capable work of Susann Braden and Mary-Jacquelyn Mann of the Na- tional Museum of Natural History SEM labora- tory. Materials and Methods All of the fossil wood specimens from the Lee Creek Mine thus far donated to the Smithsonian paleobotanical collections, represent silicified, well-eroded, and rounded fragments of secondary wood derived from trunks of trees. No branches. twigs, or shrub-like material have been observed. Transverse, radial, and tangential sections were obtained by making appropriate cuts of the ma- terial with a standard 8-inch (20-cm) diamond saw. The sawed surfaces were ground to smooth- ness with 600 grit carborundum powder on a glass plate, etched with concentrated hydrofluoric acid for 10 to 30 seconds, carefully rinsed by dipping into several changes of fresh water, air dried, and peeled by means of the Joy, Willis Lacey, technique (1956) using acetone and 76.2 jum (0.003 in) thickness cellulose acetate paper. The resulting peels were mounted in Canada balsam on standard glass microscope slides. Specimen USNM 267213, Pinus species cf. P. palustris Miller, is so porous and thoroughly mi- neralized that it was necessary to prepare stan- dard ground thin-sections that were then stained in acid Bismarck Brown solution in order to differentiate detail. This technique was described by Bartholomew, Matten, and Wheeler, 1970. The specimens of resin-like substances comprise pieces from 3 X 2 X 1 cm to 30 X 36 X 11.5 cm in overall dimensions. All 42 in the collections at this time represent eroded fragments from larger masses, even in the case of the largest specimen. Ground thin-sections were prepared from one particularly well-banded specimen (USNM 267217) for microscopic study. Epoxy 220 resin was used to attach the sections to standard glass microscope slides, and they were then ground on a glass plate to smoothness and necessary thin- ness, using water and 1200 grit aluminum oxide. The sections were covered with high-viscosity Canada balsam in order to prevent swelling of the epoxy mounting medium. A glass coverslip was used to seal the mount. Surfaces of some of the transluscent to trans- parent resinous specimens were ground and pol- ished to permit better examination of their inte- riors for possible inclusions. Sanding on a 220 fixed grit belt sander followed by a worn 600 fixed grit belt produced a surface that could then be finished to a high luster, using a soft cloth impregnated with titanium oxide final polishing compound. Unfortunately, although the polish NUMBER 53 271 was bright, it was temporary. The surfaces be- came dull with any handling and soon became opaque. Freshly broken surfaces showed the same changes with time. Fragments of plant tissue found in one speci- men of the resinous material and portions of mineral-filled cavities in another were attached to aluminum stubs using white glue (Elmer's brand). They were then coated with carbon and gold-palladium to a thickness of ~500 A, and observed and photographed in a Cambridge Stereoscan Mark II A scanning electron micro- scope. Stratigraphic Occurrence of Specimens The stratigraphic sources have been established for only four of the fifty specimens considered in this report. Analyses by T.G. Gibson of the fora- miniferal assemblages and the sedimentary char- acteristics of matrix removed from protected cav- ities in three specimens of the fossil woods and from the fillings of shells of barnacles attached to the surface of one specimen of the resin-like sub- stance (Plate 2: figure 10) are quoted here and serve to date those particular specimens. Matrix from fossil wood specimen USNM 267218 contained: Nomonella amis d'Orbigny, El- phidium clavatum Cushman, Buliminella elegantissima (d'Orbigny), Cassidulina laevigata (d'Orbigny), Nonion pizarrense Berry, Bulimina elongata d'Orbigny, Bolivina paula Cushman and Ponton, Hanzawaia concentrica Cushman, Globigerina falco- nensis Blow, small grains of secondary phosphate, and echinoid spines. This assemblage and the sedimentary characteristics indicate that this fos- sil wood specimen is from the lower part of the Yorktown Formation, in the lower 10 feet (3 m) of the Lee Creek Mine section. Matrix from fossil wood specimen USNM 298768 contained: Bulimina elongata d'Orbigny, Buccellafrigida (Cushman), Buliminella elegantissima (d'Orbigny), Elphidium clavatum Cushman, Cibi- cides lobatulus (Walker and Jacob), Epistominella pontoni Cushman, Globigerina bulloides d'Orbigny, Hanzawaia concentrica Cushman, Bolivina paula Cushman and Ponton, Virgulina fusiformis Cush- man, and small grains of secondary phosphate. This assemblage and the sedimentary character- istics indicate that this fossil wood specimen is from the lower part of the Yorktown Formation, in the lower 10 feet (3 m) of the section at the Lee Creek Mine. The matrix from fossil wood specimen USNM 298767 contained: Elphidium clavatum Cushman, Cibicides lobatulus (Walker and Jacob), Cassidulina laevigata d'Orbigny, Nomonella amis d'Orbigny, Buccella fngida (Cushman), and medium to large phosphate grains. This assemblage and the sedi- mentary characteristics indicate that this speci- men of fossil wood is probably from within the lower 4 feet (1.2 m) ofthe Yorktown Formation in the Lee Creek Mine section. Matrix filling the barnacle shells attached to the specimen of resin-like substance USNM 267220 contained: Elphidium clavatum Cushman, Uvigerina subperegnna Cushman and Kleinpell, Bu- liminella elegantissima (d'Orbigny), Cibicides lobatu- lus (Walker and Jacob), Globigerina falconensis Blow, Rosalina floridana Cushman, Bolivina floridana Cushman and Ponton, Nomonella amis (d'Orbigny), Lagena substriata Williamson, Angu- logerina occidentalis (Cushman), Bolivina plicatella Cushman, and echinoid spines. This assemblage and the sedimentary characteristics indicate that this specimen would come from the upper lower to middle part ofthe Yorktown Formation in the Lee Creek Mine section. The remainder of the specimens in this study were collected from the spoil heaps of the mine and were either clean when collected or were subsequently washed free of any matrix before donation to the museum collections. As a result no precise stratigraphic occurrence can be estab- lished for them. Systematics Division PINOPHYTA Class PINOPSIDA Order FINALES Family PINACEAE 272 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Genus Pinus Linnaeus Family CUPRESSACEAE Genus Juniperus Linnaeus Family TAXODIACEAE Genus Taxodium Richards Division MAGNOLIOPHYTA Class MAGNOLIOPSIDA Order FABALES Family CAESALPINIACEAE Genus Gleditsia Linnaeus Genus Pinus Linnaeus Pinus species cf. P. palustris Miller PLATE 1: FIGURES 1-5 This description is based on a single specimen (USNM 267213) collected and donated by J. Kaltenbach. It is 17.5 cm long and 9.5 cm wide, the latter dimension measured along the radial section of the wood. Virtually all of the organic matter of the wood has been replaced by quartz. Subsequent leaching by ground water has re- moved any remaining organic matter and has made the specimen quite porous. The nature of the pitting in the tracheid walls as well as in the cross-field contacts of the ray parenchyma were particularly difficult to demonstrate because of the replication in quartz of the cell walls. Twelve rather broad growth rings are clearly visible in the cross-section of the wood, the early wood being the broadest and most evenly tex- tured. The late wood is narrower and clearly defined by the presence of numerous longitudinal resin canals (Plate 1: figure 1). There are a few resin canals distributed unevenly through the early wood. The transition from early to late wood is gradual. Rays are very fine and can be seen only where a horizontal resin canal is in- cluded in the ray. Tracheids in cross-section are 22 to 64 jUm in diameter (average 35 urn). Bordered pits are found on the radial walls in single rows or less commonly in two rows arranged oppositely (Plate 1: figure 5). No pits were observed on the trach- eids in tangential section in the preparations at hand. Ray parenchyma with one, rarely two, large unbordered pits per cross-field is not well pre- served (Plate 1: figure 4). The rays are of two types, uniseriate and fusiform (Plate 1: figure 3) and 1 to 20 cells high (average 8). The fusiform rays contain a horizontal resin canal (Plate 1: figure 3 at arrows). It has not been possible to determine the presence of ray tracheids in this specimen. The longitudinal resin canals are large, 156 to 180 iim in diameter, and have poorly preserved remains of a thin-walled epithelium lining the walls (Plate 1: figure 2). The horizontal resin canals are much smaller, ranging from 48 to 70 jLim in diameter (average 54 jum) and are also lined with an epithelial layer or remnants of tyloses. Although the cell structure in this specimen is somewhat obscured by the effects of preservation, the detail that has been obtained compares fa- vorably with sections of modern wood of Pinus palustris Miller (longleaf pine) from the collections of the Smithsonian Division of Plant Anatomy. Comparison of cell dimensions and clearer detail of ray structure is precluded. The stratigraphic source of this specimen is not known. Genus Juniperus Linnaeus Juniperus sp, cf. J. virginiana Linnaeus PLATE 2: FIGURES 1-4 USNM 267215 has distinct growth rings marked by a dense, narrow band of late wood and a less dense broad band of early wood; transition between early and late wood abrupt. Rays are numerous and fine. Parenchyma scat- tered and gum filled (Plate 2: figure 1). Resin canals not present. Tracheids small, 23 to 35 jtim in diameter; bordered pits in one row on radial walls (poorly preserved) (Plate 2: figure 3 at arrows). Rays uniseriate, low, 1-10 cells high (mostly 4 or 5) (Plate 2: figure 2), homogeneous, cross-field pits single and large (Plate 2: figure 4) or smaller and NUMBER 53 273 paired but more specific details not determined because of poor preservation ofthe material. Comparisons of the anatomical details of this specimen with those of sections from modern Juniperus virginiana Linnaeus (eastern red cedar) in the collections of the Smithsonian Division of Plant Anatomy suggest close agreement in the identification with that species. Positive identifi- cation is precluded by the rather poor preserva- tion of finer anatomical detail in the fossil mate- rial. This specimen was collected by J.H. McLellan and bore his original collection number 237. The fragment is 3.5 X 2.75 X 2.5 cm in overall dimen- sions. The stratigraphic source of this specimen is not known. 30.5 X 17.75 X 7.5 cm in overall dimensions. Mineralization of the specimen is more of an opal-like form of quartz rather than the crystal- line type found in the other fossil wood samples examined here. The cellular detail in this specimen is excel- lently well preserved, and because of this the identification to species (bald cypress) is made with full confidence. Comparison with slides pre- pared from modern examples of the species housed in the collections of the Smithsonian Di- vision of Plant Anatomy, substantiated the iden- tification. Unfortunately, the horizon from which this specimen was obtained is not known. Genus Gleditsia Linnaeus Genus Taxodium Richards Taxodium distichum Richards PLATE 1: FIGURES 6-8 USNM 267214 has quite distinct growth rings marked by a narrow band of late wood and a very broad band of early wood. The transition between the early and late wood is abrupt (Plate 1: figure 6). Rays are conspicuous. Parenchyma is scattered randomly through the early wood and is quite obvious because of the dark resinous contents ofthe individual cells (Plate 1: figure 6). Tracheids 48 to 68 junt in diameter with bor- dered pits in single, commonly double, and only occasionally three rows on the radial walls. When in multiples the pits are opposite, and crassulae are present between vertical pairs (Plate 1: figure 8). Rays are uniseriate, 2 to 13 cells high, com- posed wholly of parenchyma and exhibit 1 to 3 bordered pits per cross-field area (Plate 1: figure 8). Longitudinal parenchyma is diffused randomly through the wood and is clearly differentiated from other cells because of the dark material filling the cell lumens (Plate 1: figures 7, 8). This specimen was collected by C.E. Ray. It is Gleditsia species PLATE 2: FIGURES 5-7, PLATE 3: FIGURES 1-4 The remaining specimens of fossil wood in this limited study belong to the Magnoliopsida, that is, to dicotyledonous broad-leafed trees. Initial examination of the specimens, following the first cutting with a diamond saw, gave the impression that all ofthe specimens were identifiable with at most two genera, each one in a different family. Subsequent microscopic examination of cellular detail led to the conclusion that only one genus may be represented from the family Caesalpini- aceae. This family comprises mostly tropical gen- era; however, our native Gleditsia Linnaeus (honey locust) and Gymnoclaudus Lamarck (Kentucky cof- fee tree) are examples of temperate climate mem- bers. Ofthe two genera, Gleditsia is the most likely one represented by the fossil woods in the present study. Structural characteristics of each of the specimens overlap so strikingly that they may be identical, and for that reason I have chosen to describe and illustrate only two (USNM 267216 and 267218) of the five specimens representing the Magnoliopsida. The specimens illustrated by means of photo- micrographs of cellulose acetate peel sections in Plate 2: figures 5-7 (USNM 267216) and Plate 3: 274 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY figures 1-4 (USNM 267218) were donated by Mr. Peter Harmatuk. Before sectioning, the for- mer was 19 cm long, 8 cm wide, and 5 cm thick. It represents a well-rounded fragment from the trunk of a tree that was at least 66 cm (26 in) in diameter as determined by projecting into a com- plete circle the arc of the growth rings along the outer margin of the specimen. It was probably a much larger tree. The latter specimen was 28 cm long, 10.5 cm wide, and 3.5 cm thick and repre- sents a fragment of a trunk at least 102.6 cm (40 in) in diameter. The wood is ring porous (Plate 2: figure 5; Plate 3: figure 1) and the transition between early and late wood is abrupt. Vessels are medium- to large-sized, 95-265 ju,m in diam- eter, mostly round in cross-section, solitary but forming a band 2 to 5 cells wide in the early wood. Late wood vessels small, 70-100 jum wide, solitary or in small clusters (Plate 3: figure 1). Paratracheal parenchyma conspicuous (Plate 3: figure 1) around vessels and in late wood passing over into the succeeding early wood layer. Vessel members are truncate (Plate 2: figure 7), lack ligulae, and have simple perforation plates (Plate 3: figure 3). Intervascular pitting crowded, an- gular (Plate 2: figure 7) pits 4.4-5.2 jum in diam- eter. Rays are high, measured up to 2.4 mm long, multiseriate 4 to 8 cells wide, occasionally uni- seriate (Plate 2: figure 6, Plate 3: figure 2), ho- mogeneous and unstoried. A very commonly ob- served characteristic is the evidence of dark gums filling many of the vessels (Plate 2: figure 5) in large areas of the woods. This is true of all of the specimens examined. Variation from the discription given above is only slight among the specimens observed. Widths of growth rings, which are quite variable, allow for changes in number of vessels per unit area, minor differences in diameters ofthe vessels, more prominent parenchyma distribution, at least in wider growth rings, and in some instances the appearance of semi-ring porous structure to the wood. But the microscopic anatomical details remain to unite the group as a whole. The age of specimen USNM 267216 could not be determined; however, USNM 267218 came from the lower 10 feet ofthe Yorktown Formation of the Lee Creek Mine. Interestingly, one of the other dated specimens (USNM 298768) came from the same horizon, while USNM 298767 came from the lower 4 feet of the formation. Comparisons made between the fossil woods and slide preparations of modern woods from the genera Gleditsia, Robinia Linnaeus, Gymnocladus and even Hymenaea Linnaeus in the collections of the Division of Plant Anatomy support the as- signment ofthe fossils to Gleditsia. The gross, but probably insignificant, variations seen in the fossil woods are not all readily observable in the mod- ern material. The finer anatomical details, how- ever, are supportive to the determination of the genus. Resin-Like Substances The collection of resin-like substances in the Smithsonian from the Lee Creek Mine comprises 42 whole or freshly broken and fragmented spec- imens. On visual examination, there appear to be four basic forms: (1) transparent, lustrous, yellow to very light reddish brown, approach most closely the appearance of amber; (2) transluscent to nearly opaque, lustrous, massive to well- banded, dark reddish brown (Plate 3: figures 5, 6, 8); (3) opaque, dull to waxy luster, light golden, well banded, and generally with a porous texture (Plate 3: figure 9); and (4) transparent to trans- luscent on thin edges, waxy luster, very dark brown to black in massive pieces, coarsely banded or swirled patterning. In all but form 1, the odor of freshly broken or slightly heated surfaces is bituminous and rather unpleasant. In form 1 the odor is not bituminous nor does it have the fragrance characteristic of most ambers when heated. Thus far it has not been possible to define the source or type of organic substances represented by these resin-like materials. The stratigraphic occurrence of the resin-like substances has been determined for only one spec- imen (Plate 3: figure 6), USNM 267220, and that is the top lower or lower middle part of the NUMBER 53 275 Yorktown Formation. The specimen is younger than any of the fossil woods from the mine for which ages have been determined. All of the specimens are fragments of even larger masses as evidenced by the banding that is abruptly truncated at the margins of the speci- mens (Plate 3: figure 8). The swirled patterns are also in sectional view (Plate 3: figure 9). It is obvious that the resin-like substances were well solidified before transport into the marine environment of deposition. Evidence of the bor- ings of marine organisms (?Pholadidae) and the attachment of barnacles (JChthalmus) and bry- ozoans to the surfaces of some specimens is illus- trated by specimen USNM 267220 (Plate 3: fig- ures 6, 7, 10). At the outset of this study the resin-like speci- mens were examined for inclusions, such as plant debris, insects, and pollen. All of the banded patterning that would suggest inclusion of foreign matter (Plate 2: figure 8; Plate 3: figures 6, 8, 9) has proved to be alternating densities of small spherical bodies (Plate 2: figures 9, 10, 11) that are neither liquid nor gas filled. Instead, they are solid and appear to be a different phase of the resin-like substance itself. In one specimen, a form 1 type as described above, donated by Mr. Swindell (USNM 267223), fragments of bark were found (Plate 4: figures 1,2). No identification is possible as to the type of tree from which the bark could have come. It does, however, represent the only inclu- sion of plant material found in any of the 42 specimens in the study collection and establishes the possibility that the form 1 type of resin-like substance is of plant origin. Bubble-like cavities are present in most of the specimens in the collection. The cavities usually are empty. However, in one specimen of form 1 material (USNM 267225) the cavities are liquid filled and in another specimen, form 3 material (USNM 267224) quartz-like crystals are present in some ofthe cavities and pyrite in others (Plate 4: figures 3-7). These liquid and mineral inclu- sions merit additional study. The entire collection of resin-like substances will be stored in the paleobotanical collections of the Smithsonian Institution and will be available for additional research to qualified specialists. Discussion At the outset of this study there was hope to establish a relationship between the resin-like substances and the fossil woods found at the Lee Creek Mine. The first assumption that proved wrong was that the woods were permineralized by phosphates. Quartz proved to be the permi- neralizing agent. The questions arose as to the source of the quartz and why the woods were the only fossil remains to be preserved by that min- eral. There were no immediately satisfying an- swers to these questions. Further the well-rounded nature of the specimens (as if tumbled, perhaps during fluvial transport) and the lack of borings by marine organisms suggest that the woods were silicified elsewhere at an earlier time, then rede- posited at the present site. Further, the one piece of resin-like substance that could be dated proved to be younger than any of the dated woods, placing in limbo any possible relationship be- tween the two materials. Thus, when it became apparent that there was no stratigraphic control on most of the specimens at hand, further efforts to identify the woods and analyze the resin-like substances seemed unwarranted. It would be presumptive to assume that all of the fossil woods and resin-like substances found at the mine were originally from the Yorktown Formation of the Lee Creek section. Some of the specimens have been arbitrarily labelled as from the Pungo River Formation. It is to be hoped that special efforts will be made in further col- lecting at the mine to obtain specimens of known or determinable stratigraphic occurrence. The analyses of the matrix will establish horizons bearing the wood and resin-like substances and some clear picture may develop that will answer some of the questions posed. The results of this study suggest that the woods and resin-like substances are probably not much older than the sediments in which they are found. Clearly the woods are identifiable with modern 276 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY genera, and, even more significantly, they are close to, if not identical to, modern species. That the resin-like substances are of plant origin still cannot be proved. Literature Cited Bartholomew, R.L., L.C. Matten and E.F. Wheeler 1970. Staining Silicified Woods. Journal of Paleontology, 44:905-907. Joy, K.W., A.J. Willis, and W.S. Lacey 1956. A Rapid Cellulose Peel Technique in Paleobotany. Annals of Botany, new series, 20:635-637. Langenheim, J., and C.W. Beck 1968. Catalogue of Infrared Spectra of Amber, Pt. I: North and South America. Harvard University Bo- tanical Museum Leaflets, 22(3):65-120. Schopf, J.M. 1975. Modes of Fossil Preservation. Review of Palaeobotany and Palynology, 20:27-53, plates 1-3. Plates 1-4 278 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 1 Pmus species cf P. palustris Miller, USNM 267213 1. Transverse section showing resin canals and general texture of the wood. Growth ring exceeds height of photograph, lower three resin canals mark the end of early and beginning of late wood, X 50. 2. Transverse section of longitudinal resin canal. Thin-walled cells lining the canal are poorly preserved. Radiating fibrilar structure in the canal is mineral growth, X 150. 3. Tangential section showing short, predominantly uniseriate rays except in those rays containing horizontal resin canals (arrows), X 50. 4. Radial section, in which ray cell walls are poorly preserved and accordingly cross-field pitting (arrows) is difficult to illustrate, X 50. 5. Radial section showing single and double rows of circular-bordered pits; pits arranged oppositely when in two rows, X 200. Taxodium distichum Richards, USNM 267214 6. Transverse section showing abrupt change from summer to spring wood. Scattered, resin- filled parenchyma cells are readily visible as dark spots, X 45. 7. Tangential section showing uniseriate rays and resin-filled longitudinal parenchyma, X 145. 8. Radial section showing arrangement of bordered pits, cross-field pits in the ray parenchyma (upper arrow) and crassulae (lower arrow), X 150. 6. NUMBER 53 279 280 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 2 Juniperus species cf y. virginiana Linnaeus, USNM 267215 1. Transverse section showing growth rings and general texture ofthe wood, X 50. 2. Tangential section showing short, uniseriate rays, X 80. 3. Radial section showing remnants of circular bordered pits (arrows) in single rows, X 150. 4. Radial section showing single, large cross-field pits (arrows) in the ray parenchyma, X 80. Gleditsia species, USNM 267216 5. Transverse section showing growth ring (between upper and lower arrows) and general texture ofthe wood, X 27. 6. Tangential section showing high multiseriate rays, X 40. 7. Tangential section showing truncate vessel members with fine, angular intervascular pitting, X 105. Resin-like substance, USNM 267217 8. View of broken and eroded surface of resin-like substance showing banding, X 34. 9. Polished thin-section of dense, light-colored band in the specimen showing small spheres; spheres are solids, not liquid or gas, X 250. 10. Polished thin-section of area between dense bands showing diminution of numbers of spheres, representing a clear area in the substance, X 250. 11. Polished thin-section showing variability in sizes and density of numbers of the spheres, which results in the banding ofthe resin-like substance, X 250. 10. NUMBER 53 281 ^fV^r-v)4^r-i^N-J-v'- ( i _|, j—— as*Mi#iJtJWMj(fl8l(A 282 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 3 Gleditsia species, USNM 267218 1. Transverse section showing growth rings and general texture ofthe wood, X 40. 2. Tangential section showing early vessels (right) and late vessels (upper left), and morphology of the uniseriate and multiseriate rays, X 40. 3. Radial section showing simple perforation plate in vessel member, X 40. 4. Radial section showing storied paratracheal parenchyma associated with vessel members, X 40. Resin-like substances and associated fauna 5. USNM 267219, lustrous broken surface of nearly homogeneous resin-like substance with large bubble cavities, X 1/2. 6. USNM 267220, broken surface showing luster, color banding and section through borings made by marine organisms (PPholadidae), X 1. 7. USNM 267220, enlargement of borings seen in figure 6, X 2. 8. USNM 267221, fragment of banded resin-like substance, banding parallel and truncated at ends of specimen, X 1/2. 9. USNM 267222, broken surface of resin-like substance showing peculiar swirled pattern of banding, X 1/2. 10. USNM 267220 (same as in figures 6 and 7) showing barnacles (?Chtlialmus sp.) attached to the surface ofthe resin-like substance; bryozoans are also present at the right margin ofthe specimen, X 2. NUMBER 53 283 284 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 4 Inclusions in resin-like specimens, all SEM photographs 1. USNM 267223, fragment of bark; surface in tangential section, X 250. 2. USNM 267223, same fragment as in figure 1; surface in transverse section, X 249. 3. USNM 267224, quartz-like crystals included in bubble cavity in resin-like substance, X 30. 4. USNM 267224, side view of quartz-like crystals in figure 3, enlarged view, X 85. 5. USNM 267224, vertical view of quartz-like crystals in figure 3, X 38. 6. USNM 267224, pyrite crystals coating inner wall of bubble cavity in resin-like substance, X 620. 7. USNM 267224, pyrite crystals enlarged from same cavity as figure 6, X 3800. 1. NUMBER 53 285 Biostratigraphy and Paleoecology of a Diatomaceous Clay Unit in the Miocene Pungo River Formation of Beaufort County, North Carolina William H. Abbott and John J. Ernissee ABSTRACT The diatomaceous clay unit from two cores from the Pungo River Formation of Beaufort County, North Carolina, contains two diatom assemblages and essentially one silicoflagellate assemblage. Based on the diatom ranges, an age equivalent to Blow's (1969) zones N8-N9 was obtained for the older diatom assemblage and an age equivalent to zone Nil for the younger as- semblage. The age range of the silicoflagellate assemblage is inclusive ofthe diatom assemblages. The diatomaceous clay was deposited in a marine near-shore environment with reducing bottom conditions and nutrient-rich surficial water. The water may have been cooler or upwelling greater during the deposition ofthe younger assemblage. Introduction Brown (1958) was the first to describe the Miocene sediments of Beaufort County, North Carolina. He considered these Miocene beds to be correlative with the Calvert Formation of Maryland on the basis of Foraminifera at the top ofthe unit. William H. Abbott, Mobil Exploration and Producing Services, Applied Stratigraphy, P.O. Box 900, Dallas, Texas 75221. John J. Ernissee, Earth Sciences and Resources Institute, Byrnes Center, University of South Carolina, Columbia, South Carolina 29208. Kimrey (1964) formally designated the Mio- cene beds of Beaufort County, North Carolina, as the Pungo River Formation. The type section is from a core near Belhaven on the Pungo River in Beaufort County, North Carolina (Figure 1). Kimrey described the lithology of the Pungo River Formation as interbedded phosphatic sands, silts and clays, diatomaceous clays, and phosphatic and nonphosphatic limestones. The Pungo River Formation was deposited in a northeast-southwest trending basin and varies in thickness from a "feather-edge" east of Wash- ington, North Carolina, to greater than 120 feet (36.5 m) near the Pamlico River in eastern Beau- fort County (Kimrey, 1965:6). The phosphorite beds of the Pungo River Formation unconform- ably overlie the Castle Hayne Limestone of Eocene age, and are unconformably overlain by the Yorktown Formation of Pliocene age. Gibson (1967) divided the Pungo River section into seven units (Figure 2), with unit 1 at the bottom and unit 7 at the top. On the basis of an abundant planktonic foraminiferal assemblage in the upper 3 meters ofthe Pungo River Formation, he correlated the uppermost beds of the Pungo River with the Globigennatella insueta zone of Trin- idad and Venezuela. Gibson (1967) also corre- lated Shattuck's (1904) "zone" 10 ofthe Calvert Formation of Maryland with the Globigennatella insueta-Globigerinoides bisphericus zone of Blow (1959) and the overlying Globorotalia fohsi bansa- 287 288 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY exFLANxrwN I 4/C£A £4ST CF THIS UNB UNOeRLilM f 6V TWf PUNGP Mid FOCMA-riON ▲ AmOWMAXe SITE or RA-li-OILL m imoMMATE sin OF PA-H-GRL FIGURE 1.—Map showing Pungo River Formation, location of type core, cores used in this study, and the Lee Creek Mine. nensis zone of Bolli (1957). These zones are equiv- alent to Blow's (1969) zones N8 and N9 and within the Langhian Stage. Gibson (1967, 1971) found that the lower units did not have an ade- quately preserved foraminiferal assemblage to al- low dating, but suggested that molluscan molds from the lower beds had a closer affinity with Oligocene than with Miocene species. METHODS.—Samples were examined from cores taken in 1962 by the Ground Water Branch of the U.S. Geological Survey, in Beaufort County, North Carolina. These cores were mega- scopically and microscopically described in great detail by Kimrey (1965). The authors used Kim- rey's description to sample siliceous portions of two cores, RA-13-GRL and PA-31-GRL (Figure 1). Though all intervals were not available, enough were obtained to derive a nearly complete sequence of siliceous material. Samples were brought to the laboratory in plastic bags to avoid contamination. Preparation of samples was done using a mod- ification of Schrader's (1974b) method: the sam- ples are placed in 250 ml beakers in a 1:1 solution of 0.5N HCl and 30-35% hydrogen peroxide. They are heated for approximately 45 minutes; then shaken, and the supernatant liquid poured into 50 ml test tubes for centrifuging. The samples are centrifuged for 2 minutes at a speed of approximately 1200 revolutions per min- ute; water with suspended clay minerals is de- canted; and the residue re-suspended with dis- tilled water. The procedure of centrifuging, de- canting, and washing is repeated 7 times or until the decantate is clear. The residue is soaked for 5 minutes in a solution of 0.5% sodium pyrophosphate in water shaken thoroughly and centrifuged. This procedure will suspend more of the clay-sized particles. The sodium pyrophosphate solution with suspended clay is decanted, and centrifuging with distilled water is repeated 4 times. The residue is diluted with distilled water and the samples are stored in glass vials. Long-term storage is facilitated by the addition of a few drops of buffered formalin and out ofthe middle ofthe bottle two or three drops NUMBER 53 289 A 50 'X) OWN FM ££T^"-Z 40 - YORKT r—Zz^T^E ION c^o ooooacx^ <0 |05 IC3 \^\<=>\ 30 - «>c^OC»OgO H tf FORM; QOOO O OOO O ^O^ ^ OO0 OL OOOCXP ooc liJ 0COC70 Ooc > oooo 0 Ooo O o o o cP O Oa 01 0 oo o o o oo 20 - ooo o oooc o OOO 0 o OOO <5 o oo o o ooo z Q Q o o o ooo Z) Q O O O O O OO a. OOOOO OC>o 0 O O O O C> 90 QOO oa000 0 oa 00 £>ot> 0 00 '^ ^ oot? 0 00 ^0Ooo 000 O^e>oo ^^0oOooa 10 - 000OOOOO S u. 1 ' 1 ' 1 UJ z >: 1 1 1 < TLE H 1 1 1 CAS 1 1 1 0 ' fl 1 1 'uYellow-green sand and bryozoan fragments (unit 7) Interbedded yellowgreen hydrozoan frognnents and phosphote lenses (unit 6) Limestone with ptiosphate grains, many molluscon molds ond casts (unit 5) Alternating limestone and phosphate beds (unit 4) -Dark greenish brown phosphatic sands (unit 3) ■ Greenish-groy diatomaceous clay (unit 2) — Dork greenish brown phosphatic sands (unit I) FIGURE 2.—Seven units ofthe Pungo River Formation, modified from Gibson (1967). are taken with a disposable bar-straw and placed on a cover glass of 18 X 18 mm. The sample is dried at 45 °C. After drying, 1 drop of hyrax is placed on a slide and the cover glass mounted and heated for 5 minutes at approximately 150°C. Several slides from each sample were scanned with a light microscope to determine the species of diatoms and silicoflagellates present. ACKNOWLEDGMENTS.