Ecology and Systematics of Foraminifera in Two Thalassia Habitats, Jamaica, West Indies MARTIN A. BUZAS, ROBERTA K. SMITH, and KENNETH A. BEEM SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY • NUMBER 31 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 Smithsonian Contributions 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 cf 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 31 Ecology and Systematics of Foraminifera in Two Thalassia Habitats, Jamaica, West Indies Martin A. Buzas, Roberta K. Smith, and Kenneth A. Beem SMITHSONIAN INSTITUTION PRESS City of Washington 1977 ABSTRACT Buzas, Martin A., Roberta K. Smith, and Kenneth A. Beem. Ecology and Systematics of Foraminifera in Two Thalassia Habitats, Jamaica, West Indies. Smithsonian Contributions to Paleobiology, number 31, 139 pages, 38 figures, 8 plates, 34 tables, 1977.—Homogeneous Thalassia beds in back-reef flat (less than 1 m) and Discovery Bay (about 3 m) were sampled for 12 successive months in Jamaica, West Indies. Living foraminifera were enumerated in each of four monthly replicates consisting of 20 ml of sediment. At the sampling times, water temperature, sediment temperature, salinity, oxygen saturation, water pH, sedi- ment pH, sediment median, sediment sorting, turbidity, particulate organic carbon, Thalassia weight, and weight percent silt plus clay were measured. In all, 18,644 individuals belonging to 143 species were picked, sorted, and identified. The back-reef flat habitat contained 7,745 individuals belonging to 115 species, while the Discovery Bay contained 10,899 individuals belonging to 117 species. Fisher's log-series fits the distribution of species abundances at both habitats well. The number of species, information function, and equitability are usually greater at Discovery Bay for individual 20 ml samples. A general linear model consisting of parameters for station differences, overall periodicity, interaction of station differences and overall periodicity, and environ- mental variables was constructed. The densities of the 19 most abundant species were statistically analyzed individually (univariate) and simultaneously (multi- variate). Univariate analyses indicate six species have significant station differences (95% level) and seven exhibit periodicity. The environmental variables are not significant for any of the species. Multivariate analyses indicate a significant difference between stations and an overall periodicity. As in the univariate analyses, environmental variables are not significant. The results suggest that in tropical habitats changes in species densities are regulated biotically. The new species Ammonia jacksoni, Elphidium norvangi, Fissurina goreaui, Discorbinella minuta, Glabratella altispira, and G. compressa are described. Taxonomic remarks are presented for most of the species. 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 Menon, A. G. H. Ecology and systematics of foraminifera in two Thalassia habitats, Jamaica, West Indies. (Smithsonian contributions to paleobiology ; no. 31) Bibliography: p. 1. Foraminifera—Jamaica. 2. Protozoa—Classification. 3. Protozoa—Ecology. 4. Protozoa Ja- maica. 5. Thalassia. I. Smith, Roberta K., 1931—joint author. II. Beem, Kenneth A., joint author. III. Title. IV. Series: Smithsonian Institution. Smithsonian contributions to paleo- biology ; no. 31. QE701.S56 no. 31 [QL368.F6] 560'.8s [593'.12'097292] 76-608169 Contents Page Introduction ... 1 Location, Field, and Laboratory Methods . 1 Environmental Variables . . .... 3 Univariate Analyses 8 Statistical Model ... 8 Sample Size . ... 12 Statistical Analyses of Species 12 Summary of Species Analyses 37 Statistical Analyses of Genera . 41 Multivariate Analyses . . 43 Relative Abundance ... 46 Species Proportions 46 Distribution of Species Abundance 50 Species Diversity 51 Discussion .... 