—The authors are grateful to Jim Sanpair and Jim Coffee of the Division of Earth Resources, Department of Natural and Economic Resources of North Carolina, who per- mitted access to the core material and assisted with the sampling. George Andrews and John Barron of the U.S. Geological Survey were ex- tremely helpful with the diatom taxonomy. Many thanks go to Tom Gibson, U.S. Geological Sur- vey, for his help, and to Alan-Jon Zupan, South Carolina Survey, who reviewed the manuscript. Ms. Sharon Huffman and Mrs. Cynthia Johns have been of great assistance in preparing the manuscript, and Camille Ransom in sample preparation. William H. Abbott prepared the diatom floral reference list and John J. Ernissee prepared the silicoflagellate floral reference list. Diatomaceous Clays of the Pungo River Formation Unit 2 of the Pungo River Formation consists of greenish gray diatomaceous clay, which can be traced areally throughout the formation. The authors sampled the diatomaceous unit from 290 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY RA-I3-GRL PA-31-GRL -V METERS FEET ri30 ^^l/--7 o ^^ i( < ^ o ^ 55--I80 o ziz: 5g X o ^//^•/ / >/ -/ ^ 7TT / - / / ^=V^ < UJ 2 o (K 2 o s "^ LJ > 5 o 6 70--230 FIGURE 3.—Cores RA-13-GRL and PA-31-GRL. Lithology based on the descriptions of Kimrey (1965). ''"'"'^f^'^'7*'''''^'^^* 7~r7 ^^ ^ ^ NUMBER 5:3 291 Kimrey's cores RA-13-GRL and PA-31-GRL (Figure 3), which is considerably thicker than the diatomaceons unit described by Gibson (1967). The diatom assemblage in this unit consists of a well-preserved marine flora (Table 1) containing 55 genera and 114 species of diatoms, of which numerous species characteristic of the Miocene were identified. In addition, a silicoflagellate as- semblage of 17 species, as well as many species of ebridians, phytoliths, and endoskeletal dinofla- gellates, were observed. Two distinct diatom assemblages are present. In core RA-13-GRL, diatomaceous samples were taken from 163 to 179 feet (48.9 to 53.7 m). The lower assemblage, assemblage A, was found in the interval between 179 and 174 feet (53.7 and 52.2 m). From approximately 174 to 165 feet (52.2 to 49.5 m), diatoms were sparse and badly broken. The higher assemblage, assemblage B, was encountered in the interval between 165 and 162 feet (49.5 and 48.6 m). Though less well preserved, both diatom assemblages were easily identified in PA-31-GRL (Table 1). Assemblages A and B share many species in common (Table 1), but important differences also exist. In assemblage B, diatoms typical of coastal upwelling, such as Thalassionema mtzschiodes, Thal- assiothrix longissima, and Denticula spp. were very abundant, whereas the same species were sparse in assemblage A. In assemblage A, Biddulphia tuomeyii, an extant subtropical species, is quite abundant, but it is not found at all in assemblage B. These diatoms suggest that there are environ- mental differences between assemblages A and B. It would be difficult to determine what climatic or environmental factors caused the differences other than to say that assemblage B was deposited during a period of greater upwelling of cool nu- trient-rich water. Biostratigraphy Stratigraphically, the most significant species of assemblage A are Raphidodiscus marylandicus Christian, Macrora stella (Azpeitia) Hanna, and Annellus californicus Tempere. Raphidodiscus marylan- dicus is restricted to Blow's (1969) zones N8 through NIO (Andrews, 1973; Kanaya and Ko- izumi, 1970). In addition, Macrora stella and An- nellus californicus are believed to be restricted to zones N8 and N9 or across the early/middle Miocene boundary (Cavallero, 1974; Opdyke, Burckle, and Todd, 1974). Assemblage A ofthe Pungo River Formation is therefore correlative with Blow's (1969) zones N8 and N9. Other diatom species that characterize this assemblage are Actinocyclus ingens Rattray, Bruniop- sis mirabilis Karsten, Cladogramma dubium Lohman, Coscinodiscus marginatus Ehrenberg, C. perforatus Eh- renberg, C. praeyabei Schrader, Cussia paleacea Schrader, C. praepaleacea Schrader, Cymatogoma am- blyoceros Hanna, Delphineis ovata Andrews, D. pe- nelliptica Andrews, Denticula hustedtii Simonsen and Kanaya, D. lauta Bailey, Hemiauluspolymorphus Grunow, Mediana splendida Sheshukova-Poret- skaya, Rhizosolema miocenica Schrader, Sceptroneis grandis Abbott, Stephanopyxis corona Ehrenberg, S. lineata Forti, S. turns Ralfs, Synedra jouseana She- shukova-Poretskaya, Thalassionema nitzschioides Grunow, Thalassiothrix longissima Cleve and Gru- now, Tnceratium condecorum Brightwell, T. tessella- tum Greville, Trinaria excavata Heiberg, and Tro- chosira concava Sheshukova-Poretskaya. Assemblage B contains several significant dia- tom species: Actinocyclus ellipticus Grunow, Cosci- nodiscus lewisianus Greville, Rouxia diploneides Schrader, and R. naviculoides Schrader. Both Actin- ocyclus ellipticus and Coscinodiscus lewisianus are common in assemblage B. Schrader (1973) in his North Pacific diatom zonation (Figure 4) places the first occurrence o{ A. ellipticus near the top of his zone 20 and the last occurrence of C lewisianus at the base of zone 20 or top of 21. Actinocyclus ellipticus was found very near the base of Schrader's zone 20 in the Miocene of South Car- olina suggesting that A. ellipticus occurs earlier in lower latitudes (Ernissee, Abbott, and Huddles- tun, 1977). Also, according to Schrader's zona- tion, Rouxia diploneides becomes extinct in the upper part of NPD zone 21 and R. naviculoides makes its first appearance. Roxia diploneides and R. naviculoides are both found in assemblage B sug- 292 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY European Standord Stages Age in m.y. Radiolorian Zones Nonnoplanklon Zones Forominiferol Zones Silicoflagellate Zones North Pacific Diotom Zones Diafom Zones (Japon) Serrovolllon 12 — -13 - -14 — -15 - -16 - -17 - NN9 NI5 Dlctyocho rhombico NPD 16 Denticula hustedti- Dentlcula lato petter ssoni -7_?_7_ N 14 NNe NPD 19 NI3 NN7 NPD 20 NI2 NN6 Nil Dorcodospyris aloto NN5 Corbisema fnocontha NPD 21 NPD 25 NIO Denticula loto o o N9 N8 NN4 Burdigalion Colocycletto costato N7 Adapted fro Ryan, el al., m 1974 Adopted from Ryon,etol,l974 ♦ From Goll, 1972 Adapted from Ryon.et al, 1974 Adapted from Ryan,etol, 1974 Adopted from Mort 101,1971,1972 ond Dumitrica, 1973 Adopted from Scliroder,l973and Berggren and *on Couvering,1974 Adopted from Koizumi, l975ond Berggren ond von Couvering,1974 FIGURE 4.—Middle Miocene microfossil correlation chart. gesting that assemblage B is equivalent to the top of Schrader's NPD zone 21 or at the zone 21/ zone 20 boundary. This interval, according to Berggren and Van Couvering (1974:170), is within zone Nil of Blow's (1969) foraminiferal zonation. Other characteristic species of assemblage B (Table 1) are Actinocyclus ingens Rattray, Brumopsis mirabilis Karsten, Coscinodiscus marginatus Ehren- berg, C. perforatus Ehrenberg, C. plicatus Grunow, C. praeyabei Schrader, Cussia paleacea Schrader, C. praepaleacea Schrader, Cymatogoma amblyoceros Hanna, Delphineis angustata (Pantocsek) Andrews, D. penelliptica Andrews, Denticula hustedtii Simon- sen and Kanaya, D. lauta Bailey, D. nicobarica Grunow, Hemiaulus polymorphus Grunow, Mediaria splendida Sheshukova-Poretskaya, Navicula pennata Schmidt, Stephanopyxis corona Ehrenberg, 6". lineata Forti, S. turris Ralfs, Synedra jouseana Sheshukova- Poretskaya, Thalassionema nitzschioides Grunow, Thalassiothrix longissima Cleve and Grunow, Tricer- atium condecorum Brightwell, T. tessellatum Greville, and Trochosira concava Sheshukova-Poretskaya. The silicoflagellate species within the two as- semblages are similar, and no stratigraphically significant distinctions between the two assem- NUMBER 53 293 blages can be made (Table 2). Both assemblages are consistent with an assignment to the Corbisema triacantha zone as defined by Martini (1971:1696, 1972) and with the subsequent use of that zone by Bukry and Foster (1973, 1974), Dumitrica (1973), and Ciesielski (1975). The upper bound- ary of the C. triacantha zone (Figure 4) has been placed within the nannofossil zone NN6 (Bukry and Foster, 1973) and NN5 (Bukry, 1973). The bottom of the zone occurs roughly at the NN4/3 boundary (Martini, 1972). These would in turn be correlative to middle N12 to N7 of the stan- dard foraminiferal zones (Ryan, 1974). Thus, the silicoflagellate zone assignments are inclusive of the diatom assignments. Distephanus stauracanthus Ehrenberg, which oc- curs only in samples from PA-31-GRL, 198' and 205', does permit much tighter control on those samples because of its narrow range. Martini (1971, 1972) and Dumitrica (1973) suggest that this species is confined entirely within the NN6 nannofossil zone and would, therefore, be correl- ative to the Nl 1-lower part of N12 foraminifera) zones (Ernissee, Abbott, and Huddlestun, 1977). The silicoflagellate assemblages from the Pungo River Formation samples, particularly those of RA-13-GRL, show all but two of the species illustrated in Tynan's (1957) paper on the Calvert Formation of Maryland, and suggest close age correlation with that deposit. However, Tynan's samples are confined to Shattuck's (1904; Tynan, 1957, fig. 1) zones 3 through 10, which is essentially the lower two-thirds of the Calvert Formation. The stratigraphically signifi- cant Distephanus stauracanthus is absent from Tynan's samples, however, and only a broad time-correlation is possible. Paleoenvironinent Microfossils in the Pungo River diatomaceous clay unit suggest a definite marine environment with no indication of freshwater forms. The dia- toms in both assemblages A and B include many coastal, neritic, and littoral forms that could be found along the present Atlantic coastline. The presence of oceanic, pelagic diatoms and silico- flagellates indicates open marine circulation. As previously noted (p. 291) diatoms typical of coastal upwelling are extremely prevalent in as- semblage B and, though present, not as abundant in assemblage A. In assemblage B, Biddulphia tuomeyii, a subtropical species, is a dominant form in the assemblage. This species is commonly found along the coast of the Carolinas today. If the extinct species were eliminated from assem- blage A, then the assemblage would not be very different from a modern coastal assemblage at the same latitude. Coastal upwelling apparently increased consid- erably during the time of deposition of assem- blage B, bringing cooler nutrient-rich waters into the area. This may have been the result of chang- ing wind patterns, changing ocean currents, or both. The clastic component of both assemblages is predominantly microfossil remains with some clay. Therefore, the basin in which these diato- maceous clays were deposited appears to have been sediment-starved. The abundance of benthic diatoms suggests that water depth was not greater than that to which light could penetrate. Since light penetra- tion is dependent upon water turbidity caused by sediment suspension and also by productivity, it is impossible to estimate actual depth. As in Miocene sediments of South Carolina (Abbott, 1974a, 1975), both assemblages A and B contained opaline phytoliths similar to those found in prairie grasses. These opaline phytoliths (Abbott, 1975) suggest that climatic conditions were drier during Pungo River time (lower-mid- dle Miocene) than the present. This is further supported by the presence of hickory and abun- dant oak pollen with an absence of pine pollen (James Darrell, Georgia Southern College, Statesboro, pers. comm., 1976). Rooney and Kerr (1967) suggest that the Pungo River Formation was deposited in a shal- low marine basin characterized by a reducing environment, with the presence of pyroclastic material. Our data do not support the finding of oa yyyyy y y yyyyoi y v:^ y y yyyyy O -31- yyyyy y yy CQ y y\ y y, y y y y y^ < OH yyyyy y y < < y yy y : lassio lassio lassio eratiu ubrotu "ssella acria hosirc thiopy « « a .'-^ i^ Q R K K E^ fi; E^ i~^ ^ h CO cn XX XX X X X X X X X X X O 1J-) 31- o CM X XX X X yyyyy X X PA- o OJ X X X X X xxx y X X X CO CO xxxxxxxx X xxx X X 00 >> x-x X X X X X X yyy X J .^ Di r- X X X X X X X X o ^^ CO tn X X X X X X X X X X < ' ' Q^ to cr> xxx X X XX X X X r- r^ X X X X X X X X X X cr. xxx X xxx X X X o o -2 _s •~ -S C! D D o V "g 3 ^ ^ 1 -a 2 S < oa 3 OJ CS i-i "3 3 2 -^ :2 *(j St j:;; o- cL .2 S rt '" . R ■^ ■^ ^ ■5 ... ^ 3 -2 >-i :~ -c .1-1 a 3 2 <« « 3 R s 2 ~ « <-— V3 .3 ^ -Cj, C g 3 a a w ^ ^ g -^ sens ombi ocha -§ I "i ~2 R 2 l: S .Q Vo S Cl R O" ^ £ ~R .=1 -CL, -55, ^ ^J 13 -R lo lo *., « 2 "o g a S/ ku^ K> ip^ /S /S rS < X X X X X X X X X X X X X XXX X < X X X X X XX X < X X X X X X < X X X X X X X X X X X < X X X X 1-5 hus a o ^ a aspedodiscus coscinodiscus ssia baleacea praepaleacea clotella kelloggi matogoma amblyoci malosira andersoni immunis malosira sp. A •Iphineis angustata lineata novaecaesaraea ovata penelliptica 'nticula hustedtii lauta nicobarica norwegica punctata icladia pylea Iploneis crabro iploneis sp. A jji^f/za hyalina idiclya robusta miothecium rogersii 'miaulus cf. polymo valodiscus laevis radiscus asperulus bipolans ova lis thodesmium minusc acrora stella ediaria splendida p/o.nra westii wicula directa hennedyit pennata tzschia sp. B t a Cj ^ CJ^CJ'CJ ^ c:)CiqciqqqqciQqc:iqQ[.'V PLATE 18 1. Dossetia hyalina Andrews, length 54 microns, width 35 microns, RA-13-164'8" 2. Chaetoceros species, ~ X 1500, RA-13-177'9" 3. Asteromphalus robustus Castracane, diameter 35 microns, PA-31-205'. 1. 338 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 19 1, 2. Cannopilus binoculus (Ehrenberg) Lemmermann: a, apical focus; b, basal ring focus; RA 13- 177'9" 3, 4. Cannopilus haeckelii Lemmermann: a, apical focus; b, basal ring focus; RA 13-177'9" 5, 6. Cannopilus hemisphaericus (Ehrenberg) Haeckel: a, apical focus; b, basal ring focus; RA 13- 177'9" (All photographs at the same magnification; scale = 0.05 mm) NUMBER 53 339 340 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 20 1. Cannopilus hemisphaericus (Ehrenberg) Haeckel: a, apical focus; b basal ring focus; RA 13- 177'9" 2. Cannopilus hemisphaericus (Ehrenberg) Haeckel: a, apical focus; b, basal ring focus; RA 13- 163'8" 3. Cannopilus sphaericus Gemeinhardt: a, low focus; b, mid-focus; c, high focus (note the large lateral windows); RA 13-163'8" 4. Cannopilus sphaericus Gemeinhardt: a, high focus; b, mid-focus (note accessory spines near the equatorial region ofthe apical apparatus); c, low focus (note the first row of lateral windows are larger than those remaining in the apical apparatus); RA 13-164'8" 5. Cannopilus triommata (Ehrenberg) Lemmermann: a, apical focus; b, basal ring focus; RA 13- 163'8" (All photographs at the same magnification, scale = 0.05 mm) NUMBER 53 341 342 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 21 1. Cannopilus triommata (Ehrenberg) Lemmermann: a, apical focus; b basal ring focus; RA 13- 177'9" 2. Cannopilus species A: a, apical focus; b, basal ring focus; RA 13-176'. 3. Cannopilus species B, RA 13-163'8" 4-8. Corbisema triacantha (Ehrenberg) Hanna: 4, PA 31-198'2"; 5, PA 31-198'2"; 6, 7, RA 13- 163'8"; 8, RA 13-177'9" (All photographs at the same magnification, scale = 0.05 mm) NUMBER 53 343 WM 0" 8 344 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 22 1-3. Dictyocha fibula Ehrenberg, sensu lato: a, apical focus; b, basal ring focus; RA 13-177'9" 4-9. Dictyocha fibula Ehrenberg, sensu lato: 4-7, 9, RA 13-177'9"; 8, RA 13-164'8" (All photographs at the same magnification, scale = 0.05 mm) NUMBER 53 345 346 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 23 1. Dictyocha fibula Ehrenberg, sensu lato: a, apical focus; b, basal ring focus; PA 31-198'2" 2, 5. Dictyocha rhombica (Schultz) Deflandre, sensu lato, PA 31-198'2" 3. Dictyocha cf Dictyocha species A, PA 31-205'4" 4. Dictyocha rhombica (Schultz) Deflandre, sensu lato: a, apical focus; b, basal ring focus; RA 13-164'8" 6-8. Dictyocha species A: 6, PA 31-205'4"; 7, 8, PA 31-198'2" 9. Dictyocha species A: a, apical focus; b, basal ring focus; PA 31-198'2" Note that figure 7 does not have the double-ended basal horns that show up well in figures 6, 8, and 9. A transitional morphologic series can be postulated from figure 4 through figures 2, 5, 3, 7, 8. (All photographs at the same magnification; scale = 0.05 mm) NUMBER 53 347 348 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 24 1. Dictyocha species B: a, apical focus; b, basal ring focus; PA 31-198'2" This specimen shows a more pronounced elongation of the basal ring and a more severe straightening of the lateral segments ofthe basal ring (cf Plate 23: figures 1,7). 2-6. Distephanus crux (Ehrenberg) Haeckel, sensu lato: 2, RA 13-177'9"; 3, PA 31-198'2"; 4, RA 13-164'8"; 5, PA 31-205'4"; 6, PA 31-198'2" 7, 9. Distephanus crux (Ehrenberg) Haeckel, sensu lato: a, apical focus; b, basal ring focus; RA 13-177'9". Note the surface ornamentation in figure 7. 8. Distephanus crux (Ehrenberg) Haeckel, sensu lato: a, apical focus; b, basal ring focus; RA 13-164'8" (All photographs at the same magnification, scale = 0.05 mm) NUMBER 53 349 350 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 25 1. Distephanus crux (Ehrenberg) Haeckel, sensu lato: a, apical focus; b, basal ring focus; PA 31-198'2" 2, 3. Distephanus crux (Ehrenerg) Haeckel, sensu lato, PA 31-198'2" 4,5, 7. Distephanus speculum (Ehrenberg) Haeckel, sensu lato: 4, 5, RA 13-163'8"; 7, PA 31- 198'2" Note 7-sided basal ring in figure 4. 6. Distephanus speculum (Ehrenberg) Haeckel, sensu lato: a, apical focus (specimen tilted); b, basal ring focus; RA 13-163'8" 8. Distephanus speculum var. pentagonus Lemmermann, RA 13-163'8" 9. Distephanus speculum var. pentagonus Lemmermann: a, apical focus; b, basal ring focus; RA 13-163'8" (All photographs at the same magnification, scale = 0.05 mm) NUMBER 53 351 352 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 26 1, 2. Distephanus stauracanthus Ehrenberg, PA 31-198'2" 3, 4, 6. Halicalyptra picassoi (Stradner) Dumitrica: 3, 4, RA 13-163'8"; 6, RA 13-164'8" 5. Halicalyptra picassoi (Stradner) Dumitrica: a, high focus; b, mid-focus; RA 13-163'8" 7-10. Mesocena cf M. elliptica (Ehrenberg) Ehrenberg: 7, RA 13-179'; 8, PA 31-205'4"; 9, RA 13-176'; 10, PA 31-198'2". Note the variability in size and shape. (All photographs at the same magnification, scale = 0.05 mm) NUMBER 53 353 Key Foraminifera from Upper Oligocene to Lower Pleistocene Strata of the Central Atlantic Coastal Plain Thomas G. Gibson ABSTRACT Biostratigraphically important planktonic and benthic foraminiferal species from strata of late Oligocene to early Pleistocene age in the central Atlantic Coastal Plain are described and illus- trated. Thirty planktonic species are used, in conjunction with a few radiometric ages, to date the strata. The ages derived are: "Silverdale" beds of latest Oligocene age; Pungo River and Calvert formations of late early to early middle Miocene age; Choptank Formation of middle middle Miocene age; St. Marys Formation of late middle to early late Miocene age; "Virginia St. Marys" beds of late Miocene age; Yorktown For- mation of early to late(?) Pliocene age; and up- permost "Yorktown," Croatan, and Waccamaw formations of late Pliocene and early Pleistocene age. Thirty-seven species and subspecies of benthic Foraminifera important for regional correlation are described, and ranges are given for this area. New species described are Bolivina pungoensis, Bo- livinopsis fairhavenensis, Epistominella pungoensis, Cib- icides cravenensis, C. croatanensis, C. pungoensis, Svrat- kina croatanensis, Nonion calvertensis, Florilus chesa- peakensis, and Elphidium neocrespinae. New subspe- cies described are Nonion advenum pustulosum and Elphidium latispatium pontium. Thomas G. Gibson (United States Geological Survey), National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Introduction The age assignments of Cenozoic strata in the Atlantic Coastal Plain have been determined pri- marily from molluscan data. Although numerous molluscan groups, including the pectens, Astarte, and Crassatella, are valuable in correlation within the Atlantic Coastal Plain, the lack of molluscan species common to North America and Europe limits correlation with the European stratotype sections for the Oligocene, Miocene, Pliocene, and Pleistocene. Many benthic foraminiferal spe- cies are common to both areas, but species ranges generally are too long for the refined correlation desired. This study used planktonic Foraminifera to correlate Atlantic Coastal Plain strata with inter- continental planktonic foraminiferal zones and the European stratotypes of Cenozoic stages. Gib- son (1967), Akers (1972), and Hazel (1977) as- signed some of the strata to planktonic zones. Strata ranging in age from late Oligocene to early Pleistocene and extending from Maryland to southern North Carolina were examined. Unfor- tunately, planktonic specimens are rare in many of the outcropping sections because of the shal- low-water environment of deposition. This scar- city, combined with the cool-water nature of most of the assemblages, makes correlation difficult. The examination of large samples, however, pro- 355 356 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY duced rare but biostratigraphically important specimens in many of the formations. Radiomet- ric ages supplement the foraminiferal data in some of the formations. Key benthic species are proposed for correla- tion within the region and should prove valuable for dating subsurface sections. Thirty-seven benthic species that characterize one or more formations are discussed, and the range of each within the study area is given. Some ofthe species have greater ranges in other geographic and en- vironmental areas. Species from well samples near the Atlantic Coast allow cross-indexing of the characteristic shallow-water species found in out- crops, and should serve as faunal links to assem- blages found in strata on the Atlantic Continental Shelf. Type specimens and all specimens illustrated here are deposited in the USNM collections of the National Museum of Natural History, Smith- sonian Institution. ACKNOWLEDGMENTS.—The drawings are by Larry Isham of the National Museum of Natural History and Jeffrey Lund and Elinor Stromberg of the U.S. Geological Survey. Susann Braden, Walter Brown, and Mary-Jacquelyn Mann ofthe National Museum of Natural History took the SEM photographs. Well samples were supplied by Philip Brown of the U.S. Geological Survey. The author benefited from stimulating discus- sions on foraminiferal taxonomy with Bruce Hay- ward ofthe New Zealand Geological Survey. The manuscript was reviewed by Laurel Bybell, Thomas Cronin, and Joseph Hazel of the U.S. Geological Survey, and their helpful suggestions are incorporated. Age of the Formations CORRELATION WITH EUROPEAN STAGES The age placements by various authors for formations in the middle and late Tertiary show relative stability in the assignments of the older units, such as the Calvert Formation, and consid- erable, relatively recent changes in the placement of younger units, such as the Yorktown and Wac- camaw formations. The correlation charts of Mansfield (1943) and Cooke, Gardner, and Woodring (1943) reflected the thinking at that time of many workers regarding the Atlantic Coastal Plain. In retrospect, however, the general stage of development of the invertebrate and vertebrate faunas appears to have been the only basis upon which these early workers correlated Atlantic Coastal Plain units with the European stages. Planktonic Foraminifera and nannoplank- ton were not used then, and radiometric dating was unavailable. Moreover, stages such as the Sarmatian and Pontian, with which the York- town Formation has been correlated (Mansfield, 1943) are represented by deposits formed in brackish to nonmarine environments in the east- ern Mediterranean region. The type areas for these stages thus contain molluscan assemblages different from those found in the shallow-marine western Atlantic environments represented in the Yorktown Formation. Even within similar shal- low marine environments, there are very few molluscan species common to both the Atlantic Coastal Plain and western Europe (Mongin, 1959:329). The changes in age assignments made during the past 20 years include changes in interpreta- tion ofthe age ofthe supposed later Miocene and younger formations. Although the recent use of planktonic groups has had a great effect on bio- stratigraphic studies, doubt about age assign- ments began because of the disparity in ages indicated by vertebrates and invertebrates. DuBar (1958:142) assigned a late Pleistocene age to the Caloosahatchee Formation of Florida, a unit that had been traditionally placed in the late Pliocene. His age assignment was based upon the occurrence of Equus, the modern horse, a genus considered indicative of a Pleistocene age. This change in age assignment disturbed those who studied invertebrates, because the large number of extinct molluscan species in the Caloosa- hatchee (as much as 50 percent, DuBar, 1958:137) suggested an older age. The resistance to accepting so many extinct species in Pleistocene strata probably is related to the Lyellian concept that less than 10 percent of the mollusks in the NUMBER 53 357 Pleistocene are extinct. This resistance also ap- pears to be related to the then accepted date of about one million years for the base of the Pleis- tocene, with the resultant rapid extinction of half ofthe assemblage. After DuBar's age assignment, common opinion among those who worked with invertebrates was as follows: If the Caloosa- hatchee were indeed Pleistocene in age, then the moderate similarity of its molluscan faunas to those of the Yorktown Formation would mean that the somewhat older Yorktown also must be interpreted as younger. Because the Yorktown Formation at that time was considered the "standard" unit of late Miocene age on the At- lantic Coastal Plain, a more recent date would indicate a Pliocene age. Although a Pleistocene age for the Caloosahatchee largely was dis- counted, the general result of this controversy was a new awareness ofthe uncertainties of previously accepted age assignments of Atlantic Coastal Plain Cenozoic strata. Another opinion on the younger age of the Yorktown Formation resulted from the author's conversations with Remington Kellogg in the early 1960s concerning the age of a new baleen whale skeleton {^'^Balaena") found in the Yorktown Formation of Rice's Pit at Hampton Roads, Vir- ginia. The baleen whale from the Yorktown For- mation was similar to living baleen whales, and, in light of known rates of whale evolution, Kel- logg could not resolve this similarity in terms of a late Miocene age for the Yorktown specimen. A possible explanation was that these beds in the upper half of the Yorktown Formation were younger than late Miocene. The age significance of vertebrate specimens reported upon by Frank C. Whitmore, Jr. during the early 1960s also supported a post-late Mio- cene age for the Yorktown Formation. Whitmore (1965:A71) reported the following: Two mammalian fossils recently given to the Survey for identification have raised questions concerning the corre- spondence of age classifications ofthe coastal plain sediments based on vertebrate and on invertebrate fossils, respectively. A single lower horse molar, collected from the Yorktown Formation (upper Miocene) at Cobham Wharf, Va., has been identified by F.C. Whitmore, Jr., ofthe U.S. Geological Survey and M.F. Skinner, ofthe Frick Laboratory, American Museum of Natural History, as Hipparion cf H. eurystyle Cope, of Clarendonian or early Hemphillian (early or early middle Pliocene) age. The possibility that part ofthe York- town may be of Pliocene age is further supported by the finding in 1960, at Hampton, Va., of a baleen whale of a type that seems too advanced for the Miocene. Part of a horse skull, found in the Caloosahatchee For- mation in its type area near Labelle, Fla., has been identified by Whitmore as Equus, probably a mid-Pleistocene species. This find is of particular interest because the rich inverte- brate fauna of the Caloosahatchee is generally regarded as being of Pliocene age. Thus, most vertebrate evidence suggested that at least part of the Yorktown Formation was Pliocene in age. The tooth collected at Cobham Wharf, Virginia, however, was a float specimen from the modern beach rather than one collected in place from the Yorktown strata. Hence, doubt remained about whether it came from the fossi- iiferous Yorktown or from overlying sand units that tentatively were considered to be Pleistocene in age. As a result of these questions concerning the age of the Yorktown Formation, the author ex- amined planktonic Foraminifera from the York- town, because this group was being used to cor- relate with the European stages. On the basis of the sparse assemblages from the lower part of the Yorktown Formation, Gibson (1967:638) indi- cated a middle late Miocene age for these depos- its. This age determination was based largely upon the identification o{ Globorotalia mayeri in the assemblages. On the basis of subsequent work on the ''Turborotalia" lineages, particularly by Blow (1969), these specimens would now be placed in "Turborotalia''^ acostaensis, a group not widely rec- ognized at that time. Even though the late Mio- cene age was applied to the lowermost part of the Yorktown, the age of the middle and upper part of the Yorktown Formation was considered un- settled at that time. Therefore, pending additional concrete evi- dence for a Pliocene age for at least part of the Yorktown, the strata were questionably assigned a late Miocene age. This state of affairs was reflected in comments such as those in Gibson (1971:10): 358 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY With the occurrence of relatively large changes in the faunas, both macro and micro, in the overlying Yorktown beds, it is entirely possible that units placed in the upper Yorktown will range considerably into the Pliocene in age, but this reassignment will be dependent upon finding ade- quate planktonic assemblages. Hazel (1971a) placed the youngest part ofthe Yorktown Formation in North Carolina, the beds along the Chowan River, into the early Pliocene. This was done on the assumption that the York- town Formation in Virginia was of late Miocene age, and that these younger beds along the Cho- wan (Mansfield, 1943) were, therefore, of early Pliocene age. How these beds correlated with the European stages of the upper Miocene and Pli- ocene was not demonstrated, except indirectly through strata in Florida. Hazel (1971a:8) stated: Thus, it may be that the upper Miocene-lower Pliocene boundary in the Atlantic and Gulf Coastal Plain eventually will be revised further downward to include in the Pliocene more of the deposits that have been traditionally assigned to the upper Miocene. Akers (1972) correlated the planktonic faunas from some Gulf and Atlantic Coastal Plain for- mations with planktonic zonations and stages in Europe. Akers (1972:32) concluded from the ex- amination of a few assemblages from the York- town Formation that it was of Pliocene age, correlating with Blow's (1969) zones N19 and N20, and that the Waccamaw Formation in North Carolina, containing Globorotalia truncatuli- noides, belonged to zone N22 of early Pleistocene age. Subsequent work by Akers and Koeppel (1973) on calcareous nannofossils supported a zone N20 date for the Yorktown Formation in Virginia and North Carolina. These age assign- ments also are supported by the present study. The Waccamaw and Croatan formations are herein considered coeval with the Caloosahatchee Formation, and are all considered late Pliocene and early Pleistocene in age. The new age assign- ments for these formations, along with that for the Yorktown, bring into agreement, at least at the epoch level, the ages of the strata and are based on both the invertebrates and the verte- brates. PLANKTONIC FORAMINIFERAL ZONATION Much of the current worldwide correlation with the European stratotypes is based upon the planktonic Foraminifera, although the use of cal- careous nannoplankton, diatom, radiolarian, and dinoflagellate correlations is increasing rapidly. Zonations based upon planktonic Foraminifera have been established in several different regions, for example by Bolli (1957) for Trinidad, Blow (1959) for Venezuela, and Jenkins (1960) for New Zealand. A summary of the land-based distribu- tions and some early Deep Sea Drilling Program samples were presented by Blow (1969), who then established a zonation that was based upon all available data. Blow named the zones, using di- agnostic or important species of planktonic For- aminifera, and also placed them in a numbered series, using N numbers for the later Cenozoic or Neogene zones and P numbers for the older Cen- ozoic or Paleogene zones. Although Jenkins and Orr (1972:1063) criticized the use of numbered zones as "unacceptable to existing stratigraphic codes," the use of the N and P zones as a form of shorthand has become widespread, although far from universal. Studies in nontropical waters, such as those of Poore and Berggren (1975) in the high latitudes of the northern Atlantic and Ken- nett (1973) in the cool-subtropical southwest Pa- cific, and even some workers in tropical areas, such as Jenkins and Orr (1972), had difficulty in recognizing some or many of Blow's zones and developed their own zonal sequence. Some at- tempt, however, was usually made to correlate the zones with Blow's (1969) sequence. The ranges of the stratigraphically important planktonic species generally are similar in the various oceanic and continental areas, but some relatively minor differences and a few major dif- ferences in the initial appearance or the extinction of species are found. The range discrepancies important to this study are noted under the spe- cies, where applicable. Studies of the planktonic Foraminifera found in the stratotype sections ofthe later Cenozoic in Europe and in the JOIDES cores have refined the NUMBER 53 359 species ranges during the past 10 years. Some European stages containing good marine faunas including planktonic Foraminifera have been used as alternatives to more commonly used stages containing faunas of restricted environ- ments. This is particularly true of the Sarmatian and Pontian stages, which were deposited in mar- ginal marine to brackish water environments. Most authors now attempt correlations with other stages containing adequate marine invertebrate assemblages. A better understanding of the ranges of plank- tonic Foraminifera has led to some changes in the stages to which zones are assigned. As a general rule, the zones have been shifted upward, either within the same stage or into a younger stage. An example of such a change, is the correlation of the Pungo River Formation of North Carolina. Gibson (1967) assigned these strata to the Globi- gerinatella insueta zone, then placed by Blow (1959:74) and Saito (1963, table 16) in the upper part of the Aquitanian. The present study sup- ports the correlation of the Pungo River Forma- tion with the G. insueta zone, but Berggren and Van Couvering (1974) placed this zone in the upper Burdigalian and lower Langhian stages. This study of the relationship of the planktonic zones to the stages in the later Cenozoic by Berg- gren and Van Couvering (1974), also integrated the radiometric dates with the sequence of plank- tonic foraminiferal zones, as shown in their fig- ure 1. The use of planktonic Foraminifera for corre- lation in the Miocene through Pleistocene strata ofthe Atlantic Coastal Plain is difficult, not only because planktonic specimens are extremely scarce (usually less than 1 percent), but also because most taxa that are present are cool-water species. Consequently it is difficult to recognize planktonic foraminiferal zones that are based on warm-water species. Most of the easternmost lo- calities of the Coastal Plain are subsurface mate- rial or artificial exposures like those found in the Pungo River Formation in the Lee Creek Mine. In some of these easternmost localities, planktonic specimens are more common, composing 10 per- cent or more of the foraminiferal assemblage, reflecting deeper water environments. The clock- wise circulation pattern of the North Atlantic does bring some warmer water faunal elements into the area, however, and a few specimens of important index species are found. Akers (1972) found two specimens of Globorotalia truncatulinoides in a large sample from Walkers Bluff in the Waccamaw Formation after 100 hours of exami- nation. Most samples in this study present similar problems because the critical species may be rep- resented by only one or several specimens at a locality. Although these low abundances are not desirable for biostratigraphic work, the use of rare specimens for zonation is the best means of in- terregional correlation at the present time. The scarcity of specimens and low reliability of cor- relation should be kept in mind when working with the placement of some of the strata in this area. An abundance of planktonic specimens at several localities yielded relatively large numbers of individuals belonging to critical species and thus provides fairly precise and reliable dates. Important localities mentioned in the text are shown in Figure 1. The interpretations ofthe ages of the strata are given in Gibson's figure 2 (p. 38 herein). A summary ofthe assemblages from each formation and the resulting age assignments are as follows. PUNGO RIVER FORMATION The only known outcrop of this formation is an artificial one in the Lee Creek Mine of Texas- gulf Inc. near Aurora, North Carolina (Figure 1, loc. 15). Additional faunas from the Pungo River Formation have been obtained from the Gates- ville Well in northeastern North Carolina and the Moores Bridge well near Norfolk in southeastern Virginia (Figure 1, loc. 11). In the Lee Creek Mine the Foraminifera are highly phosphatized in the phosphatic sands that are prevalent in the lower and middle part of the formation. Foraminifera and some mollusks, pri- marily calcitic pelecypods, are found in limy in- tervals in the upper 12 feet (3.7 m) ofthe forma- tion. 360 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Important planktonic foraminiferal species from the Pungo River strata in the Lee Creek Mine include the following: Cassigerinella chipolen- sis (Cushman and Ponton), Globigerina euapertura Jenkins, G. woodi woodi ]er\]dn?,, Globigerina species cf. G anguliofficinalis Blow, Globigerinoides altiaper- turus Bolli, Globorotalia peripheroronda Blow and Banner, G. scitula praescitula Blow, and ''^Turbo- rotalia" birnageae (Blow). The probable age for this assemblage is from the latter part of zone N7 to the latter part of zone N8 of Blow (1969) (Gibson's figure 2 on p. 38 herein). An age younger than N8 is not likely because of the presence of Globigerina euapertura and Globigerinoides altiaperturus, neither of which is known from strata younger than N8, and by the absence of Orbulina which is commonly regarded as marking the beginning of N9. The upper part of the Pungo River Formation in the Lee Creek Mine is thus considered to belong to planktonic zone N8, or latest early Miocene in age. The age of the underlying phosphatic sand units in the mine is unknown because ofthe poor preservation of the Foraminifera; the amount of time for and rapidity of deposition of the phosphatic sands and diatomaceous units is an intriguing, but pres- ently unanswerable, question. The northern limit of the Pungo River For- mation occurs in the Norfolk, Virginia, Moores Bridge well area (Figure 1, loc. 11) drilled by the U.S. Geological Survey. The strata yielded a foraminiferal assemblage similar to that found at the Lee Creek Mine. In addition, rare specimens of Globigerinatella insueta were found at a depth of 581 feet (177 m). The age of these strata also is considered to be zone N8. At this locality, the Pungo River Formation rests directly upon Pa- leogene rocks, indicating that the oldest age of the formation in this area is late early Miocene (N8). To the northeast of the Lee Creek Mine, Ab- bott and Ernissee (this volume) found strata con- taining diatoms; the lower part of the strata correlates with zone N8 or N9 and the upper part with zone Nil, indicating deposition of the younger strata of the Pungo River Formation close to the area of the Lee Greek Mine. As the top of the Pungo River section in the Lee Creek Mine has an erosional surface (Gibson, 1967:636), it is possible that the younger part of the forma- tion also was deposited there and later eroded. In summary, the distribution of strata assigned to the Pungo River Formation containing zone N8 planktonic Foraminifera is widespread, ex- tending from south of the Neuse River in North Carolina northward to the area near Norfolk, Virginia. Strata with zone Nl 1 planktonic assem- blages are found in the central part of the Albe- marle embayment in North Carolina. Strata of this age, however, may have been more extensive in the past, having been eroded in areas similar to the Lee Creek Mine and remaining unrecog- nized in other areas. Strata belonging to the intervening zones N9 and NIO have not been found. CALVERT FORMATION The only diagnostic planktonic foraminiferal assemblages in the Calvert Formation come from beds 10 to 12 of Shattuck (1904) in the Calvert Cliffs along Chesapeake Bay in Maryland. The most abundant faunas come from bed 10, whereas beds 1 to 9, and the uppermost beds, 13 to 15, yielded few specimens. The following important species are found in bed 10 in the Calvert Cliffs: Globigerina praebulloides pseudociperoensis Blow, Glo- bigerinoides altiaperturus Bolli, G. sicanus de Stefani, and Praeorbulina glomerosa glomerosa (Blow). These species indicate an age of upper zone N8 FIGURE 1—Location of collecting localities mentioned in text (MARYLAND: 1 = Randle Cliffs, 2 = Calvert Cliffs, 3 = Jones Wharf, 4 = Langleys Bluff, 5 = St. Marys River, 6 = Hammond Well. VIRGINIA: 7 = Yorktown, 8 = Cobham Wharf, 9 = Rice's Pit, 10 = Suffolk, 11 = Moores Bridge Well, Norfolk. NORTH CAROLINA: 12 = Murfreesboro, 13 = Gatesville Well, 14 = Chowan River outcrops, 15 = Lee Creek Mine, 16 = New Bern, 17 = James City, 18 = Croatan National Forest, 19 = Long Point, 20 = Barwick Farm, 21 = Natural Well, 22 = Walkers Bluff, 23 = Neils Eddy Landing, 24 = Acme, 25 = Pierce Brothers Quarry, 26 = Old Dock. SOUTH CAROLINA: 27 = Tillys Lake). NUMBER 53 361 60 KILOMETERS 362 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY to lower zone N9 for bed 10. The absence of Orbulina probably means an age of upper zone N8. As a large part of the Orbulina lineage is represented in the Calvert Cliffs, the absence of Orbulina, because of environmental causes, prob- ably can be discounted. In beds 11 and 12 ofthe Calvert Formation, specimens belonging to Praeorbulina glomerosa circularis and Orbulina suturalis appear, indicating a probable lower zone N9 placement for these strata. Beds 13 to 15 at the top of the Calvert Formation and beds 1 to 9 at the base do not contain diagnostic planktonic species, but the general similarity in the stage of evolution of molluscan species suggests closeness in age to the adjacent dated beds. In summary, the probable age for bed 10 ofthe Calvert Formation is zone N8, and that for the overlying beds 11 and 12 is probably of lower zone N9. The amount of time represented in the beds below bed 10, which includes a sequence of diatomaceous units 60 feet (18 m) thick, or more, is unknown at the present time. CHOPTANK FORMATION Planktonic specimens are extremely rare in the Choptank Formation. The only specimens found to date belong to long-ranging species of Globiger- ina, and correlation with Blow's planktonic zonal sequence is not possible. A tentative lower age limit for the formation was established on the following basis. The youngest age established for the Calvert Forma- tion in Maryland is lower zone N9 for bed 12. There is disagreement as to whether a significant unconformity exists between the Calvert and Choptank formations (Gernant, 1970:7). The ab- rupt appearance in the basal Choptank Forma- tion of several species of benthic Foraminifera that are absent in the uppermost part of the Calvert may indicate a time break (Gibson, 1962:13, 63). R.M. Forester (U.S. Geological Sur- vey, pers. comm., 1976) also noticed that some lineages of ostracodes change significantly across the Calvert-Choptank boundary, indicating a possible hiatus. For the present, the contact is considered unconformable, and the basal beds of the Choptank are tentatively dated as belonging in zone N12. Direct paleontological evidence is also inade- quate to determine the age of the upper beds of the Choptank Formation. A K/Ar age determi- nation of 12.0 ± 0.5 my from the overlying St. Marys Formation in Maryland was reported by Blackwelder and Ward (1976:5). This late middle Miocene date for the overlying St. Marys, if substantiated by further work, and the early mid- dle Miocene date for the top of the underlying Calvert places the Choptank in approximately the middle part of the middle Miocene (Gibson's figure 2 on p. 38 herein). ST. MARYS FORMATION No diagnostic planktonic foraminiferal species have been recognized among the rare specimens found in the St. Marys Formation. A K/Ar date of 12.0 ± 0.5 my (Blackwelder and Ward, 1976:5) for beds along the St. Marys River in Maryland is the only evidence on the age of this unit. The radiometric date places these beds in the upper part ofthe middle Miocene. The stratigraphically higher parts ofthe formation along the St. Marys River show sufficient change in the molluscan fauna to suggest the possibility of a younger, early late Miocene age (Gibson, in prep.). "VIRGINIA ST. MARYS" BEDS Diagnostic planktonic foraminiferal species have not been found in the "Virginia St. Marys" beds of Mansfield (1943) (see Gibson, 1971). The late middle to possible early late Miocene age for the underlying St. Marys Formation in Maryland and the early Pliocene date (planktonic zone N19) for overlying zone 1 beds of the Yorktown of Mansfield (1943) bracket these strata and indi- cate a late Miocene age. A K/Ar date of 8.7 ± 0.4 my for the middle part of the "Virginia St. Marys" sequence, reported by Blackwelder and Ward (1976:5), supports this late Miocene age assignment. Changes in the molluscan faunas NUMBER 53 363 through these strata, particularly the pectens (Gibson, in prep.), indicate that a significant amount of late Miocene time may be represented. Strata of the "Virginia St. Marys" sampled in northern North Carolina in a well near Gatesville in Gates County (Figure 1, loc. 13) contained moderately abundant planktonic specimens. Important species in the assemblage include Globorotalia merotumida Blow and Banner and "Turborotalia" acostaensis acostaensis Blow. Speci- mens approaching and within the range of vari- ation of Globorotalia plesiotumida Blow and Banner occur, but specimens of G. merotumida dominate. These species indicate a probable placement in zone N17 for the upper beds ofthe "Virginia St. Marys" in North Carolina. YORKTOWN FORMATION The Placopecten clintonius zone or zone 1 of the Yorktown Formation of Mansfield (1943) con- tains a moderately abundant planktonic assem- blage that includes some stratigraphically impor- tant species: Globigerina apertura Cushman, Globo- quadrina altispira altispira (Cushman and Jarvis), Globorotalia puncticulata (Deshayes), Globorotalia spe- cies cf. G. crassula Cushman and Stewart, Sphae- rodinellopsis seminulina seminulina (Schwager), S. sub- dehiscens subdehiscens (Blow), and ''''Turborotalia''' acostaensis humerosa (Takayanagi and Saito). The presence of Globorotalia puncticulata in these assemblages suggests a placement of zone N19 or later (see p. 373). The upper range of Sphaerodi- nellopsis seminulina seminulina in the Atlantic is considered to be the upper part of zone N20 (Poag, 1972b:492, 493, 499; Berggren and Am- durer, 1973:353, figs. 4, 7, 8). Blow (1969:338) reported an upper range of zone N20 in land- based sequences, but into lower zone N21 in deep- sea sequences. If these data are correct, this sub- species is indicative of zone N20, or older, age for the Placopecten clintonius zone of Mansfield (1943). Additional support comes from the occurrence of Globigerina apertura, which has a reported upper range of zone N20. Globoquadrina altispira altispira was considered by Blow (1969:339) to range into the lower part of zone N21 in oceanic deposits, but was probably not younger than lower zone N20 in land-based sequences. An upper range of lower zone N21 is commonly found for Sphaero- dinellopsis subdehiscens subdehiscens also. The pres- ence oV'Turborotalia" acostaensis in the assemblage also indicates an early Pliocene age. Parker (1967:165) reported that species as ranging to the end of zone N18, but Jenkins and Orr (1972:1066, 1096) extended its range into zones N19/20, and Poag (1972b:485, 511) into zone N21. This assem- blage appears to be indicative of an early Pliocene zone N19/20 age, and this is the age assigned to the Placopecten clintonius zone. A radiometric date of 4.4 ±0.2 my was obtained from Yorktown strata immediately overlying this zone (Black- welder and Ward, 1978:8), and this age closely corresponds to the planktonic foraminiferal as- signment. The overlying Yorktown strata of the Turritella alticostata zone or zone 2 of Mansfield (1943) and the Duplin Formation, the equivalent of the up- per part of the Yorktown in southern North Carolina and southward, are younger. The cor- relation of these strata with Blow's zonation is difficult at present because of the scarcity of biostratigraphically diagnostic planktonic species. Most of the planktonic specimens in these strata belong to the long-ranging species Globigerina bul- loides, Globigerinoides trilobus, and G. ruber. Among the few species that have shorter ranges is Globi- gerina apertura, which has a reported upper range through zone N20. The type area for this species is near Suffolk, Virginia, in the Yorktown deposits that Mansfield (1943) placed high in his zone 2. Subsequent work has supported this placement. Globorotalia puncticulata, which has an upper range of zone N21, also occurs in strata of Mansfield's zone 2. Specimens of G. puncticulata from zone 2 show characteristics transitional to those of ^^ Turborotalia'^ inflata, but specimens of the latter species are absent from zone 2. ''^Turborotalia''' inflata is indicative of zone 21 and younger ages. The occurrence of G. puncticulata and the absence of ^''Turborotalia''' inflata, although meager evi- dence, could indicate a late zone N19/20 to early 364 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY zone N21 age for the upper part of Mansfield's zone 2 of the Yorktown. Only the occurrence of Globorotalia hirsuta hirsuta in strata of zone 2 at the type area ofthe Yorktown Formation in the bluffs at Yorktown, Virginia, conflicts with this zonal assignment. This subspecies is usually considered characteristic of zone N22 and younger strata, but here occurs in strata tentatively assigned to late zone N20 to early zone N21. The specimens appear to fall well within the range of variation of this subspecies (Plate 3: figures 5-7). UPPERMOST "YORKTOWN," CROATAN, AND WACCAMAW FORMATIONS The uppermost part of the "Yorktown" For- mation, which is exposed along the Chowan River in northeastern North Carolina, the Croa- tan Formation at the Lee Creek Mine and south- east of New Bern, North Carolina, and the Wac- camaw Formation in southern North Carolina and northern South Carolina appear to be of approximately equivalent age (Gibson's Figure 2, p. 38 herein). This correlation is based upon the stratigraphic distribution of pectens (Gibson, in prep.) and benthic Foraminifera (Figure 2). Planktonic Foraminifera are rare in these forma- tions because of the very shallow water environ- ment of deposition. Akers (1972:36, 38) reported rare specimens of''^ Turborotalia'^ inflata and Globor- otalia truncatulinoides from the Waccamaw Forma- tion at Walkers Bluff. The first appearance of G. truncatulinoides commonly is used to mark the base of the Pleistocene, although there is some contro- versy about how close to the boundary it occurs. In the upper part of the Croatan Formation at the Lee Creek Mine, a single specimen strongly resembling G. truncatulinoides was found. At least these parts of the formations apparently belong to the lower Pleistocene or zone N22. Evidence that these deposits are no younger than zone N22 is the occurrence of Globigerinoides obliquus (Akers, 1972:34). Benthic foraminiferal species characteristic of zone 2 of the Yorktown Formation are found in the lower part of the Croatan Formation in the Lee Creek Mine, but they are not found in the upper part of the section. These faunal changes suggest that the lower part of the Croatan is somewhat older than the zone N22 assignment for the upper part. Because no specimens of G. truncatulinoides have been found in the lower beds, and in considera- tion of the changes in the benthic assemblages, the lower beds are placed in the upper Pliocene zone N21. Thus, the Croatan Formation and equivalent units appear to range in age from late Pliocene into early Pleistocene. The upper Pli- ocene strata may correlate with the Bear Bluff Formation of DuBar et al. (1974) as suggested by Hazel (1977:375). The younger strata, tentatively assigned to zone N22 of early Pleistocene age, on the basis of planktonic foraminiferal evidence, presently are recognized only in the upper part of the Croatan Formation in the Lee Creek Mine and in the Waccamaw Formation in southern North Carolina. Evidence from the pectens (Gib- son, in prep.) indicates that the uppermost "Yorktown" beds at Mt. Gould along the Cho- wan River also may be included in this zone. Key Benthic Species Although benthic foraminiferal species gener- ally have greater ranges and are more closely controlled environmentally than their planktonic counterparts, some benthic species were recog- nized by Gibson (1962) to be widespread through- out only one or several of the formations in this area. These species are useful guides, particularly in the subsurface where molluscan data are largely unavailable. Examination of several hundred samples in the Atlantic Coastal Plain from Maryland to North Carolina, both from outcrops and the subsurface, indicates that a number of benthic species are each characteristic of the upper Oligocene, Miocene, Pliocene, and lower Pleistocene formations in this area. Some diagnostic species are known only from the unit(s) in this area; others are found in coeval deposits in other areas in the Western Hemisphere or else- where; and still others have a restricted range here but broader ranges in other areas. The par- o ^ < m 7^ O 3D • > H d > o z O -< z CO Spiroplectammina mississippiensis Textulana obliqua Texiulana ultima-inHata Massilina glutinosa lilassilina maryli, yirqulinella Spiroplectammina e'xilis Bolivina pungoensis Bolivinopsi^airhavenensis Bolivina calvertensis Sagnna pulchella primivva Hopkinsina bononiensis Siphogenenna lamellata Rosalina calvernata Epistominella danvillensis Epistominella pungoensis Hotorbinella bassleri Cibicides cravenensis Cibicides pungoensis Cibicides croatanensis Svraiklna croatanensis Nonion advenum pustulosum marylandicum Nonion calvertensis Astrononion stelligerum Florilus chesapeakensis Elphidium excavatum Elphidium latispatium pontium Elphidium neocrespinae Elphidium compressulum Elphidium gunleri Elphidium limatulum FIGURE 2.—Distribution of key benthic foraminiferal species in upper Oligocene through lower Pleistocene strata in the central Atlantic Coastal Plain. 366 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY ticular situation for each species is noted in the systematic section. Hazel (1977) noted the sig- nificant ostracodes for the Pliocene and early Pleistocene in this area. Some species are present in both surface and subsurface samples, but others are found mostly only in one or the other. This results from the general eastward dip of the strata and increasing depth of deposition in the strata to the east. Thus the shallower water environments are generally found farther west in the outcrop belt. The gen- eral climatic pattern in the area was one of mild temperate conditions and some warmer fluctua- tions through the Miocene and earliest Pliocene. A general warming trend began in the early Pliocene and extended through late Pliocene and earliest Pleistocene time, when warm temperate to subtropical conditions were reached in south- ern North Carolina (Gibson, 1967:647; Hazel, 1971b:373). The warming trend allowed some species to make their first appearance at a later time in this area than in warmer areas to the south. Thus, some key species in this area have greater ranges in Florida and in other warm regions. Many of the key species are relatively common in the assemblages, greater than 5 percent, but some are rare, making up less than 1 percent of the assemblage, as noted in the systematic section. Some key species diagnostic ofthe deeper water environments of the formations in the subsurface of the Coastal Plain may aid in correlation with strata on the Continental Shelf, and serve as a bridge between the shallower water outcrop and deeper water offshore sections. Species diagnostic of each formation or part thereof are shown in Figure 2. A general distri- bution is given (pp. 360-364), and detailed dis- tributions are discussed in the systematics section. the two new subspecies, Nonion advenum pustulosum, restricted to this unit, and Elphidium latispatium pontium, which occurs in this unit and again in the younger St. Marys Formation, and Discorbitura dignata, which is not found above this unit. CALVERT AND PUNGO RIVER FORMATIONS Because of the large geographic area covered by the two formations and the variety of environ- ments represented, distributional patterns of the key species are variable. Bolivina calvertensis, Boli- vinopsis fairhavenensis, new species, and Epistominella pungoensis, new species, are restricted to these two formations. Spiroplectammina mississippiensis occurs only in these two formations in this area, but is found in other regions. Rosalina cavernata and Non- ion calvertensis, new species, are found only in the Calvert Formation. Bolivina pungoensis, new spe- cies, and Cibicides cravenensis, new species, are found only in the lower part of the Pungo River Formation. Virgulinella miocenica, Rotorbinella bas- sleri, and Florilus chesapeakensis, new species, range upwards from these formations, but serve to dif- ferentiate them from the underlying Oligocene units and, in those places where the Choptank and St. Marys formations are absent, from var- ious parts of the overlying Yorktown Formation. CHOPTANK FORMATION Because of the presently relatively restricted geographic distribution of this formation (Gibson, 1971) and generally similar shallow-water envi- ronments, the key species are found in most sam- ples. They include Spiroplectammina exilis, Textu- laria ultima-inflata, Massilina glutinosa, and Nonion marylandicum, although the latter is also found rarely in the upper part of the Calvert. SILVERDALE BEDS OF VOKES.(1967) These onshore strata represent very shallow marine environments, where a few species domi- nate the assemblages. Fortunately, the dominant species are diagnostic for this unit. They include ST. MARYS FORMATION Most assemblages from the St. Marys Forma- tion are composed of species that are found also in the underlying formations and in the lower part of the overlying Yorktown Formation, mak- NUMBER 53 367 ing it difficult to recognize this formation in the subsurface. However, the presence of Textularia obliqua, Massilina marylandica, and Elphidium latis- patium pontium, new subspecies, and the absence of the common key species in the Choptank serve to identify the unit in most samples. "VIRGINIA ST. MARYS" BEDS One important subspecies that first appears in the "Virginia St. Marys" strata and still lives today is Elphidium excavatum clavatum. This subspe- cies appears abruptly and is so abundant (form- ing 20 to 40 percent of the assemblages) that it is a valuable guide for separating these strata from earlier ones. The co-occurrence of this species with Virgulinella miocenica marks the "Virginia St. Marys." Hopkinsina bononiensis and Cibicides pun- goensis, new species, are only found in the "Virginia St. Marys" in the subsurface of north- eastern North Carolina. YORKTOWN FORMATION AND EQUIVALENTS This formation spans both considerable time and various environments, resulting in significant geographic changes in the benthic assemblages. Although some species are restricted stratigraph- ically to various parts of the formation, most appear in the underlying formations and are living today. Zone 1 of Mansfield (1943) is char- acterized by the first appearance of Textularia mayori, Quinqueloculina lamarckiana, Nodosaria cates- byi, Epistominella danvillensis, and Astrononion stelli- gerium, along with the co-occurrence o(Rotorbinella bassleri. Zone 2 of Mansfield (1943) is marked by the restricted occurrence in this area of Bolivina marginata multicostata, the last appearance of No- dosaria catesbyi and Epistominella danvillensis, and the first occurrence of Sagrina pulchella primitiva. The youngest beds of the "Yorktown" along the Chowan River, as well as the coeval Croatan and Waccamaw formations to the south, contain a number of species restricted to those strata, in- cluding three new species, Cibicides croatanensis, Svratkina croatanensis, and Elphidium neocrespinae, and Elphidium compressulum and E. limatulum, all of which are found in these strata and in the Duplin Formation. Elphidium gunteri is restricted to these units, although it has a greater range in warmer environments. Systematic Descriptions Generic concepts as applied to planktonic For- aminifera are unsettled at this time. Some of the common generic names, such as Globorotalia, have been applied to several groups of species that occur at significantly different times in the fossil record and have no known connecting species; this leads one to doubt strongly any phylogenetic linkage between the groups. Other genera, such as Turborotalia, are used as morphologic types (usually on the basis of one or possibly several distinctive features) that occur at different times in a number of apparently separate lineages, which means that they are definitely polyphy- letic. Even the commonest name, Globigerina, was considered by Fleisher (1974:1009, 1018) to in- clude a number of polyphyletic units, which he discriminated largely on the basis of wall texture. Pending further study and revision, most of the generic names currently used can be considered convenient holding names for an uncertain num- ber of species. Illuminating discussions on the generic problems are found in Fleisher (1974), Parker (1967), and Stainforth et al. (1975). Placement of species of benthic Foraminifera here follows a classification of genera and families modified from Loeblich and Tappan (1964). The rigid application by Loeblich and Tappan of wall structure as the dominant character in higher level foraminiferal systematics has led to the placement of apparently closely related species in different genera and families. Towe and Cifelli (1967) showed the difficulties in applying optical studies of wall structure, such as those of Loeblich and Tappan, to electron microscope studies ofthe wall. Buzas (1965, 1966) found both granular and radial wall structure in species of Elphidium, as did Hansen (1972a) in Turrilina and Feyling-Han- sen and Buzas (1976) in Cassidulina. Information 368 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY from studies such as these indicates that group- ings and separations based solely upon wall struc- ture should be revised. Order FORAMINIFERIDA Eichwald, 1830 Superfamily GLOBIGERINACEA Carpenter, Parker, and Jones, 1862 Family GLOBIGERINIDAE Carpenter, 1862 Genus Cassigerinella Pokorny, 1955 Genus Globigerina d'Orbigny, 1826 Genus Globigerinatella Cushman and Stainforth, 1945 Genus Globigerinita Bronnimann, 1951 Genus Globigerinoides Cushman, 1927 Genus Globoquadrina Finlay, 1947 Genus Globorotalia Cushman, 1927 Genus Orbulina d'Orbigny, 1839 Genus Praeorbulina Olsson, 1964 Genus Pulleniatina Cushman, 1927 Genus Sphaeroidinellopsis Banner and Blow, 1959 Family GLOBOROTALIIDAE Cushman, 1927 Genus Turborotalia Cushman and Bermudez, 1949 Superfamily LITUOLACEA de Blainville, 1825 Family TEXTULARIIDAE Ehrenberg, 1838 Genus Spiroplectammina Cushman, 1927 Genus Textularia Defrance, 1824 Genus Bolivinopsis Yakovlev, 1891 Superfamily MILIOLACEA Ehrenberg, 1839 Family MILIOLIDAE Ehrenberg, 1839 Genus Quinqueloculina d'Orbigny, 1826 Genus Massilina Schlumberger, 1893 Superfamily NODOSARIACEA Ehrenberg, 1838 Family NODOSARIIDAE Ehrenberg, 1838 Genus Nodosaria Lamarck, 1812 Superfamily BULIMINACEA Jones, 1875 Family BOLIVINITIDAE Cushman, 1927 Genus Bolivina d'Orbigny, 1839 Family UVIGERINIDAE Haeckel, 1894 Genus Hopkinsina Howe and Wallace, 1932 Genus Sagrina d'Orbigny, 1839 Genus Siphogenenna Schlumberger, 1882 Superfamily DISCORBACEA Ehrenberg, 1838 Family DISCORBIDAE Ehrenberg, 1838 Genus Rotorbinella Bandy, 1944 Genus Epistominella Husezima and Maruhasi, 1944 Genus Rosalina d'Orbigny, 1826 Genus Cancris Montfort, 1808 Superfamily ROTALIACEA Ehrenberg, 1839 Family ELPHIDIIDAE Galloway, 1933 Genus Elphidium Montfort, 1808 Superfamily ORBITOIDACEA Schwager, 1876 Family CIBICIDIDAE Cushman, 1927 Genus Cibicides Montfort, 1808 Superfamily CASSIDULINACEA d'Orbigny, 1839 Family CAUCASINIDAE Bykova, 1959 Genus Virgulinella Cushman, 1932 Family NONIONIDAE Schultze, 1854 Genus Astrononion Cushman and Edwards, 1937 Genus Florilus Montfort, 1808 Genus Nonion Montfort, 1808 Family ALABAMINIDAE Hoflcer, 1951 Genus Svratkina Pokorny, 1956 Genus Cassigerinella Pokorny, 1955 Cassigerinella chipolensis (Cushman and Ponton) PLATE 4: FIGURE 16 Cassidulina chipolensis Cushman and Ponton, 1932:98, pi. 15: figs. 2a-c.—Jenkins, 1971:73-74, pi. 1: fig. 30. OCCURRENCE.—Rare specimens of this species are found in the upper beds of the Pungo River Formation in the Lee Creek Mine in North Car- olina. STRATIGRAPHIC RANGE.—Blow (1969:377) re- ported a relatively long range for this species, from zone P18 (early Oligocene) to zone N13 (middle Miocene). Genus Globigerina d'Orbigny, 1826 REMARKS.—A large variety of species has been placed in this genus by various authors; the pri- mary unifying characteristic is the umbilical po- sition ofthe aperture. Fleisher (1974:1009-1012, 1018-1019) discussed the modification and split- ting of this group. Fleisher's suggestions for dif- ferent groups within those species assigned to Globigerina are based primarily, but not exclu- sively, upon the nature ofthe wall, which he felt reflected the phylogeny. Fleisher's concept of Glo- bigerina, sensu stricto, would restrict the use of this genus to a fraction ofthe various species presently placed in the genus. Until more extensive studies are made, however, Globigerina will be used for those species that have an umbilical aperture and have traditionally been placed in this genus. Globigerina apertura Cushman PLATE 1: FIGURE 10 Globigerina apertura Cushman, 1918:57, pi. 12: figs. 8a-c.— Zachariasse, 1975:119-120, pi. 16: figs. 1-2. NUMBER 53 369 Globigerina bulloides apertura Cushman.—Blow, 1969:317, pi. 12: fig. 8. REMARKS.—Globigerina apertura is distinguished from G. bulloides by a larger, rimmed aperture that is more centrally umbilicate, and by having more appressed chambers and a coarsely cancel- late wall (Fleisher, 1974:1019; Zachariasse, 1975). The species was originally described from the Yorktown Formation in Virginia. OCCURRENCE.—Rare in the Yorktown Forma- tion in outcrops near Suffolk, Virginia, the type area for the species; in the upper part of the sequence exposed along the Meherrin River near Murfreesboro, North Carolina; and in the Nor- folk, Virginia, Moores Bridge Well core at a depth of 113 feet (34.4 m). Akers (1972:30) reported it from the Yorktown Formation at Rice's Pit, Hampton, Virginia, and Copeland (1964:281) reported it from Natural Well and Barwick Farm in the Duplin Formation in North Carolina (Fig- ure l,loc. 9, 20, 21). STRATIGRAPHIC RANGE.—Blow (1969:317) re- ported a range from zone N16 to zone N19. Jenkins and Orr (1972:1086) recorded it from the equivalent of zone N18 to the upper part of zone N20. Globigerina praebulloides pseudociperoensis Blow PLATE 1: FIGURES 7-9 Globigerina praebulloides pseudociperoensis Blow, 1969:381-382, pi. 17: figs. 8-9. OCCURRENCE.—Rare to common in bed 10 of the Calvert Formation at Plum Point, Maryland. STRATIGRAPHIC RANGE.