56 Systematic Catalog 61 Appendix: Number of Individuals Observed in 20 ml Replicate Samples 110 Literature Cited 118 Plates 123 in Ecology and Systematics of Foraminifera in Two Thalassia Habitats, Jamaica, West Indies Martin A. Buzas, Roberta K. Smith, and Kenneth A. Beem Introduction Most ecological studies of foraminifera try to relate patterns of distribution and abundance to environmental variables. The results of such studies show that foraminifera can be used to identify particular environments with relative ease. Relating patterns to environmental variables has proved much more difficult. While such studies number in the hundreds, only a handful exist where the re- searcher has sampled the same area quantitatively over the period of a year or two. Moreover, none of these are in a tropical environment. The present study examines two habitats within homogeneous beds of Thalassia in Jamaica, West Indies. One habitat is a back-reef flat barely under water at low tide. The other is at a depth of about 3 m in the serene waters of Discovery Bay. Environ- mental variability is, as one would expect, greater at the former so the habitats provide a contrast along a stress gradient. Measurement of species densities and environmental variables were made monthly over the period of a year with replication to permit extensive statistical analyses. Martin A. Buzas, Department of Paleobiology, Smithsonian Institution, Washington, D. C. 20560. Roberta K. Smith, Earth Sciences, University of California at Santa Cruz, Santa Cruz, California 95060. Kenneth A. Beem, Department of Physics and Geoscience, Montgomery College, Rockville, Maryland 20850. Specimens deposited in the National Museum of Natural History, Smithsonian Institution, are listed under the acronym "USNM" (for the collection numbers of the old United States National Museum). ACKNOWLEDGMENTS.—We thank M. Abrams, T. Chin, G. Heim, L. Keller, C. McCloy, J. Sanner, K. Smith, L. Thompson, and B. Williams for help in the laboratory. The foraminiferal samples were collected by J. B. C. Jackson, who also measured the environmental variables in the field. Dante Piacesi greatly helped to facilitate the computer analyses of the data. The foraminifera were drawn by Lawrence Isham. Finally, we thank J. B. C. Jackson and M. Silver for their helpful reviews of the manuscript. This research was supported in part by the Smithsonian Research Foundation. Location, Field, and Laboratory Methods Pear Tree Bottom (station 1) is located between Discovery Bay and Runaway Bay on the north coast of Jamaica, West Indies (Figure 1). About 80 m from shore a boulder crest is usually exposed at all tides. Shoreward lies a back-reef flat built up by accumulation of calcareous sediments and stabilized by Thalassia and Diplanthera. Station 1 is located about 20 m from mean-high water on the back-reef flat within a homogeneous patch of 77° 24' SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY iflV IJ CARIBBEAN SEA SCALE 0 1 2 KILOMETERS DISCOVERY BAY PEAR TREE BOTTOM RIVER JAMAICA 77° 24' 77 21' FICURE 1.—Location of sampling areas. Thalassia. During high tides the reef flat has an open circulation with the sea, while at low tides the reef flat is isolated and has a water depth of about 10-15 cm, and a weak counterclockwise current. Discovery Bay (station 3) is roughly circular, measuring about 1 km in the east-west direction and 1.5 km north-south (Figure 1). A fringing reef coming to within 0.5 m of the surface separates the bay from the open ocean. The center of the reef is cut by a ship channel with a width of about 150 m and depth of 12 m. Station 3 is located within a homogeneous patch of Thalassia in the eastern part of the Bay about 60 m from shore and at a depth of about 3 m. The details of the sampling procedures employed are given by Jackson (1972), who took the samples and measured the environmental variables. A brief outline is presented here. Four replicate foraminif- eral samples were taken each month by inserting core liners (internal diameter 2.54 cm) into randomly chosen cells. In all, 96 such samples were obtained over 12 successive months. Buffered for- malin was added in the field, and upon returning to the laboratory the top 20 ml were washed over a 63 \i sieve and stored in alcohol. The samples so obtained were then transported to the foraminiferal laboratory at the National Museum of Natural History for further processing. The day before examination 0.1 g of rose bengal was added to the sample. After shaking several times, the sample was washed over a 63 |j. sieve, dried and floated twice in a mixture of bromoform and acetone having a specific gravity of 2.4. The NUMBER 31 3 estimated recovery of tests using this procedure is greater than 95%. The floated portion of the sample was doled out into about 15 gridded petri dishes and rewet. All foraminifera were observed under transmitted and reflected light, and those showing stained protoplasm were picked wet and placed on slides for sorting and identification. Specimens of two abundant taxa, Discorbis rosea and one milio- lid species, were retained for systematic purposes, but could not be used in this study because their test walls made it impossible to determine whether or not they were alive at the time of collection. We do not believe this omission changes the significance of the study. In all likelihood the only difference would be a slightly increased value for species diversity. At the time of the field collections, 12 environ- mental variables were measured at each of the stations (Jackson, 1972). These were (1) bottom- water temperature, (2) sediment temperature, (3) bottom-water salinity, (4) bottom-water turbidity, (5) bottom-water particulate organic carbon, (6) bottom-water oxygen, (7) bottom-water pH, (8) sedi- ment pH, (9) median sediment size, (10) sediment sorting, (11) sediment silt plus clay weight percent, (12) dry weight Thalassia J O.lm.2 Environmental Variables A thorough description of the measurement and variation of the environmental variables is given by Jackson (1972). A brief summary is excerpted here to acquaint the reader with the salient aspects. The sediment at Pear Tree Bottom (station 1) and Discovery Bay (station 3) is fine sand. Based on 48 observations the mean and standard deviation for three sedimentary parameters are: station 3 x s 0.248 0.059 2.110 0.740 12.600 5.010 station 1 x s 0.169 0.016 1.630 0.230 10.610 5.160 median (mm) sorting percent silt plus clay Mean monthly variation of bottom water and sediment temperature with yearly means and standard deviations is shown in Figure 2. Although the overall difference in mean bottom-water tem- peratures between the two areas is only about 1°C, the shallow Pear Tree Bottom exhibits much more diurnal variation. On hot summer days the bottom- water temperature may reach 41°C, and in evening as low as 25°C. Diurnal fluctuations, thus, may be as large as the annual temperature variation. At Discovery Bay, the diurnal variation is only about 1°C. Figure 2 shows that the sediment acts as a buffer to temperature change. While the difference between sediment and bottom-water temperature is minimal at Discovery Bay, at Pear Tree Bottom the difference during the summer may reach over 2°C. On hot summer days bottom-water tempera- tures may rise as much as 2°C/hr, but sediment temperatures rise only about l°C/hr. Mean monthly salinities, yearly averages, and standard deviations are shown in Figure 3. The low salinities at station 1 from April to September are due to the discharge of the Pear Tree River eastward (upcurrent) from Pear Tree Bottom. At low tide the reef flat is isolated from the open ocean and for a period of a few hours during the summer months salinities may reach as low as 15%c. Although no direct measurements were made, the sediment probably acts as a buffer and the variation in interstitial water is probably less than for bottom water. As Figure 3 indicates, salinities at Discovery Bay are nearly constant and in the range of normal marine salinities. Figure 4 shows mean monthly turbidity values, yearly averages, and standard deviations. As might be expected turbidity values are higher at station 1 than at station 3. At Pear Tree Bottom turbid conditions often last for only a single tide; however, during storms or high rainfall, turbid conditions may prevail for a week or more. The high turbidity values recorded at station 3 are a reflection of winter storms. Visibility decreases to 0 during violent storms and sometimes remains less than 5 m for several days. During the remainder of the year, however, the water is clear with visibilities in excess of 15 m. Mean monthly values, yearly averages, and standard deviations for particulate organic carbon are shown in Figure 5. Unlike temperature and salinity, the values fluctuate greatly. Most of the particulate organic carbon at station 1 is Thalassia (derived detritus). The reason for the very high value recorded in July is not clear, but substantially accurate. Most of the microbial community on such detritus consists of bacteria, various protozoa, and diatoms (Fenchel, 1970). At station 3, the bulk (70-80%) of the particulate organic carbon consists of diatoms and bacteria. Thus particulate organic SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 30 — 29 28 2 27 (— Z 3 26 % 25 STATION 1 O WATER TEMPERATURE X=29.2 S-1.4 • SEDIMENT TEMPERATURE X=28.13 S= .89 1969 N M M O 1970 30 ""29 LU Qi 3 S 28 UJ Q_ ^27- Z 3 26 STATION 3 O WATERJEMPERATURE X= 28.2 S=l.l • SEDIMENT TEMPERATURE X= 27.99 S= 1.00 25- 1969 N M M O 1970 FIGURE 2.—Monthly variations of temperature. NUMBER 31 STATION 1 SALINITY X=31.5 S=2.5 34 33 <£ 32 fc SALIN 31 Z < 30 29 28 1969 N M M O 1970 35 34" 33 o STATION 3 o~- 3? SALINITY > X=34.9 LINI1 31 S=0.4 < to z < 30 — 29 28 1969 N F M A M J J FIGURE 3.