—Blow (1969:321) re- ported a range from zone N7 to zone N12. Globigerina woodi woodi Jenkins PLATE 4: FIGURES 9-11 Globigerina woodi Jenk'ms, 1960:352, pi. 2: figs. 2a-c. Globigerina (Globigerina) woodi woodi Jenkins.—^Jenkins, 1971: 159-160, pi. 18: figs. 548-550. OCCURRENCE.—Rare in the upper part of the Pungo River Formation in the Lee Creek Mine, North Carolina. STRATIGRAPHIC RANGE.—^Jenkins and Orr (1972:1090) reported a range equivalent to zone N4 to zone N17. Kennett (1973:578, 583, 584, 588) recorded its upper limit in the lower Pleis- tocene, N22. Globigerina euapertura Jenkins PLATE 4: FIGURE 15 Globigerina euapertura Jenkins, 1960:351, pi. 1: figs. 8a-c; 1971:147, pi. 15: figs. 457-461: pi. 16, fig. 462.—Jenkins and Orr, 1972:1088, pi. 9: figs. 1-6. Turborotalia (Turborotalia) euapertura (Jenkins).—Fleisher, 1974:1035. OCCURRENCE.—Rare in the Pungo River For- mation in the Lee Creek Mine, North Carolina. STRATIGRAPHIC RANGE.—^Jenkins and Orr (1972:1088) recorded a range for this species cor- responding to zone P18 to zone N3, and smaller specimens {G. cf. euapertura) extending to the equivalent of zone N6. Fleisher (1974) reported this species from zone P22 to zone N7-N8; the upper limit is similar to the age of the present occurrence. Globigerina species cf. G. anguliofHcinalis Blow PLATE 4: FIGURES 13, 14 REMARKS.—Specimens from the Pungo River Formation are similar to G. anguliofficinalis Blow, 1969, in having ^V2 chambers in the last whorl, incised intercameral sutures, and a similar ap- pearance of the wall. Pungo River specimens differ in having a narrower and shallower umbil- icus and a lower arched aperture. In view of these differences between the Pungo River specimens and those illustrated by Blow, the specimens from the Pungo River are not included within the species, although there appears to be a close relationship. OCCURRENCE.—Rare in the upper beds of the Pungo River Formation in the Lee Creek Mine, North Carolina. STRATIGRAPHIC RANGE.—The specimens from 370 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY the Pungo River Formation occur in rocks that are assigned to zone N8 on the basis of the co- occurrence of other planktonic species. G. angu- liofficinalis, according to Blow (1969:315), has a range from zone PI7 to zone N2 and is thus restricted to the Oligocene. OCCURRENCE.—Rare in the upper part of the Pungo River Formation in the Lee Creek Mine and in the Yorktown Formation in North Caro- lina. STRATIGRAPHIC RANGE.—Blow (1969:327) re- ported a range of zone N5 to zone N23. Genus Globigerinatella Cushman and Stainforth, 1945 Globigerinatella insueta Cushman and Stainforth PLATE 6: FIGURE 17 Globigerinatella insueta Cushman and Stainforth, 1945:69, pi. 13: figs. 7-9.—Blow, 1969:330, pi. 26: figs. 1-7. REMARKS.—Only a single specimen was found. It has a series of separated bullae and appears to fit within the range of variation described by Bronnimann (1950). It is very similar to the spec- imen illustrated by Bronnimann and Resig (1971, pi. 21: fig. 1). OCCURRENCE.—The single specimen was found in the Pungo River Formation in the Norfolk, Virginia, Moores Bridge Well, at a depth of 581 feet (177m). STRATIGRAPHIC RANGE.—Blow (1969:330) re- ported a range from the beginning of zone N6 to the lower part of zone N9. Genus Globigerinoides Cushman, 1927 Globigerinoides altiaperturus Bolli PLATE 1: FIGURES 1, 2; PLATE 4: FIGURES 7, 8 Globigerinoides triloba altiapertura Bolli, 1957:113, pi. 25: figs. 7a-8. Globigerinoides altiaperturus Bolli.—^Jenkins, 1971:174-175, pi. 20: figs. 604-606. OCCURRENCE.—Rare in the upper part of the Pungo River Formation in the Lee Creek Mine, North Carolina, and in bed 10 of the Calvert Formation, Plum Point, Maryland. STRATIGRAPHIC RANGE.—Blow (r969:325) re- ported a range from zone N5 to the lower parts of zone N7. Bronnimann and Resig (1971:1441) recorded a range from zone N4 to zones N7/8, and it appears from this and other works (Fleisher, 1974:1023) that this species ranges into zone N8. The occurrence in bed 10 ofthe Calvert Formation indicates a range into upper zone N8 or lower zone N9 in this area. Genus Globigerinita Bronnimann, 1951 Globigerinita glutinata ambitacrena (Loeblich and Tappan) PLATE 1: FIGURES 11-13 Tinophodella ambitacrena Loeblich and Tappan, 1957:114, figs. 2a-3c. Globigerinita glutinata ambitacrena (Loeblich and Tappan).— Fleisher, 1974:1022, pi. 9: fig. 3. REMARKS.—Following the usage of Fleisher (1974), which is compatible with my observations on limited material of Miocene to Holocene age, forms with a simple bulla are placed in the "subspecies" G. glutinata ambitacrena. Globigerinoides sicanus de Stefani PLATE 1: FIGURES 4, 5 Globigerinoides conglobata (Brady).—Cushman and Stainforth, 1945:68, pi. 13: fig. 6. Globigerinoides sicana de Stefani, 1952:9.—Fleisher, 1974: 1024, pi. 9: fig. 10. Globigerinoides bispherica Todd, 1954:681, pi. 1: figs. la-c. Globigerinoides sicanus de Stefani.—Blow, 1969:326. OCCURRENCE.—Rare in bed 10 of the Calvert Formation at Plum Point, Maryland. STRATIGRAPHIC RANGE.—Blow (1969:327) re- ported that the range was from the beginning of zone N8 to within the lower part of zone N9. Bronnimann and Resig (1971:1251, 1441) re- corded it as low as the top of zone N6 in some cores in the Pacific Ocean. NUMBER 53 371 Globigerinoides trilobus trilobus (Reuss) PLATE 4: FIGURE 12 Globigerina triloba Reuss, 1850:374, pi. 47: figs, lla-d. Globigerinoides triloba triloba (Reuss).—Bolli, 1957:112-113, pi. 25: figs. 2a-c. Globigerinoides trilobus trilobus (Reuss).—^Jenkins, 1971:180- 182, pi. 19: figs. 571-581. OCCURRENCE.—Rare to common in the Calvert Formation in Maryland, the Pungo River For- mation in Virginia and North Carolina, the York- town Formation in Virginia and North Carolina, and the Waccamaw and Croatan formations in North Carolina. STRATIGRAPHIC RANGE.—Blow (1969:326) re- ported a range from zone N6 to zone N23. Genus Globoquadrina Finlay, 1947 Globoquadrina altispira altispira (Cushman and Jarvis) PLATE 2: FIGURES 4, 7, 8 Globigerina altispira Cushman and Jarvis, 1936:5, pi. 1: figs. 13a-c, 14. Globoquadrina altispira altispira (Cushman and Jarvis).—Bolli, 1957:111, pi. 24: figs. 7a-8b.—Stainforth et al., 1975:245, fig. 100. OCCURRENCE.—Rare in bed 10 of the Calvert Formation in Maryland, and in the lower part of the Yorktown Formation along the bluffs of the Meherrin River near Murfreesboro, North Caro- lina. Dorsey (1948:313, fig. 28) reported rare oc- currences in beds 10 to 12 ofthe Calvert Forma- tion and bed 16 of the Choptank Formation in Maryland. STRATIGRAPHIC RANGE.—Blow (1969:339) re- ported a range from zone N4 to the lower part of zone N20 and a probable range in deep oceanic areas into part of zone N21. Upon examination of many land-based sequences, Blow (1969:255) considered this species to have become extinct in the lower part of zone N20 during shallow-water sequences found on the continents. In the deep- ocean sequences sampled by JOIDES, later extinc- tion horizons than early N20 have been found, as mentioned by Blow. In the Pacific, Bronnimann and Resig (1971:1437) recorded this species from the middle of zone N20, Parker (1967:165) and Jenkins and Orr (1972:1094), the lower part of N21, and Kennett (1973:578, 580, 583, 584, 589, 596), the top of N21. In the North Atlantic, Poag (1972b:504, 514) found this species into the lower part of N21. Genus Globorotalia Cushman, 1927 As discussed by Fleisher (1974:1009-1012, 1025-1026), the genus Globorotalia has been used to include a diverse multitude of forms. The general usage of Fleisher is followed here, but not the subgeneric groupings, as there is still a need for lineage studies to demarcate the different stocks clearly. Globorotalia hirsuta hirsuta (d'Orbigny) PLATE 3: FIGURES 5-7; PLATE 5: FIGURES 9, 10 Rotalina hirsuta d'Orbigny, 1839b: 131, pi. 1: figs. 37-39. Globorotalia hirsuta hirsuta (d'Orbigny).—Blow, 1969: 398-400, pi. 8: figs. 1-3, pi. 43: figs. 1-2. OCCURRENCE.—Rare in the Yorktown Forma- tion in the type area at the cliffs at Yorktown, Virginia, and in the Croatan Formation in the Lee Creek Mine. STRATIGRAPHIC RANGE.—Blow (1969:363) re- ported a range from zone N22 to zone N23, and Bronnimann and Resig (1971:1433) recorded this species from the middle of zone N22 into the Holocene. Most other records of older ranges, such as Parker's (1967:178), from the middle of N20 upward, and Jenkins' and Orr's (1972:1099), from N19/20 upward, have to be studied with caution as these authors did not differentiate the subspecies proposed by Blow (1969:398-402). Globorotalia menardii (Parker, Jones, and Brady) PLATE 3: FIGURES 1-3 Rotalia (Rotalie) menardii d'Orbigny, 1826:273, model no. 10 [nomen nudum]. Rotalia menardii Parker, Jones, and Brady, 1865:20, pi. 3: fig. 81.—Banner and Blow, 1960:31-33, pi. 6: figs. 2a-c. 372 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Globorotalia menardii (Parker, Jones, and Brady).—^Jenkins, 1971:90, pi. 6: figs. 135-137. REMARKS.—The illustrated specimen and oth- ers from this area closely resemble G. menardii form 5 of Tjalsma (1971:60, pi. 6, figs. 3a-5c), particularly those specimens referred to this taxon by Zachariasse (1975, pis. 3, 4), although the proximal ends of the intercameral sutures are perpendicular or slightly acute to the spiral suture rather than slightly oblique. OCCURRENCE.—Rare in the Yorktown Forma- tion in Virginia and North Carolina and in the Duplin Formation in North and South Carolina. Akers (1972:36) reported it from the Waccamaw Formation at Walkers Bluff, North Carolina. STRATIGRAPHIC RANGE.—Blow (1969:359) re- ported a range from zone N14 to zone N23. Globorotalia merotumida Blow and Banner PLATE 6: FIGURES 1-13 Globorotalia (Globorotalia) merotumida Blow and Banner, in Banner and Blow, 1965:1352, fig. la-c. REMARKS.—Most specimens in the assemblages have the relatively slow increase in whorl height and uniformly enlarging chambers characteristic of G merotumida. Some specimens, however, have a much greater increase in whorl height and likely fall within the range of variation of G. plesiotumida. OCCURRENCE.—Common to rare in the "Virginia St. Marys" at depths from 131.5 to 138 feet (40 to 42 m) in the Gatesville Well, North Carolina. STRATIGRAPHIC RANGE.—Blow (1969:364) re- ported a range from just above the base of zone N16 to within zone N18. Globorotalia minima Akers PLATE 6: FIGURES 14-16 Globorotalia carariensis (d'Orbigny) var. minima Akers, 1955: 659, pi. 65: fig. 3a-d. Globorotalia minima Akers.—Blow, 1959:217-218, pi. 19: figs. 122a-c. OCCURRENCE.—Common to rare in the Pungo River Formation at depths from 131.5 to 138 feet (40 to 42 m) in the Gatesville Well, North Caro- lina. STRATIGRAPHIC RANGE.—Blow (1969:352) re- ported a range from the middle of zone N7 to within zone N13. Bronnimann and Resig (1971:1433, 1441) recorded this species from the lower part of zone N6 to within zone N13. Globorotalia peripheroronda Blow and Banner PLATE 5: FIGURES 5, 6 Globorotalia {Turborotalia) peripheroronda Blow and Banner, 1966:294, pi. 1: fig. la-c, pi. 2: figs. 1-3. OCCURRENCE. Rare to common in the upper beds of the Pungo River Formation in the Lee Creek Mine in North Carolina. STRATIGRAPHIC RANGE.—Blow (1969:354) re- ported a range from within zone N6 to within zone Nil. Bronnimann and Resig (1971:1441) recorded the species as early as zone N5. Globorotalia puncticulata (Deshayes) PLATE 3: FIGURES 9-11; PLATE 5: FIGURES 11, 12 Globigerina puncticulata Deshayes, 1832:170.—Banner and Blow, 1960:15-17, pi. 5: fig. 7a-c. Globorotalia puncticulata (Deshayes).—Zachariasse, 1975:114- 115, pi. 14: fig. 2a-c. REMARKS.—The specimens from the lower part ofthe Yorktown Formation (zone 1 of Mansfield) are well-developed members of this species, as illustrated in Plates 3 and 5. Specimens from the middle part of the Yorktown Formation (zone 2 of Mansfield) approach Turborotalia inflata in chamber arrangement and outline of the periph- ery. OCCURRENCE.—Specimens occur in the lower part ofthe Yorktown Formation (zone 1 of Mans- field) along the Meherrin River near Murfrees- boro, North Carolina, and in the Lee Creek Mine, North Carolina, and in the middle and upper parts ofthe Yorktown Formation in Virginia and North Carolina. The species also occurs in the NUMBER 53 373 Duplin and Waccamaw formations in North Car- olina. STRATIGRAPHIC RANGE.—Blow (1969:354) re- ported a range from the upper part of zone N19 through zone N23. However, most later authors found the first appearance essentially at the base of zone N19 (Berggren, 1972:970, 973; Cita and Gartner, 1973:530, 533, 536; Kennett, 1973:587; Gradstein, 1974:68,99; Zacharriasse, 1975:30-31, 37; Poore and Berggren, 1975:273, 278, 282). The upper limit of the species is uncertain. In the Mediterranean region it disappears within zone N20 (Gradstein, 1974:68, 99; Zacharriasse, 1975:30-31, 37). Kennett (1973:587) marked its extinction in the middle of zone N21 in the Pacific Ocean sequences. Berggren and Amdurer (1973, fig. 10) and Blow (1969:354) found it through zone N23. Poore and Berggren (1975:273, 278) showed an upper limit in the latest Pliocene or earliest Pleistocene (probably equivalent to up- permost N21 or lower N22). Cifelli (pers. comm., 1975) noted that G. puncticulata, in the form rec- ognized by the above authors, does not occur today in the plankton in the North Atlantic Ocean, and that forms previosuly placed under that name belong to what is now called G. cras- siformis. If G. puncticulata did become extinct some- time during the Pliocene or early Pleistocene as stated by various authors, then the occurrence of the type suite of Deshayes and d'Orbigny at Rimini (Banner and Blow, 1960:15) requires re- classification of fossil specimens. Globorotalia scitula praescitula Blow PLATE 5: FIGURES 1-4 Globorotalia scitula praescitula Blow, 1959:221, pi. 19: figs. 128a-c. Globorotalia (Turborotalia) scitula praescitula Blow.—Blow, 1969:356, pi. 4: figs. 21-23, pi. 39: fig. 9. OCCURRENCE.—Rare to common in the upper part of the Pungo River Formation in the Lee Creek Mine, North Carolina. STRATIGRAPHIC RANGE.—Blow (1969:356) re- ported the range as zone N5 to near the end of zone N9. Globorotalia species cf. G. crassula Cushman and Stewart PLATE 3: FIGURES 4, 8, 12 REMARKS.—A single specimen has more cham- bers in the final whorl than most individuals of G. crassula, although Berggren and Amdurer (1973, pi. 30: fig. 9) illustrated G. crassula with 4y2 chambers in the final whorl. Specimens similar to the one from the Yorktown Formation were fig- ured by Lamb and Beard (1972, pi. 2, figs. 10- 12; pi. 20, figs 3-7) as G. crassacrotonensis Conato and Follador, a form placed in G crassula, sensu lato, by Berggren and Amdurer (1973:367). Prob- ably the most similar form was illustrated by Rbgl (1974, pi. 5, figs. 1-9, 13-15) as Globorotalia crassaformis cf. viola Blow, 1969. Rogl did not assign his specimens to G. crassula because they lack the moderately strong-keel development found in this species. OCCURRENCE.—A single specimen was found in the lower part of the Yorktown Formation (Pla- copecten clintonius zone of Mansfield) in the bluffs along the Meherrin River near Murfreesboro, North Carolina. STRATIGRAPHIC RANGE.—Blow (1969:361, 362) reported a range from zone N18 to zone N23 for G. crassula crassula and G. crassula viola. Globorotalia species cf. G. truncatulinoides truncatulinoides (d'Orbigny) PLATE 5: FIGURES 13-15 REMARKS.—Only a single specimen was found. Blow (1969:405) distinguishes G. truncatulinoides from G tosaensis by the presence in the former "of the peripheral carina, no matter where or to what extent it is developed." Blow (1969:395) recog- nizes G. truncatulinoides truncatulinoides "at the first appearance of an 'imperforate' carina regardless ofthe extent or position ofthe true carina." Rbgl (1974) illustrates transitional forms. By using these criteria, the specimen from the Lee Creek Mine would fall into the range of G. truncatulinoides truncatulinoides. The umbilicus is less open than 374 those of well-developed specimens of G. truncatu- linoides, and this specimen seems to be an early form of the species. Because only one atypical specimen is known, it is placed in an uncertain status. OCCURRENCE.—A single specimen was found in the upper part of the Croatan Formation in the Lee Creek Mine, North Carolina. G. truncatuli- noides was reported by Akers (1972:36) from the Waccamaw Formation at Walkers Bluff, North Carolina. STRATIGRAPHIC RANGE.—Blow (1969:370), Bronnimann and Resig (1971:1248), and Kennett (1973:587 reported a range from the base of zone N22 through zone N23. Genus Orbulina d'Orbigny, 1839 Orbulina universa d'Orbigny PLATE 5: FIGURE 8 Orbulina universa d'Orbigny, 1839a:3, pi. 1. fig. 1.—Blow, 1956:66, fig. 2, nos. 8-9.—Jenkins, 1971:193-194, pi. 23: fig. 660.—Stainforth et al., 1975:328-330, fig. 150. Orbulina cornwallisi McLean, 1956:365, pi. 53: figs. 3a-b. REMARKS.—Examination of McLean's type material and additional topotypic material showed no difference in surface appearance be- tween 0. cornwallisi and 0. universa. OCCURRENCE.—Common in the "Virginia St. Marys" beds in the Gatesville Well in North Carolina in the interval from 131.5 to 138 feet (40 to 42 m). Specimens are rare in the Yorktown Formation in Virginia and North Carolina, and in the Duplin and Waccamaw formations in North Carolina. Dorsey (1948:314) found a single specimen in bed 17 ofthe Choptank Formation in Maryland; examination indicates this speci- men should be referred to 0. suturalis. STRATIGRAPHIC RANGE.—Blow (1969:334) re- ported a range from within the lower part of zone N9, through zone N23. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Genus Praeorbulina Olsson, 1964 Praeorbulina glomerosa glomerosa (Blow) PLATE 1: FIGURES 3, 6 Globigerinoides glomerosa glomerosa Blow, 1956:65, fig. 1: nos. 15-19, fig. 2: nos. 1, 2. Praeorbulina glomerosa glomerosa (Blow).—^Jenkins, 1971:198, pi. 23: fig. 668. Praeorbulina glomerosa (Blow).—Postuma, 1971:376.—Stain- forth et al., 1975:281, fig. 121. REMARKS.—Stainforth et al. (1975) accepted Postuma's (1971:376) combining of the various subspecies of P. glomerosa into one taxon. The relatively few specimens of Praeorbulina found in this area indicate that the time of initial appear- ance of the subspecies is consistent with that found in warmer areas by Blow and that there is a transition upward through the various subspe- cies similar to that described by Blow. OCCURRENCE.—A single specimen was found in bed 10 ofthe Calvert Formation at Plum Point, Maryland. Dorsey (1948:314) reported rare oc- currences of Candorbulina universa Jedlitschka from beds 11 and 12 of the Calvert Formation in Maryland. Examination ofthe material indicates that two of the specimens from bed 11 belong in this subspecies, but one from bed 11 and two from bed 12 belong to P. glomerosa circularis. STRATIGRAPHIC RANGE.—Blow (1969:333) re- ported a range from the middle part of zone N8 to the basal part of zone N9. Praeorbulina glomerosa circularis (Blow) PLATE 10: FIGURE 9 Globigerinoides glomerosa circulans Blow, 1956:65, fig. 2, nos. 3, 4. Praeorbulina glomerosa circularis (Blow).—^Jenkins, 1971:196- 197, pi. 23: fig. 665. REMARKS.—As noted under P. glomerosa glome- rosa, the lineage proposed by Blow has been ques- tioned. The few specimens found in the Calvert Formation support Blow's interpretation. Bed 10 NUMBER 53 375 contains Globigerinoides sicanus and P. glomerosa glomerosa, bed 11 contains P. glomerosa glomerosa and P. glomerosa circularis, and bed 12 contains P. glomerosa circularis. Although there is overlap in the occurrences, the over-all trend is consistent with Blow's description. OCCURRENCE.—One specimen was found in bed 11 and two in bed 12 from the collections made by Dorsey from the Calvert Formation in Maryland. STRATIGRAPHIC RANGE.—Blow (1969:333) re- ported that its range is from the upper part of zone N8 to near the top of zone N9. Genus Pulleniatina Cushman, 1927 Pulleniatina obliquiloculata obliquiloculata (Parker and Jones) PLATE 2: FIGURES 5, 6, 9 Pullenia obliquiloculata Farker and Jones, 1865:365, pi. 19: figs. 4a,b. Pulleniatina obliquiloculata obliquiloculata (Parker and Jones).— Banner and Blow, 1967:137-139, pi. 3: figs. 4a-c. Pulleniatina obliquiloculata (Parker and Jones).—Stainforth et al., 1975:385, figs. 186, 187. REMARKS.—Although these specimens are con- sistent with the subspecies concept used by Blow in Pliocene and Pleistocene assemblages, they differ slightly from living specimens found in plankton tows and may warrant a separate tax- onomic designation to signify their biostrati- graphic importance. OCCURRENCE.—Rare in the uppermost part of the "Yorktown" Formation along the western shore of the Chowan River, at Colerain Landing, Mt. Gould Landing, and Black Rock Landing in Bertie County, North Carolina. STRATIGRAPHIC RANGE.—Blow (1969:376) re- ported a range from within zone N19 through zone N23, as did Parker (1967:172). Earlier first occurrences are from the beginning of zone N20 (Bronnimann and Resig, 1971:1435) and from zone N22 onward (Kennett, 1973:587). Genus Sphaeroidinellopsis Banner and Blow, 1959 Sphaeroidinellopsis seminulina seminulina (Schwager) PLATE 2: FIGURES 10-12; PLATE 5: FIGURE 7 Globigerina seminulina Schwager, 1866:256, pi. 7: fig. 112.— Banner and Blow, 1960:24, pi. 7: figs. 2a, b. Sphaeroidinella seminulina (Schwager).—Parker, 1967:161-162, pi. 23: figs. 1-5. OCCURRENCE.—Rare in the "Virginia St. Marys" beds in the Gatesville Well, North Caro- lina, at a depth of 131.5 feet (40 m) and in the lower part ofthe Yorktown Formation {Placopecten clintonius zone of Mansfield) along the bluffs of the Meherrin River near Murfreesboro, North Carolina. STRATIGRAPHIC RANGE.—Blow (1969:338) re- ported a range from zone N6 to near the zones N19/N20 boundary in most localities, with a range possibly as high as the lower part of zone N21 in the deep sea sequences. The youngest range, as indicated by Blow, is variable, depend- ing apparently upon the location and environ- ment. A range to uppermost zone N20 was given by Poag (1972b:492, 493, 499) and Berggren and Amdurer (1973:353, figs. 4, 7, 8) in the Atlantic. A range to the lower part of zone N21 was given by Parker (1967:161), to the latter part of zone N21 by Kennett (1973:587, 591, 594), and into the lower part of zone N22 by Bronnimann and Resig (1971:1435). Berggren and Van Couvering (1974:31) placed the extinction datum in the lowest part of zone N21. Sphaeroidinellopsis subdehiscens subdehiscens (Blow) PLATE 2: FIGURES 13-15 Sphaeroidinella dehiscens subdehiscens Blow, 1959:195-196, pi. 12: fig. 7la-c, 72. Sphaeroidinellopsis subdehiscens subdehiscens (Blow).—Blow, 1969:338, pi. 30: figs. 1-3, 6. OCCURRENCE.—A single specimen was found in 376 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY the lower part of the Yorktown Formation in the Placopecten clintonius zone of Mansfield along the bluffs of the Meherrin River near Murfreesboro, North Carohna. Akers (1972:32) reported this species from the Yorktown Formation at Rice's Pit, Virginia. STRATIGRAPHIC RANGE.—Blow (1969:338) re- ported a range from the base of zone N13 to within zone N19. Some workers (e.g., Parker, 1967:160) had a similar latest occurrence, but most recent work shows considerably later occur- rences for this species: the top of N19/N20 (Ujiie and Oki, 1974:44), the top of N20 (Bronnimann and Resig, 1971:1435), the lower part of N21 (Poag, 1972b:515; Berggren and Van Couvering, 1974:67), and as recent as the lowest part of N22 (Kennett, 1973:587, 588). Genus Turborotalia Cushman and Bermudez, 1949 In addition to the type-species Globorotalia cen- tralis Cushman and Bermudez and other related species that also occur in Eocene strata, a wide variety of species from Miocene and younger ages have been placed in this genus or subgenus. A problem with Blow's (1969) concept of this genus is that species from several different lineages, which possess characteristics oV^ Turborotalia" dur- ing some stage of the lineage development, are placed together, making the genus recognizably polyphyletic. This problem was discussed by Fleisher (1974). Although many workers have used Turborotalia for species occurring in Miocene and younger strata, it appears that this name should be restricted to the Eocene lineages, be- cause the phylogenetic gap between these two groups is large. This restriction leaves a large number of younger species without a generic name. Neogloboquadrina has been used for '''Turborotalia'" dutertrei and its related forms. In this study, species possessing a rounded periphery without an imperforate carina are temporarily placed in ''Turborotalia'''' pending resolution ofthe problem. *^ Turborotalia" birnageae (Blow), new combination PLATE 4: FIGURES 1-6 Globorotalia birnageae Blow, 1959:210-211, pi. 17: fig. 108a-c. Globorotalia (turborotalia) birnageae Blow.—Blow, 1969:346, pi. 34: figs. 7, 8. REMARKS.—The number of chambers in the last whorl varies from approximately 4y2 to 5V2. The aperture varies in position, becoming more extraumbilical in some specimens, and ranges from a very low slit (Plate 4: figures 2 and 3, and to a lesser degree Plate 4: figure 4) to a somewhat arched aperture (Plate 4: figure 5). The wall texture is coarsely cancellate. OCCURRENCE.—Rare to common in the upper beds of the Pungo River Formation in the Lee Creek Mine, North Carolina, and rare in the 581 to 618 foot (177 to 188 m) interval in the Norfolk, Virginia, Moores Bridge Well. STRATIGRAPHIC RANGE.—Blow (1969:346) re- ported a range from within zone N7 to within the later part of zone N9. Bronnimann and Resig (1971:1441) recorded the species as early as zone N6, with specimens of G. aff. birnageae specimens in zone N4. "Turborotalia" acostaensis humerosa (Takayanagi and Saito), new combination PLATE 2: FIGURES 1-3 Globorotalia humerosa Takayanagi and Saito, 1962:78, pi. 28: figs, 1-2.—Stainforth et al., 1975:357-360, fig. 170. Globorotalia (Turborotalia) acostaensis humerosa Takayanagi and Saito.—Blow, 1969:345-346, pi. 33: figs. 4, 5, 7-9, pi. 34: figs. 1-3. OCCURRENCE.—Rare to common in the lower part ofthe Yorktown Formation in the Placopecten clintonius zone along the bluffs of the Meherrin River near Murfreesboro, North Carolina. Akers (1972:32) reported this species from the Yorktown Formation in Rice's Pit in Virginia and the basal part of the section (Placopecten clintonius zone) in the Lee Creek Mine, North Carolina, and in the Waccamaw Formation at Walkers Bluff and Old Dock in North Carolina. NUMBER 53 377 STRATIGRAPHIC RANGE.—Blow (1969:345) re- ported a range from the latest part of zone N16 through N23. The ranges reported in most sub- sequent studies generally indicate a later origin and an earlier extinction. Parker (1967:169) gave a range from lower N18 to the top of N21; Bronnimann and Resig (1971:1433) recorded it from the middle of N17 into N23; Jenkins and Orr (1972:1099) had a range from N17 to N22; Poag (1972b:485) recorded it from N19 to the end of N21; and Kennett (1973:580, 583, 584, 588, 591) had a range from the base of N19 to N22. *^Turborotalia" inflata (d'Orbigny), new combination PLATE 3: FIGURES 13-15 Globigerina inflata d'Orbigny, 1839b: 134, pi. 2: figs. 7-9. Globorotalia (Turborotalia) inflata (d'Orbigny).—Banner and Blow, 1967:145-146, pi. 4: figs, la-c, 11. Globorotalia inflata (d'Orbigny).—Zachariasse, 1975:116, pi. 14: fig. 3a-c.—Stainforth et al., 1975:360, fig. 171. OCCURRENCE.—A single specimen that un- doubtedly belongs to this species was obtained from the Waccamaw Formation at Walkers Bluff on the Cape Fear River, North Carolina. Several specimens placed in G. puncticulata, although they show strong transitional characteristics to G. in- flata, were found in the upper part of the York- town and Duplin formations in North Carolina. Akers (1972:36, 42) reported this species from Walkers Bluff and Old Dock in the Waccamaw Formation in North Carolina. STRATIGRAPHIC RANGE.—Although Blow (1969:350) reported a range from zones N17 through N23, most subsequent work has shown a considerably later initial appearance of the spe- cies. Boltovskoy (1974:678, 706) reported G. inflata from sediments as old as late Miocene in the Indian Ocean, but, judging from his ensuing discussion, it appears that these were confused with G. crassaformis. Bolli (1970:581, 614) and Kennett (1973:587) indicated a range from the equivalent of upper N20 through N23, and Par- ker (1967:179), Ujiie and Oki (1974:39, 44), and Zachariasse (1975:30-31, 39, 79) had the first occurrence at the base of N21. It seems that an earliest occurrence of uppermost N20 to lower N21 is best documented. Genus Spiroplectammina Cushman, 1927 Spiroplectammina mississippiensis (Cushman) PLATE 9: FIGURES 5, 9; PLATE 16: FIGURES 1-3 Textularia mississippiensis Cushman, 1922b:90, pi. 14: fig. 4. Spiroplectammina mississippiensis (Cushman).—Dorsey, 1948: 275-276, pi. 27: figs. 3a-4b. Spiroplectammina spinosa Dorsey, 1948:276, pi. 27: figs. 5a, 6b. DESCRIPTION.—Test elongate, varying in width from iy2 to 3 times as long as wide, tapering uniformly towards the initial end; periphery acute, usually keeled, and keel development, if present, varying from slight to very wide in dif- ferent populations; chambers planispirally coiled in early portion, consisting of 5 or 6 chambers, later part of test biserial; chambers compressed, with little, if any, increase in inflation of cham- bers throughout ontogeny; sutures distinct, slightly depressed, varying from straight to slightly curved downward, sutural areas filled with clear shell material that is variable in devel- opment from a slight line along the suture to broad bands covering as much as half of the surface of the test; wall finely arenaceous with much cement, giving a smooth finish; aperture a moderately low slit at the base ofthe inner margin of the last-formed chamber. REMARKS.—This species is variable, as noted in the description. S. spinosa Dorsey was distin- guished by its straight, horizontal sutures and spinose projections at the peripheral margin. Spi- nose projections can be seen on many specimens of S. mississippiensis as either the thicker parts of a broken peripheral keel or as the incipient parts of a poorly developed keel. Although the holotype of S. spinosa has horizontal sutures, as do speci- mens in some populations of 5. mississippiensis, the paratypes have slightly curved sutures. Dorsey reported both S. spinosa and S. mississippiensis in 378 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY the same samples, and it seems clear that S. spinosa is just an extreme variation within a population of S. mississippiensis. Spiroplectammina mississippiensis can be distin- guished from S. exilis Dorsey, which occurs in the Choptank Formation, because the latter has a more inflated test with a nonkeeled periphery, the sutures are more strongly curved downwards, and it has a longer and narrower shape. OCCURRENCE.—Rare to common throughout most of the Calvert Formation in Maryland, ranging from beds 3 to 14, and rare in the Pungo River Formation in the Lee Creek Mine and in the "Virginia St. Marys" beds in the Gatesville Well in North Carolina at a depth of 138 feet (42 m). STRATIGRAPHIC RANGE.—The first appearance of this species is in strata of Eocene age, ranging geographically from Texas to Virginia. It is com- mon in the Oligocene and questionable in the lowermost Miocene strata of the Gulf Coast re- gion. The occurrences in the Miocene Calvert and Pungo River formations and "Virginia St. Marys" beds are probably the most recent rec- ords. Spiroplectammina exilis Dorsey PLATE 10: FIGURES 1, 2; PLATE 16: FIGURES 4, 5 Spiroplectammina gracilis Cushman and Cahill, 1933:6, pi. 1: figs. 6, 7. Spiroplectammina exilis Dorsey, 1948:275, pi. 27: figs. 1, 2. DESCRIPTION.—Test elongate, about 2y2 times as long as wide, tapering uniformly towards the initial end; moderately inflated test with subacute to narrowly rounded periphery; chambers planispirally coiled in initial stage, numbering 4 or 5, composing small part of test, later part of test biserial consisting of 16 to 20 chambers; chambers compressed with only slight increase in inflation during ontogeny; sutures distinct, slightly depressed, varying from straight to slightly curved downward, meeting periphery at approximately a 45 degree angle; wall finely arenaceous and containing much cement, giving a smooth finish; aperture a low, narrow slit at base of inner margin of last-formed chamber. REMARKS.—This species is distinguished from S. mississippiensis (Cushman) by its more inflated cross-section, nonkeeled periphery, more curved sutures, and generally more elongated shape. OCCURRENCE.—This species is characteristic of the Choptank Formation in Maryland and north- ern Virginia. Dorsey (1948) reported an occur- rence in bed 24 ofthe St. Marys Formation along the St. Marys River, but extensive collecting by the author has not yielded any specimens belong- ing to this genus in those strata. STRATIGRAPHIC RANGE.—The age range is re- stricted to the middle Miocene. Genus Textularia Defrance, 1824 Textularia obliqua Dorsey PLATE 16: FIGURES 8, 9 Textularia obliqua Dorsey, 1948:279, pi. 28: figs. 6, 7. DESCRIPTION.—Test broadly elongate, about iy2 times as long as broad, widest part may be at last pair of chambers or at earlier stage, early portion strongly tapering, later portion almost parallel; test moderately inflated, periphery broadly rounded; 7 to 8 pairs of chambers bise- rially arranged, early chambers much longer than high, later ones increasing in height; sutures dis- tinct, straight, slightly depressed, forming an an- gle of approximately 30 degrees with horizontal; wall coarsely arenaceous, smoothly finished; ap- erture an arched slit at base of inner margin of last formed chamber. OCCURRENCE.—This species is rare throughout beds 22 to 24 of the St. Marys Formation in Maryland and also is found at one locality in the overlying "Virginia St. Marys." STRATIGRAPHIC RANGE.—The range is from up- per middle to probable lower upper Miocene. Textularia mayori Cushman PLATE 16: FIGURES 11, 12 Textularia mayori Cnshman, 1922a: 23, pi. 2: fig. 3.—McLean, 1956:320, pi. 36: figs. 1-3. DESCRIPTION.—Test broadly elongate, almost NUMBER 53 379 as broad as long, widest part at last pair of chambers, uniformly tapering toward initial end; test moderately compressed throughout, periph- ery strongly angled with variable number of spines projecting from chamber margins; 7 to 8 pairs of chambers biserially arranged, early cham- bers longer than high, later ones gradually be- coming higher; sutures indistinct, later ones slightly depressed, straight to slightly curved downward; wall finely arenaceous with relatively smooth finish; aperture a low slit at base of indented inner margin of last formed chamber. REMARKS.—Spinosity ranges from complete absence of spines to long projecting spines. This species is similar to T. gramen d'Orbigny, and it is probable that the specimens from the Choptank and St. Marys formations placed in T mayori by Dorsey (1948:278) belong to T. gramen. Textularia gramen does not have spines and differs from nonspinose forms of T mayori by having a shorter, broader test, which is more inflated in cross-sec- tion. OCCURRENCE.—Rare to very common through- out the Yorktown Formation in Virginia and North Carolina and in the Waccamaw Formation in North Carolina. STRATIGRAPHIC RANGE.—From zone N19 to present off the southeastern coast of the United States. Textularia ultima-inflata Dorsey PLATE 16: FIGURES 6, 7 Textularia ultima-inflata Dorsey, 1948:279, pi. 28: fig. 8a-c. DESCRIPTION.—Test broadly elongate, about 1V2 times as long as broad, widest part at last pair of chambers, strongly tapering toward initial part throughout test; test compressed in early portion, inflated in later; periphery subacute in early por- tion, broadly rounded in later portion; 7 to 8 pairs of chambers biserially arranged, early cham- bers much longer than high, later ones gradually becoming higher until last pair is about as high as wide; sutures indistinct, not depressed, curved slightly downward; wall finely to coarsely arena- ceous; aperture a low slit at indented base of inner margin of last formed chamber. OCCURRENCE.—Only in beds 17 to 19 of the Choptank Formation in Maryland. STRATIGRAPHIC RANGE.—The range is restricted to the middle Miocene. Genus Bolivinopsis Yakovlev, 1891 Bolivinopsis fairhavenensis, new species PLATE 20: FIGURES 1-4 Bolivinopsis curta Cushman, 1948:220 Inot Spiroplecloides curta Cushman, 1933] DESCRIPTION.—Test small, about 3 times as long as broad, early planispiral stage moderately compressed, becoming less compressed through growth with later biserial portion only slightly compressed, being oval in cross-section and hav- ing a rounded periphery; sides of test nearly parallel with biserial portion having approxi- mately the same width as planispiral; biserial stage usually straight, but may be slightly to moderately curved; chambers distinct in both stages; 7 to 8 chambers in planispiral stage, usu- ally 3 to 4 slightly inflated chambers of equal height and width in each row of biserial stage; sutures distinct, slightly depressed, forming ap- proximately a 45 degree angle with periphery; wall calcareous, radial, finely perforate, smooth, highly translucent to transparent; an elongate oval aperture in the terminal face extending to the inner margin. REMARKS.—Bolivinopsis is most common in Cre- taceous to Eocene rocks; B. fairhavenensis is one of the youngest species of the genus. The nature of the wall structure within the genus, as typified by the Russian type-species, remains uncertain. Some species placed in Bolivinopsis clearly look agglutinated, while others, including the present species, clearly appear to be perforate calcareous. Some species in the USNM collections that are placed in Spiroplectammina Cushman, 1927, also appear to be calcareous perforate. This problem of the generic wall structure is discussed in Loeb- lich and Tappan (1964:251). Examination ofthe species in the USNM collections shows three 380 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY groups of species that apparently have the same pattern of chamber arrangement. One group has a clearly agglutinated structure, ranging from coarse- to fine-grained; one has clearly calcareous perforate structure and one may be finely agglu- tinated or calcareous perforate. This last group will have to be sectioned to determine the wall Structure. If the type-species of Bolivinopsis, B. capitata Yakovlev, does prove to have a calcareous perforate wall, as has been suggested, then the calcareous perforate group does have a name. If this species proves to be agglutinated, then it probably should be the senior synonym of Spiro- plectammina, and the perforate group will need a new generic name. Bolivinopsis curta (Cushman) differs from B. fair- havenensis in that it has a broader and more inflated test (about twice as thick) containing more inflated chambers throughout and a more roughened or granular wall texture. Bolivinopsis attenuata (Cushman) from the Eocene of the Atlantic shelf differs by having a broad planispiral stage followed by a narrow biserial stage with compressed chambers, a sub- acute periphery, and limbate sutures. OCCURRENCE.—Restricted in outcrop occur- rence in Maryland to the top of bed 3 of the Fairhaven Member of the Calvert Formation in Calvert County. The abundance may be as high as 4 percent. Cushman (1948) found it in the lower 100 feet (30 m) ofthe Calvert Formation in the Hammond Well on the Eastern Shore of Maryland. It also was recorded in the upper 10 feet (3 m) of the Eocene in this well, but this probably represents down-hole contamination. In a continuous core hole at the site ofthe Baltimore Gas and Electric Company's nuclear power plant near St. Leonards, Calvert County, Maryland, this species occurs from within 8 feet (2.4 m) of the base of the Calvert Formation to the top of bed 3, an interval of 108 feet (33 m). In Maryland, this species is characteristic of the lower part of the Calvert Formation, the Fairhaven Member. The species was found in one sample in the Pungo River Formation in the Lee Creek Mine, North Carolina. STRATIGRAPHIC RANGE.—The occurrences in Maryland are in the lower part of the Calvert Formation, below the beds assigned to the latest part of planktonic zone N8 or N9. The North Carolina occurrence is in beds dated as probably late N8. Thus, the upper age limit of the species in this area is latest N8 (late early Miocene). Although the lower part ofthe Calvert Formation appears to be somewhere in zone N8, it has not been dated with certainty; therefore the lower range of this species is still uncertain. TYPE LOCALITY.—The type locality for the hol- otype, figured paratypes, and unfigured para- types is Randle Cliffs on the western shore of Chesapeake Bay, Calvert County, Maryland, in the upper part of bed 3 ofthe Calvert Formation. TYPES.—The holotype is USNM 252518; fig- ured paratypes are USNM 252519 and 252520; all from USGS 25981. Genus Quinqueloculina d'Orbigny, 1826 Quinqueloculina lamarckiana d'Orbigny PLATE 17: FIGURES 1, 2, 6 (^inqueloculina lamarckiana d'Orbigny, 1839a: 189, pi. 11: figs. 14, 15.—Cushman, 1929:26, pi. 2: fig. 6. (Quinqueloculina venusia? Cushman, 1918:70, pi. 29: fig. 3a-c. (Quinqueloculina cuvieriana Cushman, 1919:69. (Quinqueloculina seminulangulata McLean, 1956:322, pi. 37: fig. 8a, b. DESCRIPTION.—Test about IV2 times as long as wide, triangular in cross-section, periphery acute to subacute; chambers broad and flattened, mid- dle chamber large, projecting, having distinct acute edge, other chambers visible as narrow band; sutures distinct, slightly depressed; surface smooth, polished, covered with few low costae that are parallel to slightly oblique to periphery; aperture oval, having a slightly thickened rim and a short, stout tooth. REMARKS.—The degree of acuteness of the pe- riphery varies from strongly angulated to suban- gulated. The strength ofthe ornamentation varies from barely visible on the early chambers to well NUMBER 53 381 developed on all chambers. These characters vary with latitude in living populations along the At- lantic Coast ofthe United States. Specimens with a more rounded periphery and less-developed costae occur in the northern part of the range in coastal waters off North Carolina, whereas forms having a more angular periphery and more strongly developed costae are more abundant in populations off Florida. A similar north-to-south pattern in the variation is seen in the Yorktown and Waccamaw formations; the more southern localities have more strongly angulated and or- namented specimens. OCCURRENCE.—Rare to common throughout the Yorktown Formation in Virginia and North Carolina and in the Waccamaw Formation in North and South Carolina. STRATIGRAPHIC RANGE.—The range in this area is from lower Pliocene strata (zone N19) into the living fauna. In South America, the West Indies, and the southern United States, reports extend its range to the Eocene. Genus Massilina Schlumberger, 1893 Massilina glutinosa Cushman and Cahill PLATE 17: FIGURES 3, 4, 7 Massilirm glutinosa Cushman and Cahill, 1933:10, pi. 2: fig. lOa-c—Dorsey, 1948:281, pi. 29: fig. 6a-c. DESCRIPTION.—Test strongly compressed with parallel sides, periphery broadly rounded, test oval in outline, apertural end not projecting; chambers of uniform width, later chambers nearly circular in cross-section; sutures distinct, slightly depressed; wall agglutinated with much cement; aperture circular and has slightly thick- ened lip, containing bifid tooth that may be thin and elongated or short and stout. OCCURRENCE.—Known from beds 16 through 20 of the Choptank Formation in Maryland and northern Virginia. STRATIGRAPHIC RANGE.—Restricted to the mid- dle Miocene. Massilina marylandica Cushman and Cahill PLATE 16: FIGURES 10, 13, 14 Massilina marylandica Cushman and Cahill, 1933:10, pi. 2: fig. 9a-c. DESCRIPTION.—Test strongly compressed, pe- riphery rounded, test oval in outline, apertural end not projecting; chambers distinct, later ones increasing considerably in width; sutures distinct, slightly depressed; wall has well-developed lon- gitudinal costae crossing chamber at slight angle to the periphery; aperture circular, containing thickened bifid tooth. OCCURRENCE.—Found only in the upper part of the St. Marys Formation at Langley's Bluff and Chancellor Point, Maryland. STRATIGRAPHIC RANGE.—Middle to lower up- per Miocene. Genus Nodosaria Lamarck, 1812 Nodosaria catesbyi d'Orbigny PLATE 11: FIGURES 5, 6; PLATE 17: FIGURE 11 Nodosaria catesbyi d'Orbigny, 1839a: 16, pi. 11: figs. 8-10.— McLean, 1956:329, pi. 39: figs. 1-4. DESCRIPTION.—Test elongate, varying from slightly tapering to slightly expanding; initial end blunt and rounded, some specimens have an ap- ical spine; chambers vary in number from 2 to 4; sutures distinct, depressed; wall ornamented by 12 to 16, high, sharp, longitudinal costae, contin- uous across the sutures; aperture terminal, radiate on short to medium neck; wall calcareous, finely perforate. REMARKS.—Originally characterized by d'Or- bigny as composed of two chambers; specimens having three or four chambers occur in the pres- ent material. The later chambers may be as large or larger than the earlier, or considerably smaller in size. The costae vary in development from strong to very faint or absent on the last chamber. The costae may continue onto the neck where they are twisted. 382 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY OCCURRENCE.—Lower and middle parts of the Yorktown Formation in Virginia and northern North Carolina. STRATIGRAPHIC RANGE.—The occurrences in this area are restricted to the lower and middle Pliocene parts of the Yorktown Formation, al- though the species is found in upper Oligocene and Pliocene strata in other areas. Genus Bolivina d'Orbigny, 1839 REMARKS.—Loeblich and Tappan (1964:549) confine the usage of Bolivina to those forms having the test "somewhat compressed . . . basal margins of chambers with retral processes or backward directed chamber overlap .... Species without chamber overlaps, commonly keeled and strongly compressed, are placed by us in Brizalina.'" These distinctions have not been uniformly applied by subsequent authors. Bolivina pungoensis, new species, undoubtedly belong to the genus Bolivina under these concepts, but the placement of B. calvertensis would be more difficult as it is transitional between the groups. The gradational nature of the large number of species belonging to the Bolivina-Brizalina complex as interpreted by Loeblich and Tappan makes placement of the species difficult. Bolivina calvertensis Dorsey PLATE 9: FIGURES 1-4 Bolivina calvertensis Dorsey, 1948:306, pi. 36: fig. 17a-c. DESCRIPTION.—Test elongate, from 2 to 3 times as long as broad, gently tapering toward both ends and having greatest width at or slightly anterior to mid-point of test; somewhat com- pressed test that has subacute periphery in earlier stages, becoming more inflated with more rounded periphery in later stages; chambers in later part moderately distinct, slightly inflated; sutures somewhat distinct with inner part being curved and later straight, forming angle of about 30 to 45 degrees with the horizontal; in earlier part, chambers and sutures largely obscured by ornamentation; wall moderately perforate; sur- face of all but the last 2 or 3 chambers promi- nently ornamented by closely spaced longitudinal costae, varying in strength from fine to moder- ately coarse between specimens, numbering ap- proximately 12 to 20, costae may bifurcate and shortly rejoin; aperture moderately narrow, an elongate oval, opening above the base of last formed chamber, sometimes with slight lip. REMARKS.—Dorsey compared this species to B. marginata multicostata Cushman; the latter differs in being much larger, having a more compressed test with a keel, and having fewer costae that are straighter and cover less of the test. OCCURRENCE.—Dorsey reported the species as rare in beds 6 through 14 in the Calvert Forma- tion in Maryland. Additional collecting by the author also yielded few specimens in bed 5, giving a range from the middle to upper parts of the Calvert Formation. Occurrence is rare to com- mon in the upper part of the Pungo River For- mation in the Lee Creek Mine. STRATIGRAPHIC RANGE.—Found only in those parts of the Calvert and Pungo River formations regarded as equivalent to zones N8-lower N9. Bolivina pungoensis, new species PLATE 8 DESCRIPTION.—Test elongate, about 2 to 2y2 times as long as broad, gently tapering toward both ends, greatest width near anterior end, usu- ally across base of shoulders of last two chambers; moderately compressed test with subacute peri- phery in early part, becoming more inflated with more rounded periphery in later two-thirds to one-half of test; chambers distinct, moderately compressed in early stages, becoming more in- flated in later part, chambers in early portion low and broad, later increasing considerably in height, inner basal part of chambers extending as a lobe over earlier chambers and the lobular projections becoming very pronounced in later chambers; sutures distinct, depressed, curved near periphery at 50 to 60 degree angle to the horizon- tal; wall distinctly perforate except for areas sur- rounding the apertural face; imperforate areas NUMBER 53 383 around aperture generally only partially covered by succeeding chambers in earlier part ofthe test, more commonly completely overlapped in later parts; aperture narrow, elongate, opening on or near the base of the inner margin of the last- formed chamber, usually with tooth plate ex- posed. REMARKS.—The apertural face and surround- ing area is imperforate, and during ontogeny the amount of imperforate area covered by succeed- ing chambers varies. In the earlier parts of the test much of the imperforate area remains visible after succeeding chambers are formed (Plate 8: figure 5), whereas during later stages of ontogeny the amount of imperforate area exposed becomes considerably reduced (Plate 8: figure 4), and in the latest stages, only very thin areas, if any, are visible (Plate 8: figure 3). The grain size within the imperforate area is coarsest in the early stages (Plate 8: figure 9) and becomes increasingly finer in the later stages (Plate 8: figures 7, 8). This appears to be a result of the organism's activity and not of diagenetic alteration, as the same pattern of changing grain size is observable on all specimens. Diagenetic alteration should affect the areas equally as far as regrowth of crystal size is concerned. A group of species of Bolivina appears in the early and lower middle Miocene that seems to be morphologically related, and occurs over much of North America. Although some of the species are apparently restricted stratigraphically, others range well into and through the later parts of the Miocene strata. This group includes B. pungoensis, new species, and the following three closely re- lated species, among others. Bolivina advena Cushman (1925:29), described from the Monterey shale of California, differs from B. pungoensis by having the lobular exten- sions not as strongly developed, and by lacking the imperforate areas in the earlier chambers. Bolivina floridana Cushman (1918:49), described from the Choctawhatchee Formation in Florida, is longer and narrower (specimens being 3 to 4 times as long as wide), has the inflation of the chambers beginning much earlier, has a more rounded periphery in the later chambers, has multiple lobular extensions, and lacks the imper- forate areas on earlier chambers. B. floridana im- porcata Cushman and Renz and B. floridana regu- laris Cushman and Renz, from the Miocene of Venezuela, are similar to B. pungoensis in the overall shape ofthe test, but differ in the presence of multiple lobular extensions. Bolivina plicatella mera Cushman and Ponton (1932:82), from the Oak Grove Formation in Florida, differs in having the greatest width at the apertural end, giving a square appearance, having a more compressed test, and lacking the imperforate areas. OCCURRENCE.—Bolivina pungoensis comprises about 3 percent of the foraminiferal assemblages in most samples ofthe Pungo River Formation in the Lee Creek Mine. It is found also in the Pungo River Formation in the 581 to 616 foot (177 to 187.7 m) interval ofthe Norfolk, Virginia, Moores Bridge Well. STRATIGRAPHIC RANGE.—Found only in the parts of the Pungo River Formation that are assigned to planktonic zone N8 of Blow (1969:289) of late early Miocene age. TYPE-LOCALITY.—The locality for the holotype and figured paratypes is the Lee Creek Mine, North Carolina, in the upper part of the Pungo River Formation. Unfigured paratypes are from the Moores Bridge Well, Norfolk, Virginia, at depths of 581 and 610 feet (177 and 186 m). TYPES.—The holotype is USNM 240134 from USGS locality 26013; figured paratypes are USNM 240135 from USGS locality 26014 and USNM 240136 from USGS locality 26013. Un- figured paratypes are USNM 240151 from USGS locality 26002 and USNM 240152 from USGS locality 26003. Bolivina marginata multicostata Cushman PLATE 11: FIGURES 1, 2; PLATE 17: FIGURES 5, 8 Bolivina aenariensis var. multicostata Cushman, 1918:48, pi. 10: fig. 2. Bolivina marginata var. multicostata Cushman.—Cushman, 1937:87, pi. 10: figs. 71-10. 384 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Bolivina marginata multicostata Cushman.—Puri, 1953:121, pi. 22: figs. 3-6. DESCRIPTION.—Test elongate, about 2V2 times as long as broad, uniformly tapering toward ini- tial end, very compressed, variable development of keel on periphery ranging from prominent to almost absent, commonly absent on last several chambers; approximately 18 to 20 compressed chambers, early ones low and broad, later ones increasing in relative height; sutures distinct, slightly limbate and curved, intersecting margin at 45 to 60 degree angle; ornamented by a series of longitudinal costae, approximately 6 in num- ber, which vary in strength and extent; may have short basal spine; aperture narrow, elongate, opening at inner margin of base of last formed chamber; wall moderately perforate. REMARKS.—This species is variable in devel- opment of the keel and costae and in the shape of the sutures. The keel may extend along the entire periphery, most of the periphery except for the last two pairs of chambers, or a lesser distance. The costae vary in their length and strength, and in some specimens the central costae may be enlarged to form a median ridge. The shape of the sutures varies from strongly arched to straight. Although both this subspecies and B. marginata marginata Cushman occur in the same strata in Florida (Puri, 1953), B. marginata multicostata is by far the more common of the two in Virginia and North Carolina. OCCURRENCE.—Rare in the middle part of the Yorktown Formation in Virginia and North Car- olina and at Barwick Farm in the Duplin For- mation in North Carolina. Cushman (1937) re- ported this subspecies from the Calvert Forma- tion in Maryland, but none were found in out- crops ofthe Calvert Formation during this study. Specimens are found, however, in the subsurface strata of the Pungo River Formation in the Moores Bridge Well at 610 feet (186 m), Norfolk, Virginia. STRATIGRAPHIC RANGE.—The range in this study area is from uppermost lower Miocene (equivalent to zone N8) to probable middle Pli- ocene. Genus Hopkinsina Howe and Wallace, 1932 Hopkinsina bononiensis (Fornasini) PLATE 11: FIGURES 3, 4 Uvigerina bononiensis Fornasini, 1888:48, pi. 3: figs. 12-12a. Hopkinsina bononiensis (Fornasini).—Marks, 1952:288, pi. 1: fig. 25. DESCRIPTION.—Test elongate, fusiform, initial end rounded; slightly inflated chambers arranged triserially in early stages, later becoming biserial; chambers have lobulate projections over earlier chambers; sutures distinct, depressed; approxi- mately 20 costae of medium strength on each chamber, some bifurcating, others discontinuous, costae do not extend over the sutures; aperture terminal, rounded, on short, relatively broad neck containing slight lip and internal tooth plate; wall calcareous, finely perforate. REMARKS.—The occurrence in the Pungo River Formation in North Carolina is the first report of this species in North America. The species is widespread in middle Miocene through Pliocene strata of Europe and North Africa, particularly in the Mediterranean region. Several varieties and subspecies of//, bononiensis have been described, but comparative material is lacking in the USNM collections and some ofthe original illustrations are inadequate for subspe- cific discrimination of the North Carolina mate- rial. The North Carolina specimens appear to be most similar to the original material of Fornasini. The only species similar to H. bononiensis is //. quasistriata Krasheninnikov, 1961, from the mid- dle Miocene of Russia, which differs in having barely discernible, very fine longitudinal striae. OCCURRENCE.—Only in the younger, subsur- face part ofthe "Virginia St. Marys" beds in the Gatesville Well, North Carolina, at a depth of 131.5 to 132 feet (40.1 to 40.2 m). STRATIGRAPHIC RANGE.—The known range in this area is restricted to the upper Miocene. NUMBER 53 385 Genus Sagrina d'Orbigny, 1839 Sagrina pulchella primitiva (Cushman), new combination PLATE 17: FIGURES 9, 10 Bolivina pulchella var. primitiva Cushman, 1930:47, pi. 8: figs. 12a-b. Bolivina pulchella primitiva Cushman.—Puri, 1953:122, pi. 21: figs. 11-12. DESCRIPTION.—Test about iy2 times as long as broad, greatest breadth formed by last pair of chambers, periphery broadly rounded, lobulate; chambers distinct, inflated, increasing gradually in size, early and middle stages of test triserial, only last 3 chambers biserial; sutures distinct, depressed, straight, making angle of 20 to 30 degrees with horizontal; test ornamented by short, strong, longitudinal costae, usually not crossing sutures; aperture an elongate oval with distinct lip; wall coarsely perforate. OCCURRENCE.—Rare in the Waccamaw For- mation in southern North Carolina. Its distribu- tion elsewhere includes the Pliocene, Pleistocene, and Holocene of Florida and adjacent areas. STRATIGRAPHIC RANGE.—In the study area the species has a range of upper Pliocene to lower Pleistocene, although it ranges from lower Pli- ocene into the Holocene in Florida and other southern areas. Genus Siphogenerina Schlumberger, 1882 Siphogenerina lamellata Cushman PLATE 9: FIGURES 6, 10-16; PLATE 17: FIGURE 13 Siphogenerina lamellata Cushman, 1918:55-56, pi. 12: fig. 3.— Dorsey, 1948:309, pi. 36: figs. 13a, b. Siphogenerina spinosa (Bagg).—Cushman, 1926:10.—Dorsey, 1948:309, pi. 36: figs. 14a, b. DESCRIPTION.—Test elongate, about 3 to 4 times as long as wide, greatest width at or near apertural end, tapering gradually toward initial end and tapering to broadly rounded at apertural end; early portion triserial, later and greater por- tion of test uniserial, commonly with 6 to 8 uniserial chambers; chambers distinct, increasing very gradually in size as added, slightly inflated; ornamentation consisting of longitudinal costae, ranging in number from approximately 7 to 9, equally spaced, varying in strength from slightly raised to broadly flanged, and extending the length of the uniserial stage and commonly onto the initial chambers where they may project as basal spines, additional costae seldom added by intercalation; sutures distinct, flush to strongly depressed with a U-shaped pattern, being strongly curved downwards at the costae and arched upwards between costae; aperture termi- nal with a short cylindrical neck and lip. REMARKS.—This species appears to have rela- tively consistent variations over its broad geo- graphic range. The test usually expands toward the apertural end where the greatest width occurs. Less frequently the greatest width may be about two-thirds of the distance from the initial end, with a gradual taper towards the apertural end similar to the shape of S. collomi Cushman. A few specimens in the Caribbean region may have additional costae added in the later stages of the test by intercalation (the holotype from Florida has a very slight added costa over the last several chambers), but this characteristic is rare in this species in contrast to other species of Siphogenerina. Basal spines are commonly found in a small to moderate proportion ofthe population in various geographical areas. The spines are a result of the projection of the basal portion of the costae be- yond the initial chambers ofthe test. The only other reported species of Siphogenerina from the Miocene of the middle Atlantic Coastal Plain is S. spinosa (Bagg), originally described from the Choptank Formation at Jones Wharf, Maryland. The type material of this species is not in the Cushman Collection at the Smithsonian and its whereabouts is unknown. The type illus- tration of S. spinosa does not show the basal spines that are discussed in the description and indicated by the species name [although Bagg (1904:480) mentioned distal spines, he probably meant spines at the basal or proximal end]. Basal spines 386 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY are prominent in specimens such as Dorsey's from the Calvert Formation and in some specimens from the type area in the Red Bay Formation of Puri and Vernon (1964) in western Florida, and are probably just part of the range of variation within S. lamellata. The prominent intercalated costae shown in the type illustration of S. spinosa distinguish it from most specimens of S. lamellata, but it is probable that both taxa are variants of one species. If so, S. spinosa has priority. No other specimens of Siphogenerina have been reported from the Choptank Formation, including samples from Jones Wharf examined by Dorsey (1948) and Gibson (1962). Because ofthe absence of any type material of S. spinosa, the two species are not placed in synonymy. Several related species are found in Miocene strata in Florida, the Caribbean region, and Cal- ifornia. Among the most closely related are the following species. Siphogenerina transversa Cushman is very similar to S. lamellata, but differs in having a less tapered test; more inflated later chambers, which may give a slightly nodular appearance; more costae (11 to 12 being common); and intercalation and bifurcation of the costae. Siphogenerina reedi Cushman is broader; not as tapered toward the initial end, and has more costae (13), which are more closely spaced and not as strongly developed. Some specimens of S. senni Cushman and Renz closely approach S. lamellata, but the former typ- ically differs in having 9 costae that become less strongly developed during ontogeny; commonly having intercalation of costae, a spinose projec- tion at the base ofthe test (typically better devel- oped than in S. lamellata), sutures not as strongly recurved, and very little taper toward the initial end. Siphogenerina collomi Cushman differs in com- monly having intercalated costae half-way up the side ofthe test, having more costae (12-14), and in tapering anteriorly. Broken specimens more closely resemble S. lamellata, but have more costae that are not as strongly developed. OCCURRENCE.—Rare in the Calvert Formation in Maryland in beds 9 to 13; rare in the upper beds of the Pungo River Formation in the Lee Creek Mine, and rare to common in the Norfolk, Virginia, Moores Bridge Well at a depth of 610 feet (186 m) and in the "Virginia St. Marys" beds in the Gatesville Well, North Carolina, at 131.5 feet (40 m). McLean (1956:350) reported one specimen which he attributed to reworking in the Yorktown Formation in Virginia. STRATIGRAPHIC RANGE.—Specimens are found in the Calvert and Pungo River formations in levels assigned to zones N8 to Nl 1. In the middle Atlantic Coastal Plain this species appears to be useful as an index to strata of this age. Younger ages for this species are reported in other areas. In Florida, the species occurs only in the Area zone (Cushman and Ponton, 1932:86), which provisionally has been placed by Akers (1972:5, 14) in zone N17. Blow (1959:153) found the species to range throughout the Miocene section in Venezuela, which in that area included a range from zone N5 into zone N17; Blow (1959:153) cited the local environment as a possible causal factor for the upper range. Other closely related species, particularly those found in California, occur through approximately the same time range as S. lamellata in North Carolina to Maryland. Siphogenerina collomi and S. reedi have ranges of zones N8-N9 in California (Kleinpell, 1938:300, 304; Berggren and Van Couvering, 1974, fig. 1), with S. transversa being found earlier in approximately zones Nl to N7, and thus being a likely forerunner of this group. In Venezuela, Blow (1959:153) reported a time range of zones N6 to N12 for S. senm and zones N6 to NIO for S. transversa. Siphogenerina species PLATE 9: FIGURES 7, 8 REMARKS.—Some specimens from the upper beds of the Pungo River Formation in the Lee Creek Mine differ from S. lamellata in having almost horizontal sutures and in having twelve moderately developed costae. These specimens probably represent a new species because none of NUMBER 53 387 the other described species of Siphogenerina in- cludes such forms. OCCURRENCE.—Rare in the upper part of the Pungo River Formation in the Lee Creek Mine, North Carolina. STRATIGRAPHIC RANGE.—The only occurrence of this species is in beds assigned to zone N8. Genus Rotorbinella Bandy, 1944 Rotorbinella bassleri (Cushman and Cahill), new combination PLATE 18: FIGURES 4-6 Rotalia bassleri Cushman and Cahill, 1933:30, pi. 10, fig. 7a-c.—Dorsey, 1948:312, pi. 37: fig. 8a-c. DESCRIPTION.—Test trochoid, planoconvex, spi- ral side moderately convex with thickening of shell material at apex, umbilical side varies from slightly convex to slightly concave, outline circu- lar; last several chambers may be lobulate; pe- riphery subacute and limbate; umbilicus de- pressed with stout plug of clear shell material; 3y2 whorls, 20 to 25 chambers, 6 to 7 chambers in last whorl, increasing gradually in size; sutures distinct, strongly recurved and slightly limbate on spiral side, moderately recurved and depressed on umbilical side; aperture an elongate slit with slight lip, extending from near periphery to um- bilical plug, apertural slits on last 2 to 4 chambers may be visible; wall coarsely perforate. REMARKS.—The convexity of the test is vari- able. The spiral side is usually moderately convex, but the umbilical side can vary from concave to convex. In some specimens the last chamber is strongly inflated. This species is similar to R. colliculus Bandy and R. campanulata (Galloway and Wissler) and clearly belongs to Rotorbinella as emended by Douglass and Sliter (1965). OCCURRENCE.—Rare throughout the Calvert, Choptank, and St. Marys formations in Mary- land. A single specimen was found in the lower part of the Yorktown Formation in Virginia. STRATIGRAPHIC RANGE.—The range is from up- permost lower Miocene (zone N8) to lower Pli- ocene (zone N19); however, all occurrences except for one specimen are pre-Pliocene. Genus Epistominella Husezima and Maruhasi, 1944 Epistominella danvillensis (Howe and Wallace) PLATE 20: FIGURES 10-12; PLATE 21: FIGURES 9-13 Pulvinulinella danvillensis Howe and Wallace, 1932:71, pi. 13: fig. 7a-c. Epistominella pontoni sensu Schnitker, 1970:72-73, pi. 6: fig. 5a-c. DESCRIPTION.—Test trochospiral, small, ap- proximately equally biconvex with spiral side usually more convex, circular in outline, very slightly lobulate, periphery rounded, umbilical area not depressed, but small clear shell area present where sutures converge in center; 2V2 whorls, 20 to 26 chambers, 8 to 10 in last whorl, chambers slightly inflated, gradually increasing in size, last chamber may project above general umbilical surface; sutures distinct, straight to slightly curved and strongly oblique to periphery and slightly limbate on spiral side, umbilical sutures radial to slightly recurved in later cham- bers, essentially radial in earlier chambers and slightly limbate; aperture a narrow opening par- allel to the periphery of the test; wall finely perforate with pores located in center of surface granules. REMARKS.—This species was described from upper Eocene strata in Louisiana. Subsequent records in Eocene strata are from Georgia (McBean Formation) and Virginia (Chickahom- iny Formation). Epistominella danvillensis is re- ported from Oligocene and lower Miocene strata of Europe; but the illustrations show specimens differing from the paratypes oi E. danvillensis, and it is doubtful that they are conspecific. Four paratypes of E. danvillensis are in the USNM collections, and the present material from the Yorktown and Pungo River formations falls well within the range of morphologic variation of these specimens. 