—Monthly variations of salinity. O 1970 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY ,30 O 1970 1969 N h z OO CO |_ ,20 * U 10 2z STATION 1 TURBIDITY X=.108 S = .088 1969 N O 1970 STATION 3 TURBIDITY X = .049 $ = .044 FIGURE 4.—Monthly variations of turbidity. NUMBER 31 STATION 1 Mg c/m3 X=786 $=1012 Z 800 O 700 OQ < y 600 z < o Qi 500 o LU f— u400 a; *■ 300 z < UJ 5 200 1969 N M M O 1970 800 O 700 CO < u u 600 z < o O 500 STATION 3 Mg c/m X=316 S = 100 3 3 U 400 < 300 z < 200 1969 NDJFMAMJJA FIGURE 5.—Monthly variations of particulate organic carbon. O 1970 8 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY carbon can be thought of as a crude measure of potential food for the foraminifera. Figure 6 shows mean monthly values, yearly averages, and standard deviations for oxygen satura- tion. Both areas are supersaturated with respect to oxygen. The somewhat lower values at Discovery Bay are probably due to lower light intensities at the deeper station. Although no measurements were made at night, values presumably decreased below saturation due to the respiratory needs of the biota. Figure 7 shows mean monthly bottom water and the sediment pH values, yearly averages, and standard deviations. As might be expected pH values in the sediment are lower than in the over- lying water. Differences between bottom water and sediment pH are greater at station 1, and reducing conditions are usually present within two cm of the sediment surface at the back reef flat. Mean monthly Thalassia dry weight, yearly averages, and standard deviations are shown in Figure 8. Like particulate organic carbon, values vary greatly during the year. The high value recorded in summer at Discovery Bay may reflect a mid- to late-summer peak. Jackson (1972) calculated F ratios for the vari- ances of the environmental variables. For salinity, particulate organic matter, turbidity, and oxygen saturation, station 1 has a significantly greater variance than station 3. At Pear Tree Bottom (station 1) mean values for salinity are lower, and turbidity and oxygen saturation higher than at Discovery Bay (station 3). Jackson used stress as a collective term for mean values, variance, and the pattern of variance. Considering all three, there is less stress at Discovery Bay than at Pear Tree Bottom. Univariate Analyses STATISTICAL MODEL.—We recall that four rep- licate samples were taken each month for 12 successive months at two stations to estimate foram- iniferal species densities. At the same time 12 environmental variables were measured. An a priori general linear model was constructed to test hy- potheses for: (1) an overall difference in species densities between the two stations, (2) an overall periodicity at the two stations, (3) different period- icities at the two stations, (4) a linear relationship between species densities and environmental var- iables. The construction of such a model is discussed by Seal (1964) and Finn (1974). In matrix notation the general linear model, CI, is written x Z' B , e. m (N X 1) " (N X q) (q X 1) (N X 1) y The dependent variable, x, is a vector of species densities for the N = 96 observations. Each ele- ment, xijk, is the observation (number of individ- uals) for the ith month, i = 1, . . . , 12, the jth station, j = 1 and 3, and the kth replicate, k = 1, . . . , 4. The matrix Z' is composed of q = 22 columns consisting of 12 covariates (environmental variables) and 10 instrumental or "dummy" vari- ates whose makeup will be discussed below. The vector B has q = 22 parameters to "explain" the N observations of species densities. Finally, the vector e, assumed to be N(0, o2), is a vector of "errors" or "residuals" not accounted for by the model. Table 1 shows the composition of the matrix 21, The vector zx is a column of units and because each of the other z' adds to zero, B\ is the mean of the 96 observations. The vector z2 tests for any overall difference in species densities between the stations. Vectors z3, . . . , z14 are the observed en- vironmental variables. The vectors z15 and z16 re- peat a sine wave for each of the stations, and test for overall periodicity. Vectors z17 and z18 also test for overall periodicity but with a periodicity of LT/3 instead of n/6. Bliss (1958) discusses the use of such functions for analysis of periodicity. By multiplying z, X z15 and z16, and z2 X x17 and z18, we obtain two sets of interaction vectors that test whether or not