388 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Epistominella pontoni (Cushman) is a similar spe- cies from strata of Miocene and Pliocene age in northwestern Florida (Plate 21: figures 7, 8). It is characterized by having 6 to 7 inflated chambers in the last whorl, depressed, nonlimbate sutures on both the spiral and umbilical sides, lobulate periphery, and a depressed umbilicus. Epistomi- nella danvillensis has less convexity on the umbilical side; lacks a depressed umbilicus; has more lim- bate sutures on the umbilical side and unde- pressed sutures on the spiral side; has fewer and more compressed chambers in the last whorl; and a less lobulate periphery. Epistominella danvillensis differs from E. pungoen- sis, new species, by having more chambers in the last whorl, a less lobulate periphery, less limbate sutures, and an imperforate umbilical area. OCCURRENCE.—This species composes less than 1 percent of the assemblage in the lower and middle parts of the Yorktown Formation in Vir- ginia and North Carolina and in one sample from the Pungo River Formation in North Carolina; however, in the Yorktown Formation in the Lee Creek Mine in North Carolina it composes as much as 9 percent of the assemblage. The occur- rences appear to be in strata and areas represent- ing greater water depths in the Yorktown and Pungo River formations. STRATIGRAPHIC RANGE.—The occurrences in the Pungo River and lower and middle parts of the Yorktown formations give a documented range of upper lower Miocene (zone N8) to pos- sibly middle Pliocene (zones N20-N21), in addi- tion to the Eocene range ofthe species in the Gulf Coast. Epistominella pungoensis, new species PLATE 20: FIGURES 13-15; PLATE 21: FIGURES 1-6 DESCRIPTION.—Test trochospiral, small, bicon- vex with spiral side slightly to considerably more convex, circular to slightly oval in outline, slightly to moderately lobulate, periphery rounded, um- bilical area flush to slightly depressed, containing moderate to large area of imperforate clear shell material from coalescing of limbate sutures; 3 whorls, 22 to 26 chambers, 6 to 7 in last whorl, chamber slightly inflated, gradually increasing in size; sutures distinct, slightly limbate and straight to slightly curved and strongly oblique to periph- ery on spiral side, radial and moderately to strongly limbate on umbilical side; aperture a moderately wide slit parallel to the periphery of the test; wall coarsely perforate. REMARKS.—This species is characterized by having 6 to 7 chambers in the last whorl and moderately to strongly limbate sutures on the umbilical side with a large area of imperforate shell material in the umbilical region. Epistominella pungoensis differs from E. danvillen- sis (Howe and Wallace) by having fewer and wider chambers in the last whorl (6 to 7 compared with 8 to 10), a more lobulate periphery, more strongly limbate sutures on the umbilical side, a larger area of imperforate shell material in the umbilical region, and possibly a slightly depressed umbilicus. Epistominella pontoni (Cushman) differs in hav- ing more inflated chambers, less limbate sutures on the umbilical side, and lacking the imperforate shell material in the umbilical area. OCCURRENCE.—The species is common (ap- proximately 8 percent) in the Pungo River For- mation in southeastern Virginia in the Norfolk Moores Bridge Well at depths of 581 to 616 feet (177 to 187.7 m) and in the upper part of bed 3 of the Fairhaven Member of the Calvert Forma- tion at Randle Cliffs, Maryland, along the Ches- apeake Bay. STRATIGRAPHIC RANGE.—The range is upper- most lower Miocene (zone N8) to lower middle Miocene (zone N9). TYPE-LOCALITY.—The type-locality for the hol- otype and figured and unfigured paratypes is from a core sample at 610 foot (186 m) depth in the Norfolk, Virginia, Moores Bridge Well. TYPES.—The holotype is USNM 252521; fig- ured paratypes are USNM 252522-252525; 10 unfigured paratypes are USNM 252526; all of which are from USGS locality 26003. NUMBER 53 389 Genus Rosalina d'Orbigny, 1826 Rosalina cavernata (Dorsey) PLATE 10: FIGURE 4; PLATE 17: FIGURES 14-16 Discorbis cavernata Dorsey, 1948:311, pi. 37: fig. 2a-c. DESCRIPTION.—Test planoconvex, spiral side of low convexity with all chambers visible, umbilical side with large cavernous depression containing large bulbous growths on broad chamber flaps, test oval in outline, periphery narrowly rounded; 5 to 6 chambers in last whorl, increasing rapidly in size; sutures distinct, slightly depressed, strongly recurved on spiral side, less recurved and slightly limbate on umbilical side; aperture a low slit under plate-like extension of last chamber, extending to periphery, previous apertures some- times visible; wall coarsely perforate with a brown chitinous inner lining. DISCUSSION.—The presence of a relatively open umbilicus, broad chamber flaps, and occasional remnants of earlier apertures places this species in Rosalina d'Orbigny as interpreted by Loeblich and Tappan (1964:584). OCCURRENCE.—Rare throughout the Plum Point Marl Member of the Calvert Formation in Maryland and in the lower part of the Calvert Formation in the Hammond Well on the Eastern Shore of Maryland. STRATIGRAPHIC RANGE.—The range is upper- most lower to lower middle Miocene (zones N8 to N9). Genus Cancris Montfort, 1808 Cancris sagra (d'Orbigny) PLATE 18: FIGURES 1-3 Rotalina sagra d'Orbigny, 1839a:77, pi. 5: figs. 13-15. Cancris sagra (d'Orbigny).—McLean, 1956:359, pi. 48: figs. 3-5, 7. DESCRIPTION.—Test trochospiral, biconvex, moderately elongated to almost oval in outline; 6 to 7 chambers in last whorl, moderately com- pressed to highly inflated, rapidly enlarging; pe- riphery broadly rounded to moderately com- pressed with keel, entire or lobulate; umbilical aperture a low slit under broad lip, which may extend over much of umbilicus; wall finely per- forate. OCCURRENCE.—This species is common to rare in most samples of the Yorktown, Duplin, and Waccamaw formations in Virginia and North Carolina. STRATIGRAPHIC RANGE.—The range is lowest Pliocene, zone N19, into the modern fauna. Genus Elphidium Montfort, 1808 REMARKS.—Because of the presence of areal openings on the septal face, several workers sug- gested that the following new species be placed in Cribroelphidium. The other features of the test, however, show a close relationship to the type- species and related species of Elphidium. In addi- tion, examination of another species of Elphidium from the Croatan Formation in the Lee Creek Mine reveals that specimens of undoubtedly a single species range from forms with a row of 3 or 4 openings at the base of the septal face to forms having both basal and areal openings. The char- acter of the aperture alone, therefore, is not con- sidered sufficient justification to place such closely related species in different genera. This view is also stated in Loeblich and Tappan (1964:637): "Some species previously placed in Cribroelphidium by reason of the presence of a multiple areal aperture belong to Elphidium, as shown by the presence of retral processes and a complex canal system. ..." In addition, Loeblich and Tappan (1964:632), in the description of Elphidium, stated that the "aperture consist[s] of row of pores at base of septal face, earlier septa may also have areal foramina due to resorption." The illustrated specimens in the following species have a broken last chamber, exposing the previous septal face, and this may be the cause for the exposed face showing areal apertures. 390 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Elphidium neocrespinae, new species PLATES 7, 19: FIGURES 3, 4 DESCRIPTION.—Test free, planispiral, involute, much compressed, sides flattened, nearly parallel; periphery subacute with a broadly rounded keel; umbilical regions flattened, not excavated, con- taining a variable number of small bosses; cham- bers usually distinct, although in earlier parts of last whorl sutures may be partially obscured by development of retral processes; number of cham- bers in final whorl variable from about 10 to 15, chambers relatively narrow and curved; sutures distinct, slightly elevated, strongly curved; retral processes distinct, numbering 11 to 15 in the later chambers, covering much of the surface of the chamber, extent variable from almost entirely across the chamber in some specimens to less than halfway in others; retral processes contain spines on their sides, extending to considerable depths into the canal region, but none on the top; low, rounded papillae common on the chamber sur- face; apertural face roughly triangular, slightly concave with numerous low, rounded papillae; aperture consists of a varying number of small rounded openings, both at the base of the aper- tural face, usually 3 in number, and as areal openings above the base, from 3 to 10; apertural openings have a complete raised rim, which may be accompanied by low papillae on the sides; wall calcareous, radial, finely perforate. REMARKS.—Elphidium crespinae Cushman, known from middle Oligocene to middle Miocene strata in Australia, differs from E. neocrespinae in having more chambers in the last whorl (16 to 18), narrower chambers, more strongly biconvex shape with a more acute periphery and a stronger keel, in having a depressed umbilicus in most specimens, and commonly a more lobulate pe- riphery. The morphologic variation of most char- acters within a population sample of E. crespinae is considerably greater than in samples of E. neocrespinae. The wide range of morphology in E. crespinae places the extreme specimens of that species very close to those of E. neocrespinae. Elphidium subplanatum Cushman, described from the upper Oligocene to lower Miocene of Ger- many, differs in having more chambers, about 20 to 22 in the final whorl, and a wider area of more conspicuous bosses in the umbilical region. The probable continuation of the lineage in the Aus- tralasian area is through E. crassatum Cushman, described from middle Miocene to middle Plio- cene strata in Australia, which differs in having more chambers in the last whorl, about 20, a slightly depressed umbilicus containing a low umbo, and a more biconvex test with a more acute periphery. Species morphologically similar to E. neocrespinae are found among later Cenozoic faunas in widely distributed areas. Species de- scribed include: E. novo-zealandicum Cushman, known from late Miocene to Holocene faunas in New Zealand and a likely continuum of the lineage in the Australasian area, which differs from E. neocrespinae in having 20 or more cham- bers in the final whorl and a more biconvex test with a depressed umbilical region containing re- ticulate ornamentation; E. jenseni (Cushman), from living assemblages in Samoa and from Pli- ocene and Pleistocene strata of Japan (Asano, 1938) which differs in having more chambers and a more acute and keeled periphery; E. pustulosum Cushman and McCullock, from the Holocene of the west coast of North and South America, which differs in having more inflated later cham- bers, a stronger and sharper keel, and less well- developed retral processes; and E. earlandi Cush- man, from the Atlantic Ocean off the coast of Spain, which differs in having fewer chambers and much less-developed retral processes. The entire group of species of Elphidium with strong retral process development, including the above discussed species and numerous others in the E. macellum-E. crispum complex, have a wide geographic and stratigraphic distribution. Al- though Cushman (1939) states that the Eocene species of Elphidium have poorly developed retral processes, subsequent descriptions have shown specimens in the Pacific area from Eocene strata with well-developed retral processes. This group possibly originates with E. hampdenense Finlay, from the upper lower Eocene of New Zealand, NUMBER 53 391 and continues in geographic range through slightly younger E. aguafrescaense Todd and Kni- ker and E. skyringense Todd and Kniker, from the middle Eocene of Chile. The above three species are strongly biconvex; among the first species with a compressed test are the middle to late Oligocene forms, including E. crespinae from Aus- tralia and E. subplanatum from Germany, indicat- ing a spread into Europe. The development of species with strong retral processes continued, but the geographic extent, as in the earlier species, is mainly in the Pacific region, including the west coast of North America and through the Medi- terranean area into the Atlantic coast of Europe or Africa. The species of Elphidium in Eocene and Oligocene strata in the Atlantic Coast of the United States generally have poorly developed retral processes, and the E. macellum-E. crispum complex does not make an appearance in this area, even in the Holocene, except for a distant relative, E. advenum (Cushman) [or E. fimbriatulum (Cushman)], which is found in Pliocene and younger strata and E. neocrespinae. The ancestral form of E. neocrespinae does not occur in the Atlan- tic Coast area of North America as far as can be determined from the earlier species of Elphidium. Although E. neocrespinae is found during an inter- val in the late Pliocene to early Pleistocene, it is absent from the later Pleistocene and Holocene deposits of the Atlantic Coast. This relatively short time range for the species makes it useful for correlation. Faujasina compressa Margerel (1971) is similar in many morphologic features to E. neocrespinae, such as canal system and sutural arrangement, aper- tural characteristics, and general surface orna- mentation, but differs in being slightly plano- convex with a spiral side. The group of three species of Faujasina discussed by Margerel appears in the Pliocene and Pleistocene deposits of north- ern Europe, close in time to the appearance of E. neocrespinae in the eastern United States. OCCURRENCE.—Moderately rare throughout the area, composing up to 2 percent of the fora- miniferal assemblages, but it is found in most samples and, with its relatively large size, is a conspicuous member of the fauna. Specimens are found in the following units: the Croatan For- mation at the Lee Creek Mine, North Carolina; the Waccamaw Formation in southern North Carolina at Walkers Bluff and Neils Eddy Land- ing on the Cape Fear River, an outcrop 1 mile (1.6 km) east of Neils Eddy Landing, a marl pit at Acme, marl pits at Old Dock, and the Pierce Brothers Quarry 8 miles (12.8 km) southwest of Wilmington; the Waccamaw Formation in South Carolina at Tillys Lake; the James City Forma- tion of DuBar and Solliday (1963) at James City, North Carolina; and the uppermost part of the "Yorktown" Formation along the Chowan River, North Carolina at Colerain Landing, Mt. Gould Landing, one-half mile (0.8 km) south of Mt. Gould Landing, and Black Rock Landing. STRATIGRAPHIC RANGE.—Based upon the ranges of the co-occurring planktonic Foramini- fera, E. neocrespinae has a probable range of zones N21 (upper Pliocene) and N22 (lower Pleisto- cene). Elphidium neocrespinae is absent from the many samples examined from the Duplin For- mation in North and South Carolina and the Yorktown Formation in North Carolina and Vir- ginia, most of which are dated as belonging to zones N19/N20 and probably lower N21. The species is also absent from the upper Pleistocene strata in North Carolina, such as exposed at Planners Beach and Terra Ceia, from upper Pleis- tocene deposits in Virginia and Maryland (Lan- gleys Bluff and Cornfield Harbor), and from the living faunas off the coast. TYPE-LOCALITY.—The locality for the holotype and figured paratypes is the Lee Creek Mine, North Carolina, in the Croatan Formation. Un- figured paratypes are from the Waccamaw For- mation at Old Dock and Walkers Bluff, North Carolina, and from the "Yorktown" Formation one-half mile (0.8 km) below Mt. Gould Landing on the Chowan River, North Carolina. TYPES.—The holotype is USNM 240133; fig- ured paratypes are USNM 240130, USNM 240131, and USNM 240132; all of which are from USGS locality 25997. Unfigured paratypes are USNM 240148 from USGS locality 26021, 392 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY USNM 240149 from USGS locality 25926, and USNM 240150 from USGS locality 25927. ocene) to the lower Pleistocene strata of the Croa- tan and Waccamaw formations. Elphidium compressulum Copeland PLATE 19: FIGURES 5, 6; PLATE 21: FIGURES 14, 15 Elphidium compressulum Copeland, 1964:262-263, pi. 37: figs. 3a,b. DESCRIPTION.—Test planispiral, involute, strongly compressed; outline an elongate oval, not lobulate in early portion but may be slightly to moderately so in later portions; periphery nar- rowly rounded; umbilical region may be slightly depressed, commonly containing one small boss or may be completely filled with clear shell ma- terial flush with chamber walls; chambers com- pressed in early portion of final whorl, later be- coming slightly inflated; 9 to 12 chambers in last whorl; sutures distinct, slightly depressed to flush with surface, may be thickened, slightly to mod- erately recurved; short retral processes well de- veloped; aperture composed of small rounded openings at base of apertural face as well as areal openings; wall moderately coarsely perforate. DISCUSSION.—The extreme compression of the test, the presence of short and broad, but well- defined retral processes, thickened sutures below the retral processes, and coarse wall porosity char- acterize this species. Some specimens from the Waccamaw and Croatan formations are less com- pressed and have an umbilical umbo projecting above the chamber surface. OCCURRENCE.—The initial description of this species was from the Duplin Formation in south- ern North Carolina, and is most widely found there, in abundances up to 3 percent. It is fairly widespread in abundances of less than 1 percent in the Waccamaw and Croatan formations in central and southern North Carolina. The species occurs at some localities in the upper part of the Yorktown Formation, particularly those near Suf- folk in southeastern Virginia. STRATIGRAPHIC RANGE.—The known range of this species is from the upper part ofthe Yorktown and Duplin formations (probably middle Pli- Elphidium gunteri Cole PLATE 19: FIGURES 7, 10 Elphidium gunteri Cole, 1931:34, pi. 4: figs. 9, 10.—Cushman, 1939:49, pi. 13: fig. 10.—Copeland, 1964:263, pi. 37: fig. 4a,b. DESCRIPTION.—Test free, planispiral, involute, slightly compressed; periphery broadly rounded, not lobulate; umbilical regions usually filled with a series of large pustules, as many as 15, occasion- ally containing only a single large boss or entirely lacking any filling; chambers distinct, 12 to 15 in last whorl; sutures usually radial, sometimes slightly recurved, well marked by retral processes; retral processes well developed along entire su- tural area, may be raised above chamber surface; wall rather coarsely perforate; aperture composed of rounded openings at base of apertural face. DISCUSSION.—Test shape, type of umbilical fill- ing, and development of the retral processes vary in E. gunteri. The test varies from a very broad oval shape with almost parallel sides and a broadly rounded periphery to a more compressed shape that tapers from the umbilical region to- ward the more narrowly rounded periphery. The most common type of umbilical filling is a series of about 10 moderate-sized clear pustules; how- ever, in some specimens there may be several pustules or just one large pustule or boss, or in smaller specimens nothing in the umbilical area. The retral processes vary in development; they may be very strong and extend across much of the chamber and have a rounded shape that is raised above the chamber surface or be consider- ably shorter and have a flattened shape that is flush with the chamber surface. OCCURRENCE.—This species is commonly found in low frequencies, up to 3 percent, in most ofthe samples from the Croatan Formation in the Lee Creek Mine, the Duplin and Waccamaw forma- tions in southern North Carolina, and the upper- most part of the "Yorktown" Formation in the NUMBER 53 393 beds along the Chowan River in northeastern North Carolina. STRATIGRAPHIC RANGE.—In addition to its range from the middle Pliocene (Duplin) to the lower Pleistocene (Croatan, Waccamaw) in North Carolina, the species is known from the Caloosa- hatchee Formation (upper Pliocene and lower Pleistocene) in Florida and also is found living in the Gulf of Mexico. The range for the southeast- ern United States would be from middle Pliocene to Holocene. Elphidium latispatium latispatium Poag, new status PLATE 22: FIGURES 3, 11, 12, 15, 16 Elphidium latispatium Poag, 1966:415, pi. 7: figs. 3, 4. Elphidium cf E. poeyanum (d'Orbigny).—Cushman and Mc- Glamery, 1938:106, pi. 25: figs. 5, 7. DESCRIPTION.—Test planispiral, involute, mod- erately compressed; periphery broadly rounded, slightly lobulate; umbilical regions flush to slightly depressed, commonly containing 3 or 4 irregular bosses; chambers distinct, increasing slowly in size, 10 to 12 in last whorl; sutures incised, slightly recurved; retral processes well developed along most of suture except for periph- ery; processes relatively short, may have slight inflation at proximal end giving bulbous appear- ance; aperture composed of small, rounded open- ings at base of apertural face; wall calcareous, finely perforate. REMARKS.—A comparison was made between Cushman and McGlamery's specimens and other specimens from the Chickasawhay Formation at Choctaw Bluff and a topotypic population sam- ple of E. latispatium latispatium, described from the overlying Paynes Hammock Formation. There is a close similarity in most characteristics, except that the specimens from the Chickasawhay at Choctaw Bluff generally have slightly shorter re- tral processes and a more bulbous shape (Plate 22: figures 11, 12, 15). This subspecies has shorter retral processes than those found in E. latispatium pontium. OCCURRENCE.—Poag (1966) recorded this form from the Chickasawhay Formation in Mississippi, the Paynes Hammock Formation in Alabama and Mississippi, and the Tampa Limestone in Florida. The subspecies is also found in Cushman and McGlamery's material from the Chickasaw- hay Formation in Alabama. Examination of the Smithsonian collections yielded specimens iden- tified as Elphidium species from the Frio Clay in Texas, which also belong to this taxon. The spec- imens are from a depth of 5805 to 5810 feet (1769 to 1770.8 m), described as 600 feet (182.8 m) below the top of the Frio in Magnolia Petroleum Company's Corpus Christi Bank #2, Plymouth Field, San Patricio County, Texas. A note on the back of the slide states that this form is found throughout the Frio. This form has not been recognized in the Atlantic Coast. STRATIGRAPHIC RANGE.—This subspecies ranges through the Chickasawhay and Paynes Ham- mock formations, which were dated by Poag (1972a:266) as late Oligocene in age, equivalent to zones N2-N3. Elphidium latispatium pontium, new subspecies PLATE 12: FIGURES 15, 16; PLATE 13: FIGURES 8, 9; Plate 19: FIGURES 8, 9; PLATE 22: FIGURES 9, 10, 13, 14 Elphidium poeyanum d'Orbigny.—Dorsey, 1948:302, pi. 35: figs. 7a, 8b. DESCRIPTION.—Test planispiral, involute, slightly to moderately compressed; periphery broadly rounded, slightly lobulate; umbilical re- gions flush to slightly depressed, commonly con- tain 3 or 4 irregular bosses; chambers distinct, increasing slowly in size, 10 to 12 in last whorl; sutures incised, usually radial to slightly recurved in later chambers, slightly to moderately recurved in earlier; retral processes well developed along entire sutural area, bridges vary in width from moderately thin to broad; aperture composed of small, rounded openings at base of apertural face; wall finely perforate, calcareous, optically radial. REMARKS.—This species is the oldest Elphidium 394 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY in the Oligocene and Miocene of the central Atlantic region, and is followed by a number of species in the uppermost Miocene and Pliocene of Virginia and North Carolina. This subspecies differs from E. latispatium lati- spatium by having longer retral processes that continue across the periphery, and by the later sutures being less recurved to radial. The popu- lation from the Silverdale beds of Vokes (1969), although belonging to E. latispatium pontium, has individuals with short retral processes approach- ing those found in E. latispatium latispatium. Dorsey (1948) and Cushman and McGlamery (1938) incorrectly placed this subspecies in E. poeyanum. Elphidium poeyanum differs in having fewer chambers (7 to 9) in the last whorl, a more coarsely perforate wall, a more depressed umbil- icus, a more elongate test outline, and more re- curved sutures in the later part of the test. Several species described from the Oligocene and Miocene of Europe have characters similar to this species. They include Elphidium antoninum (d'Orbigny), E. listerii (d'Orbigny), E. hauerinum (d'Orbigny), E. rugosum (d'Orbigny), E. minutum (Reuss), E. latidorsatum (Reuss), and E. articulatum (d'Orbigny). These species all differ from E. lati- spatium in variable combinations of characters, but do belong in a closely related grouping. Twelve specimens on a slide in the Smithsonian USNM collections, labeled only '^E. rugosum d'Orbigny, Sarmat," do not belong to that species but do belong to E. latispatium pontium, confirming the European connection of this group. Elphidium gunteri Cole differs in having longer and stronger retral processes, a more inflated test shape, and generally more umbilical bosses. Elphidium excavatum (Terquem) includes 5 sub- species (Feyling-Hanssen, 1972). In addition to the nominate form, they are E. excavatum clavatum Cushman, E. excavatum selseyensis (Heron-Allen and Earland), E. excavatum alba Feyling-Hanssen, and E. excavatum lidoensis Cushman. Although these taxa show considerable variation among themselves, they differ as a group from E. lati- spatium by having poorly developed retral processes, fewer chambers, papillae in umbilical. apertural, and sutural areas, more convexity of test in umbilical regions, and generally either a large boss or many small bosses in the umbilicus. OCCURRENCE.—This subspecies occurs in the Silverdale beds of Vokes (1967) in Jones, Carteret, Craven, and parts of adjacent counties in the central part of eastern North Carolina, and in the lower and middle parts of the St. Marys Forma- tion in Calvert and St. Marys counties, Mary- land. STRATIGRAPHIC RANGE.—The range of this sub- species is from the upper Oligocene and lower Miocene (zones N3-N4), to probable middle Mio- cene (post-zone Nil). TYPE-LOCALITY.—The type locality for the hol- otype and some of the figured and unfigured paratypes is Langley's Bluff on the western shore of Chesapeake Bay, St. Marys County, Maryland, in the St. Marys Formation. The other figured and unfigured paratypes are from the Silverdale beds of Vokes (1967) southeast of Mays ville, Jones County, North Carolina, on the Long Point Road to the White Oak River. TYPES.—The holotype is USNM 252571 from USGS locality 25955; figured paratypes are USNM 252572 and 252573 from USGS locality 25955, and USNM 252574 and 252575 from USGS locality 22294. Unfigured paratypes are USNM 252576 from USGS locality 25955, and USNM 252577 from USGS locality 22294. Elphidium limatulum Copeland PLATE 11: FIGURES 9-12 Elphidium limatulum Copeland, 1964:263-264, pi. 37: fig. 5a-b. DESCRIPTION.—Test free, planispiral, involute, moderately compressed; periphery broadly rounded; umbilical regions slightly to moderately depressed, containing 20 to 40 pustules of varying sizes; chambers increasing gradually in size, num- bering 9 to 11 in last whorl; sutures deeply incised, moderately recurved; retral process development varies from the presence of 4 to 6 short, broad processes in the later chambers of larger speci- NUMBER 53 395 mens to the total absence in medium to small specimens; aperture composed of small, rounded openings at the base of the apertural face; wall glossy, finely perforate. REMARKS.—The development of the retral pro- cesses is variable in this species. Large adult spec- imens usually have short, broad retral processes, particularly near the umbilicus in the later cham- bers (Plate 11: figure 11), but many medium- to small-sized individuals lack them completely (Plate 11: figure 9). The glossy wall, large number of umbilical pustules, and weak development of retral processes characterize the species. OCCURRENCE.—Rare to common throughout the Croatan Formation in the Lee Creek Mine, the Waccamaw and Duplin formations in south- ern North Carolina, and in the uppermost part of the "Yorktown" Formation as exposed along the Chowan River in North Carolina. STRATIGRAPHIC RANGE.—The known range of this species is from the Duplin Formation (prob- ably middle to upper Pliocene) to the lower Pleis- tocene (Croatan and Waccamaw formations). Genus Cibicides Montfort, 1808 REMARKS.—During studies of the species of Cibicides on the scanning electron microscope, the presence of structures in the large pores was noted, first in C. pungoensis, new species, and subsequently in the other species as they were examined. The structures in the pores are sieve- like in appearance. The pore pattern in the sieve plate is generally consistent within each of the species, although it differs among the species. The jieve plates are most readily visible on the umbil- ical side, particularly on the last several cham- bers. They are less prominent on the spiral side, also occurring most noticeably in the last several chambers, probably because the sieve plates are near the surface in the last two or three chambers, but are more deeply recessed in the earlier cham- bers in the whorl. The arrangement of the pores in the sieve plates differs among species. In C. pungoensis, new species, the pores are small and numerous, as many as 20 to 25, mostly round with some elon- gate, and some of the interpore areas are consid- erably raised (Plate 13: figures 3, 10). In C. cravenensis, new species, the pores are large and few in number, from 1 to 4, mostly irregular in shape, and have a raised collar around the outside ofthe wall pore (Plate 14: figures 5, 7, 9). In C. croatanensis, new species, the sieve pore pattern is more variable, particularly between <^ht spiral and umbilical sides. Some specimens have 8 to 10 medium-sized rounded sieve pores with raised ridges between (most commonly on the umbilical side), and others have a thin, flat plate with several irregular openings (most commonly on the spiral side) (Plate 10: figures 8, 10). Several descriptions of structures within the pores of Foraminifera have been made, including the "foraminal plugs" in Elphidium crispum (Jepps, 1942:625-627), "bouchons" in Planorbulina medi- terranensis (Le Calvez, 1938:236, 1947), "dark disks" in Discorbis erecta (Le Calvez, 1947), "sieve plates" in several species (Jahn, 1953), "pore plugs" in Discorbinopsis aguayo (Arnold, 1954a, 1954b), and "sieve plates" in Amphistegina (Han- sen, 1972b). The reported pore structures vary in position and shape, and include structures with similarity to the presently figured ones; however, all the previously described pore structures are of organic composition and thus differ from the present ones, which are largely calcareous. It is not known at present whether calcareous "sieve plates" are widespread in species of Cibicides or restricted to this group of species from the Oligocene through lower Pleistocene in the Atlan- tic Coastal Plain. Cibicides cravenensis, new species PLATE 14 DESCRIPTION.—Test trochospiral, biconvex, spi- ral side slightly to moderately convex, umbilical side moderately to strongly convex; broad, prom- inent umbo on umbilical side; periphery very slightly lobulate, keeled, acute; chambers com- pressed, increasing gradually in size, 11 to 13 in 396 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY last whorl; sutures strongly recurved on both sides, moderately limbate and raised on both sides; aperture a narrow slit along inner margin of spiral side of final chamber and extending as an arch onto umbilical side, with thickened rim; wall calcareous, optically indistinctly radial, coarsely perforate on both sides, pores on both sides commonly containing "sieve plates" with 1 to 5 irregular openings, pores commonly sur- rounded by raised collar. REMARKS.—The distinguishing characters of this species are the broad umbo on the spiral side, convex umbilical side, limbate, raised, curved sutures on the umbilical and spiral sides, keel throughout, and coarse pores on both sides. Cibicides cravenensis, new species, and C. pungoen- sis, new species, differ by the latter having a strongly lobulate and irregular adult form with nonlimbate, depressed sutures on the umbilical side and a flat spiral side that has more strongly limbate and raised sutures. Cibicides elomoensis Rau differs in being smaller in size and having a less conspicuous keel. Cibicides cravenensis is distin- guished from C. falconensis Renz by having a stronger umbo and more limbate sutures on the umbilical side and a more convex spiral side. Cibicides cravenensis is similar to C. floridanus (Cush- man), but has a stronger keel, broader and more highly raised sutures on both sides, and is more coarsely perforate on the umbilical side. OCCURRENCE.—This species is abundant in the subsurface in the limey facies of the Pungo River Formation in the Croatan National Forest area between New Bern and Morehead City, North Carolina. STRATIGRAPHIC RANGE.—The range is upper lower Miocene (zone N8), with a possibility of a slightly lower limit in the lower Miocene. TYPE-LOCALITY.—The holotype and figured and unfigured paratypes are from a well core at a depth of 48.5 to 49.5 feet (14.7 to 15.0 m) near Great Lake in Craven County, North Carolina. TYPES.—The holotype is USNM 252527, fig- ured paratypes are USNM 252528 through 252532, 10 unfigured paratypes are USNM 252533; all of which are from USGS locality 26018. Cibicides croatanensis, new species PLATE 10: FIGURES 5-8, 10; PLATE 20: FIGURES 5-9 DESCRIPTION.—Test trochospiral, plano-convex with umbilical side moderately convex, spiral side commonly has undulating surface consisting of depressions at sutures and strong overgrowths on early chambers; small umbo on umbilical side, more pronounced in early stages; periphery slightly to moderately lobulate, subacute, slight keel throughout; chambers compressed, increas- ing gradually in size, 9 to 10 in last whorl, later chambers sometimes irregular in shape; sutures strongly recurved on spiral side, less recurved on umbilical side; moderately limbate on both sides in early stages, becoming slightly limbate and depressed in later stages; aperture a narrow slit along inner margin of spiral side of final chamber and extending as a broad arch with thickened rim onto umbilical side; wall calcareous, optically indistinctly radial, coarsely perforate on both sides, pores deeply set in wall with thickened ridges between pores, thickening of wall in earlier chambers of last whorl covers many of the pores and is more prevalent on umbilical side, pores on both sides commonly containing sieve plates; on umbilical side, most sieve plates have as many as 10 rounded openings with raised ridges inbe- tween, on spiral side plates have as many as 5 irregular openings in a generally flat plate. REMARKS.—This species is characterized by having coarse pores, deeply set in thickened wall on both sides, a flat spiral side with thickened surface over early chambers, and a moderately convex umbilical side with recurved limbate su- tures. Cibicides croatanensis differs from C. altamiraensis Kleinpell in having coarser pores and more strongly limbate and raised sutures on the umbil- ical side. Cibicides croatanensis is distinguished from C. ornatus (Cushman) by having fewer chambers in the last whorl, coarser pores, and lacking pus- tules on inner chambers on the spiral side. OCCURRENCE.—Rare to common (as much as 4.5 percent) in the Croatan Formation in the Lee Creek Mine. It also occurs in abundances of less than 1 percent in scattered localities in the Wac- NUMBER 53 397 camaw Formation in southern North Carolina, including Acme and Old Dock, and in the up- permost beds ofthe "Yorktown" Formation along the Chowan River in northeastern North Caro- lina at Mt. Gould Landing and Black Rock Land- ing. STRATIGRAPHIC RANGE.—Upper Pliocene to lower Pleistocene (zones N21-N22). TYPE-LOCALITY.—The holotype and figured and unfigured paratypes are from the Lee Creek Mine, North Carolina, in the Croatan Formation. TYPES.—The holotype is USNM 252534, fig- ured paratypes are USNM 252535 to 252539, 10 unfigured paratypes are USNM 252540; all of which are from USGS locality 25997. Cibicides pungoensis, new species PLATE 13: FIGURES 1-7, 10 DESCRIPTION.—Test medium to large, trocho- spiral, spiral side flat to strongly concave, umbil- ical side slightly to strongly convex, becoming partially evolute in later chambers; strongly pro- jecting umbo on umbilical side; in early stages periphery subrounded, slightly lobulate with keel, becoming strongly lobulate and acute with less- developed keel in later stages; chambers increas- ing gradually in size, 9 to 11 in last whorl, later chambers irregular in shape; sutures strongly re- curved on spiral and umbilical sides, limbate and raised in early stages on umbilical side, gradually becoming nonlimbate and moderately depressed in later stages on both sides; aperture a narrow slit along inner margin of spiral side of final chamber and extending as an arch onto umbilical side, with thickened rim; wall calcareous, opti- cally indistinctly radial, coarsely perforate on both sides, pores on both sides containing sieve plates that have small circular to oval pores, as many as 25, and raised ridges between pores. REMARKS.—This species is characterized by the prominent umbo, strongly limbate and raised sutures in the early stages, which become nonlim- bate and depressed in the later stages on the umbilical side; a circular and slightly lobulate early shape, which becomes highly lobulate and somewhat evolute in the later stages, and coarse pores on both sides. The test may have a strongly concave spiral side, a shape attributed in C. loba- tulus (Walker and Jacob) to attachment to sea- weed during growth. The change in outline during growth from circular to irregular and strongly lobulate also is similar to that found in C. lobatulus by Nyholm (1961). Cibicides pungoensis differs in having a strongly developed umbo and limbate sutures, and generally is more coarsely perforate, espe- cially on the umbilical side. Cibicides pungoensis differs from C. cravenenisis in its more lobulate adult shape and a flat to concave spiral side with less limbate and raised sutures. OCCURRENCE.—Rare to common in the "Virginia St. Marys" beds in the Gatesville Well, North Carolina, at depths of 131 to 138 feet (39.9 to 42.0 m). STRATIGRAPHIC RANGE.—Upper Miocene. TYPE-LOCALITY.—The holotype and figured and unfigured paratypes are from the Gatesville Wefl, North Carolina, at a depth of 131.5-132 feet (40.0 to 40.2 m) in the "Virginia St. Marys" beds. TYPES.—The holotype is USNM 252541, fig- ured paratypes are USNM 252542 to 252544, 10 unfigured paratypes are USNM 252545; all of which are from USGS locality 25992. Genus 'Virgulinella Cushman, 1932 Virgulinella miocenica (Cushman and Ponton), new combination PLATE 10: FIGURE 3; PLATE 17: FIGURES 12, 17, 18 Virgulina miocenica Cushman and Ponton, 1931:32, pi. 4: figs. 14-16. Virgulina (Virgulinella) miocenica Cushman and Ponton.— Cushman, 1937:35, pi. 5: figs. 15-16.—Dorsey, 1948:305, pi. 36: fig. 12. DESCRIPTION.—Test elongate, about 3 times as long as wide, tapering toward both ends; early part of test may be straight or curved; chambers inflated, distinct, with numerous arcuate projec- 398 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY tions to previous chambers; sutures distinct, with arcuate pattern; aperture elongate, narrow, straight to slightly curved, extending from base of final chamber nearly to apex, with toothplate; wall smooth, glassy, finely perforate. OCCURRENCE.—Rare to common throughout the Calvert, Choptank, and St. Marys formations in Maryland and northern Virginia, the Pungo River Formation in North Carolina, and through the "Virginia St. Marys" in Virginia. STRATIGRAPHIC RANGE.—The range in the cen- tral Atlantic Coastal Plain is from uppermost lower Miocene (zone N8) to the top of the upper Miocene (probably near the end of zone N17). Genus Astrononion Cushman and Edwards, 1937 Astrononion stelligerum (d'Orbigny) PLATE 11: FIGURES 13-16; PLATE 12: FIGURES 1, 2 Nonionina stelligera d'Orbigny, 1839:128, pi. 3: figs. 1, 2. Astrononion stelligerum (d'Orbigny).—Cushman and Edwards, 1937:31, pi. 3: fig. 7a,b.—Hornibrook, 1964:334, pi. 1: figs. 5-9, 14, 15.—Le Calvez, 1974:37-38, pi. 9: figs. 1-4. DESCRIPTION.—Test free, relatively small, plan- ispiral, involute, moderately to strongly com- pressed; periphery rounded, umbilical regions slightly depressed; chambers increasing gradually in size, 7 to 9 in last whorl; sutures distinctly incised, slightly recurved; in early part of last whorl relatively narrow tubes are present along sutures, extending from umbilicus about one- third of way along the sutures with outer ends of tubes opening into sutures; in later part of last whorl tubes are modified into flattened essentially triangular plates that are attached to chamber on anterior side and open down outermost part of posterior side; plates join in umbilical region forming large plate over area; aperture is low, simple, rimmed, arched slit at base of final cham- ber. REMARKS.—A neotype for this species has been selected by both Hornibrook (1964) and Le Cal- vez (1974). Each noted that the type specimen had "disintegrated." Topotypic material from Teneriffe Island was not available to Hornibrook, so he picked a specimen from the nearby island of Las Palmas as the neotype. As all the conditions for the establishment of a neotype appear to have been met by Hornibrook, his designation of the neotype would stand even though Le Calvez subsequently designated a specimen from the d'Orbigny collection from Teneriffe as neotype. A comparison ofthe illustrations of each neotype shows that both appear to be well within the variation of the species. The specimens in the present material gener- ally have fewer chambers, ranging from 7 to 9 in comparison to 10, and a more pronounced um- bilicus than those illustrated by Hornibrook and Le Calvez. OCCURRENCE.—Except for one specimen found in the lower Pliocene part of the Yorktown For- mation in Virginia, this species is rare in the Croatan Formation in North Carolina (upper Pliocene and lower Pleistocene). STRATIGRAPHIC RANGE.—Although found in the Pliocene and Pleistocene of the Atlantic Coastal Plain, this species has been recorded from lower Miocene strata in Europe and New Zea- land, and continues into the modern fauna. Genus Florilus Montfort, 1808 Florilus chesapeakensis, new species PLATE 11: FIGURES 7, 8; PLATE 18: FIGURES 7, 8, 11, 12 Nonion medio-coslatum sensu Dorsey, 1948:300, pi. 35: fig. 4a-c. DESCRIPTION.—Test relatively large, planispiral to slightly asymmetrical, longer than broad, with width increasing rapidly in later stages, periphery subrounded to subacute, umbilical regions slightly to moderately depressed, containing slight to moderate amount of pustulose material, which may extend onto apertural face; chambers distinct, about 13 in last whorl, much higher than wide, increasing abruptly in width as seen in apertural face, marked by slightly to sharply raised areas between sutures, raised areas extend NUMBER 53 399 from the umbilicus to about halfway to the pe- riphery; sutures distinct, slightly depressed, mod- erately to strongly limbate, especially in earlier chambers in last whorl, moderately recurved, con- taining single row of pustules extending as far as halfway to periphery; apertural face broadly heart-shaped, aperture a low, curved slit at base of apertural face; wall calcareous, finely perforate. REMARKS.—This species is characterized by the slightly to sharply raised areas between the inner parts of the sutures, the limbate sutures, and the abrupt increase in width of the heart-shaped apertural face. It differs from Florilus medio-costa- tum (Cushman) by having generally stronger raised areas between the sutures with the raised areas being found earlier in the last whorl, by more strongly limbate sutures, and in having a much more flaring test in the later chambers with the apertural face almost twice as wide. Florilus costiferus (Cushman) differs in having more cham- bers in the last whorl (approximately 20), lacking the raised chamber areas, and having more strongly limbate sutures. Florilus incisa (Cushman) is similar except for the absence of the raised areas between the sutures. OCCURRENCE.—Rare to common in samples from the Calvert, Choptank, and St. Marys for- mations in Maryland and Virginia, and the Pungo River Formation in North Carolina. STRATIGRAPHIC RANGE.—This species first oc- curs in beds placed in the uppermost lower Mio- cene (zone N8) and continues into upper Miocene strata. TYPE-LOCALITY.—The holotype is from Wind- mill Point on the St. Marys River, Maryland, in the St. Marys Formation. Figured paratypes are from the Calvert Formation at Governors Run, Maryland, and figured and unfigured paratypes are from the "Virginia St. Marys" beds in the Gatesville Well in North Carolina. Unfigured paratypes are from the Choptank Formation at Flag Pond, Maryland. TYPES.—The holotype is USNM 252565 from USGS locality 25992; figured paratypes are USNM 252566 from USGS locality 25983, and USNM 252567 from USGS locality 25969; un- figured paratypes are 5 specimens, USNM 252568 from USGS locality 25992 and 5 speci- mens, USNM 252569 from USGS locality 25962. Genus Nonion Montfort, 1808 Nonion advenum pustulosum, new subspecies PLATE 22: FIGURES 1, 2, 4-6 DESCRIPTION.—Test planispiral, circular to oval in side view, moderately compressed; periphery broadly rounded, slightly lobulate; umbilical re- gion not depressed, filled with a small to occa- sionally moderate-sized boss and moderate to large amount of small pustules, with pustules extending onto excavated sutures and apertural face; 10 to 13 chambers in last whorl; sutures deeply incised on chamber sides, slightly so on periphery, with deep excavation near boss, mod- erately recurved; aperture composed of small, rounded openings at the base of the apertural face; wall calcareous, optically granular, finely perforate. DISCUSSION.—This subspecies is characterized by a generally small umbilical boss, abundant pustulose material, excavated sutures, and 10 to 13 chambers in the last whorl. Populations of Nonion advenum advenum (Cush- man) differ in generally having a larger umbilical boss, a more discoid shape with a subacute pe- riphery, more extensive excavated sutures, and less pustulose material. A specimen from the Chickasawhay Formation at Choctaw Bluff on the Alabama River is illustrated (Plate 22: figures 7, 8). Some specimens of N advenum pustulosum in the populations from North Carolina have a larger umbilical boss and less pustulose material and approach the minimum development of those characters in specimens of N advenum ad- venum from Alabama. Nonion advenum pustulosum is larger than N. cal- vertensis, new species, and has more chambers in the last whorl, a smaller boss, and more pustulose material. Nonion inexcavatum (Cushman and Applin) dif- 400 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY fers in having a more compressed test with a subacute periphery, a larger umbilical boss, and the absence of excavated sutures in the umbilical area. OCCURRENCE.—This subspecies occurs in the "Silverdale-age" beds on the central Atlantic Coastal Plain area of North Carolina, particularly exposures in Onslow, Craven, and Carteret coun- ties. The subspecies is usually abundant, compos- ing more than 50 percent of the total assemblage in some samples. STRATIGRAPHIC RANGE.—The subspecies is re- stricted to the upper Oligocene Silverdale beds of Vokes (1967) (probably zones N2-N3). It is not found in the overlying uppermost lower Miocene Pungo River Formation (zones N8-N10). TYPE-LOCALITY.—The type-locality for the holotype, figured paratypes, and unfigured par- atypes is near Long Point on the White Oak River, Carteret County, North Carolina. TYPES.—The holotype is USNM 252561, fig- ured paratypes are USNM 252562 and 252563, and 10 unfigured paratypes are USNM 252564, all of which are from USGS locality 22294. Nonion calvertensis, new species PLATE 12: FIGURES 3-8; PLATE 19: FIGURES 1, 2 Nonion advenum sensu Dorsey, 1948:299-300, pi. 35: fig. la-c. DESCRIPTION.—Test planispiral, small, nearly circular in side view, moderately compressed; periphery broadly rounded; umbilical region filled with a large clear boss that may be even in height with sides of chambers or project well beyond them; the boss is joined to the chamber sides in the earlier chambers ofthe last whorl, but not in the last 3 or 4 chambers; a small amount of pustulose material is found in the umbilicus, extending up the sutures and onto the apertural face; chambers increasing gradually in size, num- bering 8 to 10 in the last whorl; sutures deeply incised on chamber sides, not on periphery, with deep excavation nearing boss, moderately re- curved; aperture composed of small, rounded openings at the base of the apertural face; wall finely perforate, calcareous, optically granular. DISCUSSION.—This subspecies is characterized by its small size and circular shape with a rounded periphery, 8 to 10 chambers in the last whorl, a prominent umbilical boss in most specimens and having the inner part of the sutures excavated and containing a small amount of pustulose ma- terial. Nonion advenum advenum (Cushman), found in Eocene and Oligocene strata in the southeast- ern United States, shows considerable variation both within and between populations, but gen- erally differs by being larger, as much as twice as large, having more chambers in the last whorl, 10 to commonly 15, and by having a more discoid- shaped test with a subacute periphery. Nonion advenum pustulosum, new subspecies, from the upper Oligocene of North Carolina differs by being larger in size, having more chambers in the last whorl, 10 to 13, by largely lacking a promi- nent boss, and by having a larger amount of pustulose material in the umbilical, sutural, and apertural areas. OCCURRENCE.—The distribution is limited to the lower and middle part of the Calvert For- mation in Maryland, from beds 3 to 10 of Shat- tuck. The highest frequencies are found in the lower beds, reaching a peak of 14 percent in bed 8, and the lowest, less than 1 percent, in bed 10. STRATIGRAPHIC RANGE.—Upper part of bed 3 (probably zone N8) through bed 10 (zone N9) of Calvert Formation. TYPE-LOCALITY.—The type-locality is Randle Cliffs on the western shore of Chesapeake Bay, Calvert County, Maryland, in bed 6 of the Cal- vert Formation. TYPES.—The holotype is USNM 252554; fig- ured paratypes are USNM 252555-252558; all of which are from USGS locality 25980 in bed 6. Unfigured paratypes are USNM 252559 from USGS locality 25980 in beds 5 and 6, and USNM 252560 from USGS locality 26022. Nonion marylandicum Dorsey PLATE 12: FIGURES 9-14; PLATE 18: FIGURES 9, 10 Nonion marylandicum Dorsey, 1948:301, pi. 35: fig. 2a-c. DESCRIPTION.—Test planispiral, involute, mod- erately compressed; periphery broadly rounded; NUMBER 53 401 umbilical regions slightly depressed, filled with small pustules that also extend outward along the sutural areas; chambers may increase gradually in size in last whorl or in large specimens may have a relatively constant size; 7 to 9 chambers in last whorl; sutures deeply incised, moderately recurved; aperture a narrow slit at base of aper- tural face; wall finely perforate. REMARKS.—This species is variable in the size and shape of chambers in the last whorl, largely dependent upon the size of the individual. Smaller specimens have 7 relatively compressed chambers with increasing chamber size and a small to moderate amount of pustulose material (Plate 12: figures 11, 13). Large specimens have 8 or 9 more inflated and fairly equal-sized cham- bers in the last whorl, and a large amount of pustulose material in the umbilical area (Plate 12: figure 9). OCCURRENCE.—Rare throughout the upper part of the Calvert Formation in Maryland and the Choptank Formation in Maryland and Vir- ginia. Dorsey (1948) also reported this species in the overlying St. Marys Formation in Maryland. None of her specimens from the St. Marys For- mation could be found in the Smithsonian collec- tions, but specimens collected by the author from the same localities include two species that could have been misidentified as N marylandicum. One is a species of Nonion, represented by only a few specimens, which differs from N marylandicum in having very few umbilical pustules. The other, represented by more specimens, is a species of Anomalinoides, characterized by a low trochospiral coiling and small amount of pustulose material in the umbilical region. STRATIGRAPHIC RANGE.—The range is from bed 10 ofthe Calvert Formation upward through the Choptank Formation, which gives a middle Mio- cene age, ranging from uppermost zone N8 to lowermost N9 to later than zone Nil. Genus Svratkina Pokorny, 1956 REMARKS.—The only previously known occur- rences of this genus in the New World are of S. lajollaensis Sliter from the Upper Cretaceous of southern California and the living S. decorata (Phleger and Parker) from the Gulf of Mexico. Three new occurrences of this genus in North America were found: S. croatanensis, new species from the upper Pliocene part of the Croatan Formation of North Carolina and Virginia, an as yet undescribed species from upper Oligocene strata in North Carolina, and S. crassicoria (Poag) from the Paynes Hammock Formation (upper Oligocene) of Alabama and Mississippi. The depth distribution of the living species within this genus is variable; S. tuberculata Balkwill and Wright is found in shelf waters and S. decorata is characteristic of depths from 1300 to 3550 m. Four species were placed in Svratkina by Po- korny (1956). One of these, Discorbina turris Kar- rer, 1868, does not appear to belong to this genus based upon examination of a topotype specimen in the Smithsonian USNM collections and com- parison with the original illustration. In D. turris, the suture pattern is radial on both sides and the aperture is round and umbilical in position, char- acteristics not consistent with the generic concept of Svratkina. D. turris probably belongs in Glabra- tella Dorreen, 1948. Svratkina croatanensis, new species PLATE 15 DESCRIPTION.—Test free, trochospiral, oval in outline; about 7 chambers in last whorl; all cham- bers visible on the spiral side, only those of the last formed whorl visible on the umbilical side; both left and right coiled specimens; sutures mod- erately curved on spiral side, radial to slightly curved on umbilical side; spiral side varies from essentially planar to moderately convex, umbili- cal side from slightly to moderately convex; pe- riphery broadly rounded to subrounded; aperture a long narrow opening of variable height and distinctness, extending from near the umbilicus to near the periphery, without lip, but having funnel-like projection into previous chamber; wall calcareous, spiral side coarsely perforate, pores with thickened rims opening on top of low tubular necks; size of pores variable, long slits 402 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY common, probably as a result of connection of a series of pores; pores scattered over most of spiral surface except for early chambers, with linear concentrations sometimes found along sutures; umbilical side largely imperforate, but some large pores are occasionally found near the periphery. REMARKS.—This species is distinguished by the large pores that commonly coalesce into slit-like openings on the spiral side, and few pores on the umbilical side. Svratkina tuberculata (Balkwill and Wright) is distinguished by having strongly ele- vated tubercles over much of the umbilical sur- face. Svratkina australiensis (Chapman, Parr, and Collins) differs in having more pores on the um- bilical side; even, moderate-sized pores on the spiral side; a very convex spiral side; even, mod- erate-sized pores on the spiral side; a very convex spiral side; and more chambers (9) in the last whorl. Svratkina decorata (Phleger and Parker) dif- fers in having few, scattered fine tubercles. Svrat- kina perlata (Andreae) is the most similar species, but has finer pores, which also cover much of the umbilical side. Also, a lesser degree of alignment of pores is seen along the sutural areas on the spiral side. OCCURRENCE.—Abundances of less than 1 per- cent in the lower part of the Croatan Formation in the Lee Creek Mine and at depths between 112.5 and 113 feet (34 and 34.4 m) in the Norfolk, Virginia, Moores Bridge Well. STRATIGRAPHIC RANGE.—The species is re- stricted to the Croatan Formation and uppermost part of the "Yorktown" Formation, indicating a late Pliocene and early Pleistocene age. TYPE-LOCALITY.—The locality for the holotype and some of the figured paratypes is the Lee Creek Mine, in the lower part of the Croatan Formation. Additional figured paratypes are from the Norfolk, Virginia, Moores Bridge Well at a depth between 112.5 and 113 feet (34 and 34.4 m). TYPES.—The holotype is USNM 252546 from USGS locality 25997; figured paratypes are USNM 252547 and 252548 from USGS locality 25997 and USNM 252549-252553 from USGS locality 26001. Register of USGS Localities (Each locality is listed in the following order: number of locality, formation, location and de- scription of collection site, collector, and date of collection.) 22294. Silverdale beds of Vokes (1967). Few inches (cm) of outcrop in bottom of gully south of Long Point Road about 0.5 miles (0.8 km) west of intersection with Haywood Landing Road; in Croatan National Forest, Jones County, North Carolina. P.M. Brown, Druid Wilson, and Alan Rubin, 17 November 1959. 25921. Chickasawhay Formation. Marl above hard lime- stone, Choctaw Bluff, Alabama River, Alabama (from Cushman Foraminifera Collection). 25922. Choctawhatchee Formation. Alice Creek, near N.E. corner of SE VA Sec. 8, T. IN, R. 19W about 1 mile (1.6 km) NE of Old Parmenter Place, Walton County, Florida (from Cushman Foraminifera Col- lection). 25923. Duplin Formation. Natural Well, 2 miles (3.2 km) west of Magnolia, Duplin County, North Carolina. Four feet (122 cm) of bluish clayey sand exposed, sample 3 feet (91 cm) above base. T. Gibson, 1959. 25924. Duplin Formation. Marl pit on Barwick Farm southwest of Kenansville, Duplin County, North Carolina. Four feet (122 cm) of yellow clayey sand exposed, sample 3 feet (91 cm) above base. T. Gibson, 1959. 25925. Waccamaw Formation. Walker's Bluff on southern bank of Cape Fear River 9 miles (14.5 km) below Elizabethtown, Bladen County, North Carolina. Five feet (1.5 m) exposed; lowermost 2 feet (61 cm) of clayey sand with large pelecypods, upper 3 feet (91 cm) of yellow sand contains much shell hash: Sample 1 foot (30 cm) above base of section. T. Gibson, 1959. 25926. (same formation/location as 25925); sample 4 feet (122 cm) above base of section. T. Gibson, 1959. 25927. "Yorktown" Formation. Bluff at Mt. Gould on western bank of Chowan River in Bertie County, North Carolina. Five feet (1.5 m) of yellow clayey sand are exposed in upper part of bluff; sample 3 feet (91 cm) above base. T. Gibson, 1959. 25929. "Yorktown" Formation. Bluff on western side of Chowan River at Black Rock Landing in Bertie County, North Carolina, 1 mile (1.6 km) north of Highway 17 bridge. 9-foot (2.7-m) section exposed; sample IV2 feet (46 cm) above base of 3-foot (91- cm) unit of yellow fossiiiferous sand that extends from 5 to 8 feet (1.5 to 2.4 m) above base of section. T. Gibson, 1959. NUMBER 53 403 25930. (same formation/location as 25929); sample in mid- dle of 2-foot (61-cm) bluish clayey sand unit that extends from 3 to 5 feet (0.9 to 1.5 m) above base of section. T. Gibson, 1959. 25933. Yorktown Formation. South bank of Yorktown River about 1 mile (1.6 km) downstream from 25957. Yorktown at base of Moore House bluff in James County, Virginia. 24 feet (7.3 m) of section exposed: sample 1 foot (30 cm) above base of 3-foot (91- cm)fossiiiferous gray sand unit, which extends from 6 to 9 feet (1.8 to 2.7 m) above base of section, T. Gibson, 1959. 25960. 25934. (same formation/location as 25933); sample 2 feet (61 cm) above base of 4-foot (122-cm) brown sand unit, which extends from 2 to 6 feet (0.6 to 1.8 m) above base of section and contains few fossils. T. Gibson, 1959. 25937. Yorktown Formation. Bluff at Morgart's Beach on south bank of James River, 5 miles (8 km) north of 25962. Smithfield, Isle of Wight County, Virginia. 21.5 feet (6.6 m) of section exposed; sample 1.5 feet (46 cm) above base of 4-foot (122-cm) yellow sandy shell hash unit, which extends from 17.5 to 21.5 feet 25965. (5.3 to 6.6 m) above base of section. T. Gibson, 1959. 25941. Yorktown Formation. Bluff on western bank of Nansemond River 1 '^h miles (2.4 km) NE of High- way 258 bridge in the northern part of Suffolk, Nansemond County, Virginia. 13 feet (4 m) of 25966. yellow clayey sand exposed; lowermost 2 feet (61 cm) sparingly fossiiiferous and upper 11 feet (3.4 m) very fossiiiferous; sample 7 feet (2.1 m) above base of section. T. Gibson, 1959. 25942. (same formation/location as 25941): sample 8 feet (2.4 m) above base of section in Mulinia congesta bed. T. Gibson, 1959. 25948. Yorktown Formation. Maddry's Bluff on south bank ofthe Meherrin River 1 mile (1.6 km) down- 25969. stream from Highway 258 bridge, 1 mile (1.6 km) northeast of Murfreesboro, Hertford County, North Carolina. 37 feet (11.3 m) of section exposed; sam- ple 1 foot (30 cm) above base of 3-foot (91-cm) bluish clayey sand unit at base of section. T. Gibson, 1959. 25971. 25950. St. Marys Formation. Bluff V^ mile (1.2 km) south of Little Cove Point on western shore of Chesapeake Bay, Calvert County, Maryland. 46 feet (14 m) of section exposed; sample 1 foot (30 cm) above base of 5-foot (1.5-m) bluish sandy clay unit that extends from 11 to 16 feet (3.4 to 4.9 m) above base of section. T. Gibson, 1959. 25972. 25955. St. Marys Formation. Langleys Bluff, 5V2 miles (8.8 km) south of Cedar Point on western shore of Chesapeake Bay, St. Marys County, Maryland. 2V2 25974. feet (76 cm) of section exposed; lowest 6 inches (15 cm) of bluish sandy clay contains scattered fossils, upper 2 feet (61 cm) of yellow sandy clay contains concentrated fossil bands; sample 2V2 feet (76 cm) above beach. T. Gibson, 1959. St. Marys Formation. Small bluff V4 mile (0.4 km) SE of Chancellor Point, St. Marys River, St. Marys County, Maryland. 7.5 (2.3 m) feet of section ex- posed; sample 6 inches (15 cm) above base of 2-foot (61-cm) bluish sandy clay unit, which is at base of section. T. Gibson, 1959. Choptank Formation. Bluff V2 mile (0.8 km) south of Flag Pond on western shore of Chesapeake Bay, Calvert County, Maryland. 105 feet (32 m) of section exposed; sample 2 feet (61 cm) above base of 15-foot (4.6-m) unit of fossiiiferous sand (zone 19), which extends from 17 to 32 feet (5.2 to 9.8 m) above base of section. T. Gibson, 1959. (same formation/location as 25960); sample 1 foot (30 cm) above base of 2 feet (61 cm) of bluish green sand exposed to base of section at beach level (zone 17). T.Gibson, 1959. Choptank Formation. Nomini Cliffs, 12.5 miles (2.4 km) west of eastern end of cliffs, on Potomac River, Westmoreland County, Virginia. 70 feet (21 m) of section exposed; sample 25 feet (7.6 m) above base of 30-feet (9.1-m) unit of bluish sandy clay which is at base of section. T. Gibson, 1959. Paynes Hammock Formation. Stop 6, 1975 GCAGS Guidebook. Exposure along western bank of Chick- asawhay River, about 200 feet (61 m) north of U.S. Highway 84 bridge about 2.5 miles (4 km) west of center of downtown Waynesboro, Mississippi; sam- ple in bed 12 of measured section, about 3.5 feet (1.1 m) above the contact with the underlying Chickasawhay Formation. L. Bybell, R. Christo- pher, and C. Smith, 25 October 1975. Calvert Formation. Bluff V4 mile (0.4 km) south of Governor's Run on western shore of Chesapeake Bay in Calvert County, Maryland. 26.5 feet (8.1 m) of section exposed; sample 1.5 feet (46 cm) above base of 3.5-foot (1.1-m) unit of bluish sandy clay, which is at base of section. T. Gibson, 1959. Calvert Formation. Bluff Vt mile (0.8 km) south of Parker Creek on western shore of Chesapeake Bay, Calvert County, Maryland. 30 feet (9.1 m) of sec- tion exposed; sample in middle of 1-foot (30-cm) unit of blue sandy clay, fossiiiferous, which extends from 4 to 5 feet (122-152 cm) above base of section (zone 12). T. Gibson, 1959. (same formation/location as 25971); sample 2.5 feet (76 cm) above base of 4-foot (122-cm) blue clay unit at base of section (zone 11). T. Gibson, 1959. Calvert Formation. Bluff 1 mile (1.6 km) south of 404 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 25975. 25977. 25980. 25981. 25982. 25983. 25985. 25989. 25992. 25993. Plum Point on western shore of Chesapeake Bay, Calvert County, Maryland: sample 2 feet (61 cm) above beach level in 7 feet (2.1 m) of greenish fossiiiferous sand of zone 10. T. Gibson, 1959. Calvert Formation. Bluff 1 mile (1.6 km) north of Plum Point on western shore of Chesapeake Bay, Calvert County, Maryland. 28.5 feet (8.7 m) of section exposed: sample 1 foot (30 cm) above base of 10-foot (3-m) unit of fossiiiferous sand (zone 10), which extends 18.5 to 28.5 feet (5.6 to 8.7 m) above base of section. T. Gibson, 1959. (same formation/location as 25975); sample 6 feet (1.8 m) above base of 10-foot (3-m) unit of bluish green clayey sand (zone 8), which extends from 6 to 16 feet (1.8 to 4.9 m) above base of section. T. Gibson, 1959. Calvert Formation. Bluff south of Chesapeake Beach at Randle Cliffs Community Association Church, western shore of Chesapeake Bay, Calvert County, Maryland: sample 8 feet (2.4 m) above Ostrea percrassa bed (zone 4), 1 foot (30 cm) above base of thickly packed Corbula bed 3 feet (91 cm) thick. T. Gibson, J. Ayres, 1 July 1963. (same formation/location as 25980); sample in dia- tomaceous sediments, 4 feet (122 cm) below Ostrea percrassa bed. T. Gibson, J. Ayres, 1 July 1963. Calvert Formation. Bluffs about 1.5 miles (2.4 km) south of Plum Point on western shore of Chesapeake Bay, Calvert County, Maryland. 2 feet (61 cm) of bed 10 exposed above beach level: sample 1 foot (30 cm) below top of bed 10. T. Gibson, J. Ayres, 1 July 1963. St. Marys Formation. Bluff at Windmill Point on west side of St. Marys River, St. Marys County, Maryland. 5 feet (1.5 m) of section exposed in bank; sample 1 foot (30 cm) above base of 4-foot (122-cm) blue clayey sand unit, which is at base of section. T. Gibson, J. Ayres, 1 July 1963. Choptank Formation. Section at step area at Bal- timore Gas and Electric Company Nuclear Power Plant site south of St. Leonards, Calvert County, Maryland, on western shore of Chesapeake Bay. Sample taken 1 foot (30 cm) below top of zone 17. T. Gibson, 1969. Waccamaw Formation. Pierce Bros, marl pit, about 3 miles (4.8 km) west of Town Creek on road to Winnabow, Brunswick County, North Carolina. 6.5 feet (2 m) of section exposed; sample 1 foot (30 cm) below top of 3-foot (91-cm) blue sand unit, which extends to water level. T. Gibson, D. Wilson, 22 June 1963. Pungo River Formation. Gatesville Well, North Carolina; hole located on south end of Gatesville, Gates County: sample at 131.5 feet (40.1 m) below well collar. T. Gibson and others, 1967. (same formation/location as 25992); sample 138.0 feet (42.1 m) below well collar. T. Gibson and others, 1967. 25995. "Yorktown" Formation. Texasgulf Inc. Lee Creek Mine; NW wall of test pit, Beaufort County, North Carolina. 21 feet (6.4 m) of section at top of pit; sample 1 foot (30 cm) below top of 3- to 4-foot (92- 122-cm) greenish blue clayey sand unit which is 2 feet (61 cm) below top of section; top of section marked by brown cross-bedded sand. T. Gibson, 5 December 1963. 25997. (same formation/location as 25995); sample 5 feet (1.5 m) below top of 6- to 7-foot (1.8-2.1-m) blue clayey sand unit, which extends from 7 to 14 feet (2.1 to 4.3 m) below top of section. T. Gibson, 5 December 1963. 26001. "Yorktown" Formation. Moores Bridge Well, Moores Bridge Pumping Station, Norfolk, Virginia; sample 112.5 feet (34.3 m) below collar, blue-green clayey fine sand, fossiiiferous. T. Gibson, 1967. 26002. Pungo River Formation, (same location as 26001); sample 581 feet (177.1 m) below collar, dark olive- green silty clay. T. Gibson, 1967. 26003. (same formation/location as 26002); sample 610 feet (239.3 m) below collar, olive-green phosphatic sand. T. Gibson, 1967. 26009. Yorktown Formation. Texasgulf Inc. Lee Creek Mine, NW wall of test pit, Beaufort County, North Carolina; sample 6 feet (1.8 m) below top of 10-foot (3-m) unit of blue clayey sand which extends from 1 to 11 feet (0.3 to 3.4 m) above base of Yorktown Formation. T. Gibson, 1963. 26010. (same formation/location as 26009); sample 9 inches (22.9 cm) below top of 1-foot (30-cm) blue sand unit containing Placopecten clintonius, which is lowest unit of Yorktown Formation. T. Gibson, 1963. 26012. Pungo River Formation, (same location as 26009); sample 5 feet (1.5 m) below top of 6-foot (1.8-m) yellow-green shell hash and phosphatic sand unit, which is at top of Pungo River Formation. T. Gibson, 1963. 26013. (same formation/location as 26012); sample at base of 6-foot (1.8-m) yellow-green shell hash and phos- phatic sand unit, which is at top of Pungo River Formation. T. Gibson, 1963. 26014. (same formation/location as 26012); sample at top of 3-foot (91-cm) unit of interbedded limestone and phosphatic sand beds that extend from 9 to 12 feet (2.3 to 3.7 m) below the top of the Pungo River Formation. T. Gibson, 1963. 26018. Pungo River Formation. Well C181, near Great Lake, Craven County, North Carolina; 48.5 feet (12.6 m) below collar; in unit of 6.5 feet (2 m) of yellow-green sandy shell hash. T. Gibson and oth- ers, 1967. 26019. Choptank Formation, Bartein's Landing, Mary- 26018. NUMBER 53 405 land. From interior of molluscan shells in Smith- sonian USNM collections. 26020. Calvert Formation. Well core at Baltimore Gas and Electric Company Calvert Cliffs Nuclear Power Plant south of St. Leonards, Calvert County, Mary- land; sample 43 at 165'10"-166'1" below (50.5-50.6 m) collar. T. Gibson and F. Whitmore, September 1967. 26021. Waccamaw Formation. Marl pit Vi mile (0.8 km) north of Old Dock, Columbus County, North Car- olina; sediment is largely coarse quartz sand; me- gafossils fairly common. T. Gibson, 1959. 26022. Calvert Formation. Bluff south of Chesapeake Beach at Randle Cliffs Community Association Church, western shore of Chesapeake Bay, Calvert County, Maryland; sample in 7-foot (2.1-m) unit with bands of Corbula more common near top; this unit immediately above Ostrea percrassa bed and corresponds to zone 5; sample 1 foot (30 cm) above base. T. Gibson, J. Ayres, July 1, 1963. 26023. Calvert Formation. 1 mile (1.6 km) south of Plum Point, Calvert County, Maryland: zone 12. A. Dor- sey. 26022. Literature Cited Akers, W.H. 1955. Some Planktonic Foraminifera of the American Gulf Coast and Suggested Correlations with the Caribbean Tertiary. Journal of Paleontology, 29(4):647-664, plate 65, 3 figures. 1972. Planktonic Foraminifera and Biostratigraphy of Some Neogene Formations, Northern Florida and Atlantic Coastal Plain. Tulane Studies in Geology and Paleontology, 9(1-4): 1-139, plates 1-60. Akers, W.H., and P.E. Koeppel 1973. Age of Some Neogene Formations, Atlantic Coastal Plains, United States and Mexico. In L.A. Smith, and Jan Hardenbol, editors, Proceedings of Symposium on Calcareous Nannofossils, pages 80-93, plates 1-4. Houston, Texas: Society of Economic Paleontologists and Mineralogists, Gulf Coast Sec- tion. Arnold, Z.M. 1954a. Discorinopsis aguayoi (Bermudez) and Discorinopsis vadescens Cushman and Bronnimann: A Study of Variation in Cultures of Living Foraminifera. Con- tributions from the Cushman Foundation for Foraminiferal Research, 5(1)^-13. 1954b. A Note on Foraminiferan Sieveplates. Contributions from the Cushman Foundation for Foraminiferal Research, 5(2):77. Asano, K. 1938. On the Japanese Species of Elphidium and Its Allied Genera. Journal of the Geological Society of Japan, 45 (538):581-591, plate 14. Bagg, R.M., Jr. 1904. Foraminifera. In W.B. Clark, G.B. Shattuck, and W.H. Dall, The Miocene Deposits of Maryland. Maryland Geological Survey, 2:460-483, plates 131-133. Banner, F.T., and W.H. Blow 1960. Some Primary Types of Species Belonging to the Superfamily Globigerinaceae. Contributions from the Cushman Foundation for Foraminiferal Research, 11(12):1-41, plates 1-8. 1965. Two New Taxa ofthe Globorotaliinae (Globiger- inacea, Foraminifera) Assisting Determination of the Late Miocene/Middle Miocene Boundary. Nature, 207(5004): 1351-1354, figures 1-3. 1967. The Origin, Evolution and Taxonomy ofthe For- aminiferal Genus Pulleniatina Cushman, 1927. Micropaleontology, 13(2): 133-162, plates 1-4. Berggren, W.A. 1972. Cenozoic Biostratigraphy and Paleobiogeography of the North Atlantic. In A.S. Laughton et al., Initial Reports of the Deep Sea Drilling Project, 12:965-1001, 13 plates. Berggren, W.A., and M. Amdurer 1973. Late Paleogene (Oligocene) and Neogene Plank- tonic Foraminiferal Biostratigraphy of the Atlan- tic Ocean (Lat. 30°N to Lat. 30°S). RivisIa Italiana di Paleontologia, 79(3):337-392, plates 25-33. Berggren, W.A., and J.A. Van Couvering 1974. The Late Neogene. Palaeogeography, Palaeoclimatol- ogy, Palaeoecology, 16(1 /2): 1 -216. Blackwelder, B.W., and L.W. Ward 1976. Stratigraphy of the Chesapeake Group of Maryland and Virginia. (Guidebook 7b: Northeast-Southeast Sec- tions Joint Meeting 1976.) 55 pages. Arlington, Virginia: Geological Society of America. Blow, W.H. 1956. Origin and Evolution ofthe Foraminiferal Genus Orbulina d'Orbigny. Micropaleontology, 2(l):57-70, figures 1-4. 1959. Age, Correlation, and Biostratigraphy of the Up- per Tocuyo (San Lorenzo) and Pozon Formations, Eastern Falcon, Venezuela. Bulletins of American Paleontology, 39(178):59-252, plates 6-19. 1969. Late Middle Eocene to Recent Planktonic Fora- 406 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY miniferal Biostratigraphy. Proceedings of the First International Conference on Planktonic Microfossils, Ge- neva, 1967, pages 199-422, plates 1-54. Blow, W.H., and F.T. Banner 1966. The Morphology, Taxonomy and Biostratigraphy of Globorotalia barisanensis LeRoy, Globorotalia fohsi Cushman and Ellisor, and Related Taxa. Micro- paleontology, 12 (3): 286-302, plates 1-2. Bolli, H.M. 1957. Planktonic Foraminifera from the Oligocene-Mio- cene Cipero and Lengua Formations of Trinidad, B.W.I. United Stales National Museum Bulletin, 215:97-123, plates 22-29. 1970. The Foraminifera of Sites 23-31, Leg 4. In R.G. Bader et al., Initial Reports of the Deep Sea Drilling Project, 4:577-643, plates 1-9. Boltovskoy, E. 1974. Neogene Planktonic Foraminifera of the Indian Ocean (DSDP, Leg 26). In T.A. Davies, B.P. Luyendyk, et al., Initial Reports of the Deep Sea Dnlling Project, 26:675-741, plates 1-14. Bronnimann, P. 1950. Occurrence and Ontogeny of Globigerinatella insueta Cushman and Stainforth from the Oligocene of Trinidad, B.W.I. Contributions from the Cushman Foundation for Foraminiferal Research, 1:80-82, plates 13-14. Bronnimann, P., and J. Resig 1971. A Neogene Globigerinacean Biochronologic Time- scale ofthe Southwestern Pacific. In E.L. Winterer et al.. Initial Reports of the Deep Sea Drilling Project, 7 (2): 1235-1469, plates 1-51. Buza^, M.A. 1965. The Distribution and Abundance of Foraminifera in Long Island Sound. Smithsonian Miscellaneous Collections, 149:1-89, plates 1-4. 1966. The Discrimination of Morphological Groups of Elphidium (Foraminifera) in Long Island Sound through Canonical Analysis and Invariant Char- acters. Journal of Paleontology, 40:585-594, plates 71-72. Cita, M.B., and S. Gartner 1973. IV: The Stratotype Zanclean Foraminiferal and Nannofossil Biostratigraphy. Rivista Italiana di Pa- leontologia, 79(4):503-558, plates 50-53. Cole, W.S. 1931. The Pliocene and Pleistocene Foraminifera of Florida. Florida State Geological Survey Bulletin, 6:1-58, plates 1-7. Cooke, C.W., J. Gardner, and W.P. Woodring 1943. Correlation of the Cenozoic Formations of the Atlantic and Gulf Coastal Plain and the Carib- bean Region. Geological Society of America Bulletin, 54:1713-1724. Copeland, C.W. 1964. Eocene and Miocene Foraminifera from Two Lo- calities in Duplin County, North Carolina. Bulle- tins of American Paleontology, 47 (215): 205-324, plates 23-43. Cushman, J.A. 1918. Some Pliocene and Miocene Foraminifera ofthe Coastal Plain of the United States. United States Geological Survey Bulletin, 676:1-100, plates 1-31. 1919. Fossil Foraminifera from the West Indies. In T.W. Vaughan, Contributions to the Geology and Pa- leontology of the West Indies. Carnegie Institute of Washington Publication, 291:23-71. 1922a. Shallow-Water Foraminifera of the Tortugas Re- gion. Carnegie Institute of Washington Publication, 311:1-85, plates 1-14. 1922b. The Foraminifera of the Byram Calcareous Marl at Byron, Mississippi. United States Geological Survey Professional Paper, 129-E:87-122, plates 14-28. 1925. Some Textulariidae from the Miocene of Califor- nia. Contributions from the Cushman Laboratory for For- aminiferal Research, 1(2):29-35, plate 5. 1926. Foraminifera of the Genera Siphogenerina and Pa- vonina. Proceedings of the United States National Mu- seum, 67(25): 1-24, plates 1-6. 1929. The Foraminifera of the Atlantic Ocean, Part 6: Miliolidae, Ophthalmidiidae, and Fischerinidae. United States National Museum Bulletin, 104:1-29, plates 1-22. 1930. The Foraminifera of the Choctawhatchee Forma- tion of Florida. Florida State Geological Survey Bulle- tin, ^-.1-63, plates 1-12. 1937. A Monograph of the Subfamily Virgulininae of the Foraminiferal Family Buliminidae. Cushman Laboratory for Foraminiferal Research Special Publica- tion, 9:1-228, plates 1-24. 1939. A Monograph of the Foraminiferal Family Non- ionidae. United States Geological Survey Professional Paper, 191:1-100, plates 1-20. 1948. Foraminifera, Hammond Well. Maryland Depart- ment of Geology, Mines, and Water Resources Bulletin, 2:213-267, plates 15-26. Cushman, J.A., and E.D. Cahill 1933. Miocene Foraminifera of the Coastal Plain of the Eastern United States. United Slates Geological Survey Professional Paper, 175-A:l-35, plates 1-13. Cushman, J.A., and P.G. Edwards 1937. Astrononion, a New Genus of the Foraminifera, and Its Species. Contributions from the Cushman Laboratory for Foraminiferal Research, 13(l):29-36, plate 3. Cushman, J.A., and P.W. Jarvis 1936. Three New Foraminifera from the Miocene, Bow- den Marl, of Jamaica. Contributions from the Cushman Laboratory for Foraminiferal Research, 12(l):3-5, plate 1. Cushman, J.A., and W. McGlamery 1938. Oligocene Foraminifera from Choctaw Bluff, Al- NUMBER 53 407 abama. United States Geological Survey Professional Paper, 189-D: 103-119, plates 24-28. Cushman, J.A., and G.M. Ponton 1931. A New Virgulina from the Miocene of Florida. Contributions from the Cushman Laboratory for Forami- niferal Research, 7(2):32-33, plate 4. 1932. The Foraminifera of the Upper, Middle and Part of the Lower Miocene of Florida. Florida Geological Survey Bulletin, 9:1-147, plates 1-17. Cushman, J.A., and R.M. Stainforth 1945. The Foraminifera of the Cipero Marl Formation of Trinidad, British West Indies. Cushman Labora- tory for Foraminiferal Research Special Publication, 14:1-74, plates 1-16. Deshayes, G.P. 1832. Encyclopedic methodique: Histoire naturelle des Vers, 2:1-594, Paris: Mme. V. Agasse. d'Orbigny, A.D. 1826. Tableau Methodique de la Classe des Cephalo- podes. Annates des Science Naturelles Paris, 7:245-314, plates 10-17. 1839a. Foraminiferes. In R. de la Sagra, Histoire physique, poltitique et naturelle de I'lle de Cuba, 2:1-224, plates 1-12. Paris: Bertrand. 1839b. Foraminiferes des lies Canaries. In M.P. Barker- Webb and S. Berthelot, Histoire naturelle des lies Canaries, 2(2): 119-152, plates 1-3. Dorsey, A. 1948. Miocene Foraminifera, Chesapeake Group. Mary- land Department of Geology, Mines, and Water Resources Bulletin, 2:268-321, plates 27-39, figures 28-30. Douglass, R., and W.V. Sliter 1965. Taxonomic Revision of Certain Discorbacea and Orbitoidacea (Foraminiferida). Tulane Studies in Geology, 3(3): 149-164, plates 1-3. DuBar, JR. 1958. Stratigraphy and Paleontology of the Late Neo- gene Strata of the Caloosahatchee River Area of Southern Florida. Florida Geological Survey Bulletin, 40:1-267. DuBar, J.R., H.S. Johnson, Jr., Bruce Thom, and W.O. Hatchell 1974. Neogene Stratigraphy and Morphology, South Flank of the Cape Fear Arch, North and South Carolina. In R.Q. Oaks, Jr., and J.R. DuBar, editors, Post-Miocene Stratigraphy, Central and Southern Atlantic Coastal Plain, pages 139-173. Logan: Utah State University Press. DuBar, J.R., and J.R. Solliday 1963. Stratigraphy of the Neogene Deposits, Lower Neuse Estuary, North Carolina. Southeastern Geol- ogy, A(^)-.212,-233. Feyling-Hanssen, R.W. 1972. The Foraminifer Elphidium excavatum (Terquem) and Its Variant Forms. Micropaleontology, 18(3): 337-354, plates 1-6. Feyling-Hanssen, R.W., and M.A. Buzas 1976. Emendation of Cassidulina and Islandiella Helenae New Species. Journal of Foraminiferal Research, 6(2): 154-158. Fleisher, R.L. 1974. Cenozoic Planktonic Foraminifera and Biostratig- raphy, Arabian Sea Deep Sea Drilling Project, Leg 23A. In R.B. Whitmarsh et al.. Initial Reports ofthe Deep Sea Drilling Project, 23:1001-1072, plates 1-21. Fornasini, C. 1888. Tavola Paleo-protistografica. Bollettino della Societa Geologica Italiana, 7:44-48, plate 3. Gernant, R.E. 1970. Paleoecology of the Choptank Formation (Mio- cene) of Maryland and Virginia. Maryland Geolog- ical Survey Report of Investigations, 12:1-90. Gibson, T.G. 1962. Benthonic Foraminifera and Paleoecology of the Miocene Deposits of the Middle Atlantic Coastal Plain. 198 pages. Ph.D. dissertation, Princeton University. 1967. Stratigraphy and Paleoenvironment of the Phos- phatic Miocene Strata of North Carolina. Geolog- ical Society of America Bulletin, 78(5) :631-650. 1971. Miocene of the Middle Atlantic Coastal Plain. Maryland Geological Survey Guidebook, 3:1-15. In prep. Miocene and Pliocene Pectinidae (Bivalvia) from the Lee Creek Mine, North Carolina, and Adjacent Areas. Gradstein, F.M. 1974. Mediterranean Pliocene Globorotalia: A Biometrical Approach. Utrecht Micropaleontological Bulletin, 7:1- 128, plates 1-8. Hansen, H.J. 1972a. Two Species of Foraminifera ofthe Genus Turrilina with Different Wall Structure. Lethaia, 5:39-45. 1972b. Pore Pseudopodia and Sieve Plates oi Amphistegina. Micropaleontology, 18(2) :223-230. Hazel, J.E. 1971a. Ostracode Biostratigraphy ofthe Yorktown For- mation (Upper Miocene and Lower Pliocene) of Virginia and North Carolina. United States Geolog- ical Survey Professional Paper, 704:1-13. 1971b. Paleoclimatology ofthe Yorktown Formation (Up- per Miocene and Lower Pliocene) of Virginia and North Carolina. In H.J. Oertli, editor, CoUoque sur la paleoecologie des ostracodes, Pau. Centre Recherches Pau-SNPA Bulletin, 5:361-375. 1977. Distribution of Some Biostratigraphically Diag- nostic Ostracodes in the Pliocene and Lower Pleis- tocene of Virginia and Northern North Carolina. Journal of Research, 5(3):373-388. Hornibrook, N.B. 1964. The Foraminiferal Genus Astrononion Cushman 408 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY and Edwards. Micropaleontology, 10(3):333-338, plate 1. Howe, H.V., and W.E. Wallace 1932. Foraminifera of the Jackson Eocene at Danville Landing on the Ouachita, Catahoula Parish, Louisiana. Louisiana Department of Conservation Geo- logical Bulletin, 2:1-118, plates 1-15. Jahn, B. 1953. Elektronenmikroskopische untersuchungen an Foraminiferenschalen. Zeitschrift fiir Wissenschaft- liche Mikroskopie und fur Microskopische Technik, 61(5): 294-297. Jenkins, D.G. 1960. Planktonic Foraminifera from the Lakes Entrance Oil Shaft, Victoria, Australia. Micropaleontology, 6(4):345-371, plates 1-5. 1971. New Zealand Cenozoic Planktonic Foraminifera. New Zealand Geological Survey Paleontological Bulletin, 42:1-278, plates 1-23. Jenkins, D.G., and W.N. Orr 1972. Planktonic Foraminiferal Biostratigraphy of the Eastern Equatorial Pacific-DSDP Leg 9. In J.D. Hays et al.. Initial Reports of the Deep Sea Drilling Project, 9:1059-1193, plates 1-41. Jepps, M.W. 1942. Studies on Polystomella Lamarck (Foraminifera). Journal ofthe Marine Biological Association ofthe United Kingdom, 25:607-666. Kennett, J.P. 1973. Middle and Late Cenozoic Planktonic Foramini- feral Biostratigraphy of the Southwest Pacific— DSDP Leg 21. In R.E. Burns et al., Initial Reports of the Deep Sea Drilling Project, 21:575-639, plates 1-21. Kleinpell, R.M. 1938. Miocene Stratigraphy of California. 450 pages, 22 plates. Tulsa, Oklahoma: American Association of Petroleum Geologists. Lamb, J.L., and J.H. Beard 1972. Late Neogene Planktonic Foraminifers in the Ca- ribbean, Gulf of Mexico, and Italian Stratotypes. University of Kansas Paleontological Contributions, Ar- ticle, bl-A-^l, plates 1-36. Le Calvez, J. 1938. Recherches sur les Foraminiferes, 1: Development et reproduction. Archives de Zoologie Experimentale et Generate, 80:163-333. 1947. Les perforations du test de Discorbis erecta (Fora- minifere). Bulletin du Laboratoire Maritime de Dinard, 29:1-4. 1974. Revision des Foraminiferes de la Collection d'Orbigny, I: Foraminiferes des lies Canaries. Ca- hiers Micropaleontologic, 2:1-107, plates 1-28. Loeblich, A.L., Jr., and H. Tappan 1957. The New Planktonic Foraminiferal Genus Tino- phodella, and an Emendation oi Globigerinita bron- nimann. Journal ofthe Washington Academy of Sciences, 47(4): 112-116, 3 figures. 1964. Sarcodina Chiefly "Thecamoebans" and Forami- niferida. In Raymond C. Moore, editor. Treatise on Invertebrate Paleontology, C(l-2): 900 pages. Law- rence: University of Kansas Press for Geological Society of America. Mansfield, W.C. 1943. Stratigraphy of the Miocene of Virginia and the Miocene and Pliocene of North Carolina, /n Julia Gardner, Mollusca from the Miocene and Lower Pliocene of Virginia and North Carolina, Part 1: Pelecypoda. United States Geological Survey Profes- sional Paper, 199-A: 1-19. Margerel, J.P. 1971. Le Genre Faujasina d'Orbigny dans le Plio-Pleis- tocene du Bassin Nordique Europeen. Revue de Micropaleontologic, 14(2): 113-120, plates 1-3. Marks, P. 1952. Miocene Smaller Foraminifera from the Region of Constantine Algeria. Geologic en Mijnbouw, 8: 282-291, plate 1. McLean, J.D., Jr. 1956. The Foraminifera of the Yorktown Formation in the York-James Peninsula of Virginia, with Notes on the Associated Mollusks. Bulletins of American Paleontology, 36(160):261-394, plates 35-53. Mongin, D. 1959. Study of Some American Miocene Lamellibranchs and Comparisons with Related European Species. Bulletins of American Paleontology, 39(180): 279-344, plates 24-27. Nyholm, K.-G. 1961. Morphogenesis and Biology of the Foraminifer Cibicides lobatulus. Zoologiska Bidrag Fran Uppsala, 33:157-196, plates 1-5. Parker, F.L. 1967. Late Tertiary Biostratigraphy (Planktonic Fora- minifera) of Tropical Indo-Pacific Deep-Sea Cores. Bulletins of American Paleontology, 52(235): 111-208, plates 17-32. Parker, W.K., and T.R. Jones 1865. On Some Foraminifera from the North Atlantic and Arctic Oceans, including Davis Straits and Baffin's Bay. Royal Society of London Philosophical Transactions, 155:325-441, plates 12-19. Parker, W.K., T.R. Jones, and H.B. Brady 1865. On the Nomenclature ofthe Foraminifera, Pt. 10 (continued).—The Species Enumerated by d'Orbigny in the Annales des Sciences Naturelles, vol. vii. 1826. Ill: The Species Illustrated by Models. Annals and Magazine of Natural History, 16:15-41, plates 1-3. Poag, C.W. 1966. Paynes Hammock (Lower Miocene?) Foramini- NUMBER 53 409 fera of Alabama and Mississippi. Micropaleontology, 12(4):393-440, plates 1-9. 1972a. Planktonic Foraminifers of the Chickasawhay Formation, United States Gulf Coast. Micropaleon- tology, 18(2):257-277, plate 1. 1972b. Neogene Planktonic Foraminiferal Biostratigra- phy of the Western North Atlantic: Deep Sea Drilling Project, Leg 11. In CD. Hollister et al.. Initial Reports of the Deep Sea Drilling Project, 11:483-543, plates 1-11. Pokorny, V. 1956. New Discorbidae (Foraminifera) from the Upper Eocene Brown Pouzdrany Marl, Czechoslovakia. Universitas Carolina, Geologica 2(3):257-278. Poore, R.Z., and W.A. Berggren 1975. Late Cenozoic Planktonic Foraminiferal Biostra- tigraphy and Paleoclimatology of Hatton-Rockall Basin. Journal of Foraminiferal Research, 5(4): 270- 293, plates 1-5. Postuma, J.A. 1971. Manual of Planktonic Foraminifera. 420 pages. Am- sterdam: Elsevier Publishing Company. Puri, H.S. 1953. Contribution to the Study of the Miocene of the Florida Panhandle. Florida Geological Survey Bulletin, 36:1-345, plates 1-30, 1-17. Puri, H.S., and R.O. Vernon 1964. Summary of the Geology of Florida and a Guide- book to the Classic Exposures. Florida Geological Survey Special Publication, 5:1-312. Reuss, A.E. 1850. Neues Foraminiferen aus den Schichten des oster- reichischen Tertiarbeckens. Denkschiften der Kaiser- lichen Akademie der Wissenschaften, Mathematisch-Na- turwissenschaftliche Classe, 1:365-390, plates 46-51. Rogl, F. 1974. The Evolution ofthe Globorotalia truncatulinoides and Globorotalia crassaformis Group in the Pliocene and Pleistocene of the Timor Trough, DSDP Leg 27, Site 262. In J.J. Veevers, et al.. Initial Reports of th Deep Sea Drilling Project, 27:743-767, plates 1-5. Saito, T. 1963. Miocene Planktonic Foraminifera from Honshu, Japan. Science Reports ofthe Tohoku Univensty, Sendai, second series (geology), 35(2): 123-209, plates 53-56, 15 figures, 16 tables. Schnitker, D. 1970. Upper Miocene Foraminifera from near Grimes- land, Pitt County, North Carolina. North Carolina Department of Conservation and Development Special Publication, 3:1-128, plates 1-11. Schwager, C. 1866. Fossile Foraminiferen von Kar-Nicobar. Novara Expedition, Geologic, 2:187-268, plates 4-7. Shattuck, G.B. 1904. Geological and Paleontological Relations, with a Review of Earlier Investigations. In W.B. Clark, G.B. Shattuck, and W.H. Dall, The Miocene De- posits of Maryland. Maryland Geological Survey, 2:xxxiii-cxxxvii. Stainforth, R.M., J.L. Lamb, H. Luterbacher, J.H. Beard, and R.M. Jeffords 1975. Cenozoic Planktonic Foraminiferal Zonation and Characteristics of Index Forms. University of Kansas Paleontological Contributions, 62: 425 pages, 213 fig- ures, de Stefani, T. 1952. Su alcune manifestasioni di idrocarburi in prov- incia di Palermo e descrizione di foraminiferi nuovi. Plinia, 3:9. Takayanagi, Y., and T. Saito 1962. Planktonic Foraminifera from the Nobori Forma- tion, Shikoku, Japan. Science Reports of the Tohoku University Sendai, Japan, second series (Geology), special volume 5:67-106, plates 24-28. Tjalsma, R.C. 1971. Stratigraphy and Formaminifera ofthe Neogene of the Eastern Guadalquivir Basin (Southern Spain). Utrecht Micropaleontological Bulletin, 4: 161 pages. Todd, R. 1954. Appendix: Descriptions of New Species. In R. Todd, P.E. Cloud, Jr., D. Low, and R.G. Schmidt, Probable Occurrence of Oligocene on Saipan. American Journal of Science, 252:673-682, plate 1. Towe, K.M., and R. Cifelli 1967. Wall Ultrastructure in the Calcareous Foramini- fera: Crystallographic Aspects and a Model for Calcification. yoMrna/ of Paleontology, 41:742-762. Ujiie, H., and K. Oki 1974. Uppermost Miocene-Lower Pleistocene Plank- tonic Foraminifera from the Shimajiri Group of Miyako-jima, Ryukyu Islands. Memoirs of the Na- tional Science Museum, 7:31-52, plates 1-6. Vokes, E.H. 1967. Cenozoic Muricidae ofthe Western Atlantic Re- gion, Pt. 3: Chicoreus (Phyllonotus). Tulane Studies in Geology, 5(3): 133-166. Whitmore, F.C, Jr. 1965. Significant Finds of Vertebrate Fossils in Virginia and Florida. United States Geological Survey Profes- sional Paper, 525A:A71. Zachariasse, W.J. 1975. Planktonic Foraminiferal Biostratigraphy of the Late Neogene of Crete (Greece). Utrecht Micropa- leontological Bulletin, 11:1-171, plates 1-17. 410 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 1 I, 2. Globigerinoides altiaperturus Bolli, Calvert Formation, Plum Point, Maryland, USNM 240082, USGS 25975: 1, spiral view; 2, apertural view. Both X 93. 3, 6. Praeorbulina glomerosa glomerosa (Blow), Calvert Formation, Plum Point, Maryland, USNM 240083, USGS 25974: 3, spiral view; 6, side view. Both X 93. 4, 5. Globigerinoides sicanus de Stefani, Calvert Formation, Parker Creek, Maryland, USNM 240084, USGS 25972: 4, umbilical view; 5, spiral view. Both X 148. 7-9. Globigerina praebulloides pseudociperoensis Blow, Calvert Formation, Plum Point, Maryland, USNM 240086, USGS 25975: 7, umbilical view; 8, edge view; 9, spiral view. All X 93. 10. Globigerina apertura Cushman, Yorktown Formation, Suffolk, Virginia, USNM 240085, USGS 25941: umbilical view, X 150. 11-13. Globigerinita glutinata ambitacrena (Loeblich and Tappan), Yorktown Formation, near Murfreesboro, North Carolina, USNM 240087, USGS 25948: 11, spiral view; 12, umbilical view; 13, side view. All X 93. NUMBER 53 411 412 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 2 1-3. ''''Turborotalia'' acostaensis humerosa (Takayanagi and Saito), Yorktown Formation, near Murfreesboro, North Carolina, USNM 240088, USGS 25948: 1, spiral view; 2, edge view; 3, umbilical view. All X 93. 4, 7, 8. Globoquadrina altispira altispira (Cushman and Jarvis), Yorktown Formation, near Mur- freesboro, North Carolina, USNM 240089, USGS 25948: 4, spiral view; 7, edge view; 8, umbilical view. All X 93. 5, 6, 9. Pulleniatina obliquiloculata obliquiloculata (Parker and Jones), Yorktown Formation, Black Rock Landing, Bertie County, North Carolina, USNM 240090, USGS 25929: 5, umbilical view; 6, edge view; 9, spiral view. All X 93. 10-12. Sphaeroidinellopsis seminulina seminulina (Schwager), Yorktown Formation, near Murfrees- boro, North Carolina, USNM 240091, USGS 25948: 10, umbilical view; 11, edge view; 12, spiral view. All X 93. 13-15. Sphaeroidinellopsis subdehiscens subdehiscens (Blow), Yorktown Formation, near Murfrees- boro, North Carolina, USNM 240092, USGS 25948: 13, umbilical view; 14, edge view; 15, spiral view. All X 93. NUMBER 53 413 414 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 3 1-3. Globorotalia menardii (Parker, Jones, and Brady), Yorktown Formation, Suffolk, Vir- ginia, USNM 240093, USGS 25941: 1, spiral view; 2, edge view; 3, umbilical view. All X 93. r, 8, 12. Globorotalia species cf. G. crassula Cushman and Stewart, Yorktown Formation, near Murfreesboro, North Carolina, USNM 240094, USGS 25948: 4, spiral view; 8, edge view; 12, umbilical view. All X 93. 5-7. Globorotalia hirsuta hirsuta (d'Orbigny), Yorktown Formation, Yorktown, Virginia, USNM 240095, USGS 25934: 5, spiral view; 6, edge view; 7, umbilical view. All X 93. 9-11. Globorotalia puncticulata (Deshayes), Yorktown Formation, near Murfreesboro, North Carolina, USNM 240096, USGS 25948: 9, spiral view; 10, edge view; 11, umbilical view. All X 93. 13-15. ''''Turborotalia'''' inflata (d'Orbigny), Waccamaw Formation, Walker's Bluff, North Car- olina, USNM 240097, USGS 25925: 13, spiral view; 14, edge view; 15, umbilical view. All X 93. NUMBER 53 415 416 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 4 1-6. "'Turborotalia" birnageae (Blow), Pungo River Formation, Lee Creek Mine, North Caro- lina: 1, USNM 240098, USGS 26013, umbilical view, X 440; 2, USNM 240099, USGS 26013, umbilical view, X 400; 3, USNM 240100, USGS 26012; umbilical view, X 300; 4, USNM 240101, USGS 26013, umbilical view, X 300; 5, USNM 240102, USGS 26012, edge view, X 450; 6, USNM 240102, USGS 26012, umbilical view, X 410. 7, 8. Globigerinoides altiaperturus Bolli, Pungo River Formation, Lee Creek Mine, North Caro- lina: 7, USNM 240103, USGS 26012, umbilical view, X 290; 8, USNM 240104, USGS 26013, umbilical view, X 290. 9-11. Globigerina woodi woodi Jenkins, Pungo River Formation, Lee Creek Mine, North Carolina, USGS 26013: 9, USNM 240105, umbilical view; 10, USNM 240105, edge view; 11, USNM 240106, umbilical view. All X 300. 12. Globigerinoides trilobus trilobus (Reuss), Pungo River Formation, Lee Creek Mine, North Carolina, USNM 240107, USGS 26012, umbilical view, X 240. 13, 14. Globigerina species cf G anguliofficinalis Blow, Pungo River Formation, Lee Creek Mine, North Carolina: 13, USNM 240108, USGS 26012, umbilical view, X 300; 14, USNM 240109, USGS 26013, umbilical view, X 300. 15. Globigerina euapertura Jenkins, Pungo River Formation, Lee Creek Mine, North Carolina, USNM 240110, USGS 26012, umbilical view, X 300. 16. Cassigerinella chipolensis (Cushman and Ponton), Pungo River Formation, Lee Creek Mine, North Carolina, USNM 240111, USGS 26013, side view, X 400. 15. NUMBER 53 417 418 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 5 1, 2. Globorotalia scitula praescitula Blow, Pungo River Formation, Lee Creek Mine, North Carolina, USNM 240112, USGS 26012: 1, side view, X 420; 2, umbilical view, X 410. 3, 4. Globorotalia scitula praescitula Blow, Pungo River Formation, Lee Creek Mine, North Carolina, USNM 240113, USGS 26012: 3, umbilical view, X 330; 4, side view, X 360. 5, 6. Globorotalia peripheroronda Blow and Banner, Pungo River Formation, Lee Creek Mine, North Carolina, USNM 240114, USGS 26012: 5, umbilical view, X 410; 6, side view, X 450. 7. Sphaeroidinellopsis seminulina seminulina (Schwager), Pungo River Formation, Gatesville, North Carolina well core, USNM 240115, USGS 25992, umibilical view, X 170. 8. Orbulina universa d'Orbigny, Pungo River Formation, Gatesville, North Carolina well core, USNM 240116, USGS 25992, X 77. 9-10. Globorotalia hirsuta hirsuta (d'Orbigny), Croatan Formation, Lee Creek Mine, North Carolina, USNM 240117, USGS 25995: 9, umbilical view; 10, side view. Both X 300. 11, 12. Globorotalia puncticulata (Deshayes), Yorktown Formation, Lee Creek Mine, North Caro- lina, USNM 240118, USGS 26009: 11, umbilical view; 12, side view. Both X 270. 13-15. Globorotalia species cf G. truncatulinoides truncatulinoides (d'Orbigny), Croatan Formation, Lee Creek Mine, North Carolina, USNM 240119, USGS 25995: 13, umbilical view, X 300; 14, side view, X 300; 15, close-up of periphery, X 1375. NUMBER 53 419 420 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 6 1-13. Globorotalia merotumida Blow and Banner, "Virginia St. Marys" beds, Gatesville Well, North Carolina, USGS 25992: 1, USNM 240120, spiral view, X 260; 2, USNM 240120, side view, X 250; 3, USNM 240121, side view, X 250; 4, USNM 240121, umbilical view, X 250; 5, USNM 240122, spiral view, X 190; 6, USNM 240122, close-up of spiral side, X 490; 7, USNM 240122, side view, X 240; 8, USNM 240123, spiral view, X 250; 9, USNM 240124, umbilical view, X 190; 10, USNM 240124, side view, X 200; 11, USNM 240125, spiral view, X 170; 12, USNM 240125, side view, X 170; 13, USNM 240126, spiral view, X 170. 14, 15. Globorotalia minima Akers, "Virginia St. Marys" beds, Gatesville, North Carolina, well core, USNM 240127, USGS 25992: 14, umbilical view, X 340; 15, side view; X 340. 16. Globorotalia minima Akers, "Virginia St. Marys" beds, Gatesville, North Carolina, well core, USNM 240128, USGS 25992, umbilical view, X 340. 17. Globigerinatella insueta Cushman and Stainforth, Pungo River Formation, Norfolk, Virginia, Moores Bridge Well, USNM 240129, USGS 26002, umbilical view, X 220. 16. NUMBER 53 421 422 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 7 Elphidium neocrespinae, new species, Croatan Formation, Lee Creek Mine, North Carolina, USGS 25997 1, 2. Paratype, USNM 240130: 1, side view; 2, apertural view. Both X 190. 3, 4. Paratype, USNM 240131: 3, side view; 4, apertural view. Both X 160. 5, 6. Paratype, USNM 240132: 5, side view; 6, apertural view. Both X 150. 7-11. Holotype, USNM 240133: 7, apertural view, X 150; 8, side view, X 140; 9, close-up of apertural face, X 1200; 10, close-up of canal in penultimate chamber, X 6000; 11, close- up of penultimate chamber, X 1000. Micrographs reduced to 65y2% for publication. NUMBER 53 423 424 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 8 Bolivina pungoensis, new species, Pungo River Formation, Lee Creek mine. North Carolina 1-5. Holotype, USNM 240134, USGS 26013: 1, apertural view, X 435; 2, side view, X 150; 3, upper right part of specimen, X 600; 4, lower center part of specimen, X 600; 5, initial part of specimen, X 750. 6-10. Paratype, USNM 240135, USGS 26014: 6, side view, X 200; 7, close-up of apertural face, X 6000; 8, imperforate area of third last chamber, X 6000; 9, imperforate area in fourth chamber of test, X 6000; 10, apertural view, X 410. 11. Paratype, USNM 240136, USGS 26013, side view, X 160. Micrographs reduced to 66V2% for publication. 426 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 9 1-4. Bolivina calvertensis Dorsey, Pungo River Formation, Lee Creek Mine, North Carolina: 1, USNM 240137, USGS 26014, side view, X 140; 2, USNM 240137, USGS 26014, apertural view, X 500; 3, USNM 240138, USGS 26013, side view, X 220; 4, USNM 240139, USGS 26014, side view, X 200. 5, 9. Spiroplectammina mississippiensis (Cushman), "Virginia St. Marys" beds, Gatesville Well, North Carolina, USNM 240140, USGS 25992: 5, apertural view, X 75; 9, side view, X 50. 6. Siphogenenna lamellata Cushman, Pungo River Formation, USNM 240141, USGS 26013, side view, X 220. 7, 8. Siphogenerina sp., Pungo River Formation, Lee Creek Mine, North Carolina, USNM 240142, USGS 26014: 7, side view, X 110; 8, apertural view, X 245. 10-16. Siphogenerina lamellata Cushman, Pungo River Formation, USGS 26003: 10, USNM 240143, side view, X 110; 11, USNM 240144, side view, X 100; 12, USNM 240145, side view, X 115; 13, USNM 240146, side view, X 75; 14, USNM 240147, apertural view, X 130; 15, USNM 240145, apical view, X 225; 16, USNM 240146, apical view, X 310. Micrographs reduced to 66% for publication 428 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 10 1, 2. Spiroplectammina exilis Dorsey, Choptank Formation, Flag Pond, Maryland, USNM 252578, USGS 25985: 1, side view, X 55; 2, apertural view, X 115. 3. Virgulinella miocenica (Cushman and Ponton), Pungo River Formation, Lee Creek Mine, North Carolina, USNM 252579, USGS