the stations have different periodic- ities. By matrix transposition, multiplication, and in- version, the betas can be obtained from the rela- tionship B = (Z Z')"1 Z x. (2) Once the betas are obtained, the sum of squares of the residuals, I «, is solved for by (N-q)*2 =i n = ee' = x'x - B' Z Z' B. (3) Hopefully, some of the parameters in the Q, model will prove to be superfluous and can be deleted. To test whether this is so, several restricted models are postulated by equating desired values of /? to zero. These restricted models have s parame- ters and are called w models: x = Z' B + e. (4) (NX1) NXs) (sXl) (NX1). NUMBER 31 STATION 1 OXYGEN % SATURATION X=154 S= 25 180 160 O >140 O 5120 100 1969 N M M O 1970 180 160 STATION 3 OXYGEN % SATURATION X=H8 S=8 O >- X O z < 140 120 100 JL 1969 N ± F M A M j j FIGURE 6.—Monthly variations of oxygen. O 1970 10 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY O WATER PH X= 8.35 S =0.18 STATION 1 SEDIMENT PH X=7.26 S= 0.23 9.00 x Z 8.00 7.00 1969 N M M O 1970 9.00 O WATER PH X= 8.27 S= -10 STATION 3 SEDIMENT pH X=7.48 S = 0.20 < 8.00 7.00 1969 N ± F M A M J J FIGURE 7.—Monthly variations of pH. O 1970 NUMBER 31 11 «, fe ^400 O < 300- < x < 200 STATION 1 THALASSIA DRY WEIGHT X=256 S = 100 1969 N M M O 1970 400 STATION 3 THALASSIA DRY WEIGHT X = 239 S=73 x UJ <: < 300 < x < 200 1969 N M M O 1970 FIGURE 8.—Monthly variations of Thalassia dry weight. 12 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 1.—Composition of matrix Z' zl z2 z3 z4 z5 z6 z7 z8 z9 z10 zll z12 z13 z14 z15 z16 z17 z18 z19 z20 z21 z22 a vector of units +1 for station 1 observations, -1 for station 3 observations salinity (o/oo) water temperature (°C) sediment temperature (°C) particulate organic carbon (Mg C/m3) turbidity {% extinction) sediment median (mm) sediment sorting Thalassia dry weight (g/0.1 m2) weight percent silt plus clay percent oxygen saturation water pH sediment pH sin(m X ir/6) m = 1,. ...12 cos(m X TT/6) m = 1,. ..,12 sin(m X TT/3) m = 1,. ..,12 cos(m X ir/3) m = 1,. ...12 z2 x z15 z2 I z16 z2 X z17 Zo X Zio Utilizing equations (2) and (3), the sum of squares of the residual, I n, for the w models are then evaluated. The significance of any set of B's can be tested for by comparing the restricted model to the general model by means of the relationship (iq>-in)/(q-s) i«/(N-q) *<*"><*-«>■ SAMPLE SIZE.—Most of the 143 species recorded in this study are represented by only a few indi- viduals. For both statistical and biological reasons, analyses based on a few occurrences in 96 observa- tions are of dubious value. The decision of how abundant a species should be before being included in statistical analyses, however, is somewhat arbi- trary. Greig-Smith (1964) showed that plots of the Poisson distribution became very asymmetrical as the mean number of individuals became less than 4. While few natural populations are distributed according to the Poisson distribution, nevertheless, his suggestion that the number of observations with no individuals should not exceed those with one individual is valid. When the mean number of individuals is very low, no transformation can restore symmetry. In the present instance, the num- ber of observations with no individuals was exam- ined for various species densities. In general, when the mean number of individuals is greater than one or two, more than 50% of the observations con- tained at least one individual. Consequently, for the purposes of this study abundant species are defined as those having a mean number of individ- uals of at least 2 for the 96 observations. Jackson (1972) also chose a mean value of 2 in his studies of molluscs based on his observations and theoreti- cal considerations discussed by Taylor (1961). Of the 143 species recorded, 19 meet this requirement and are statistically analyzed below. To stabilize the variance and make the distributions more normal, all counts were transformed to In (xljk + 1). STATISTICAL ANALYSES OF SPECIES.—Bolivina stri- atula has a grand mean of 30.55 and is the most abundant species recorded in this study. The mean number of individuals observed each month, their yearly average, and standard deviations at the two stations is shown in Figure 9. At station 1, high mean values of 57.00 and 49.75 occur in November and May. Minima of 22.75 and 21.75 occur in March and April. At station 3, maxima of 39.75 and 45.25 occur in May and August. A minimum of 15.00 was recorded in January. Table 2 presents the results of the statistical analyses in the form of an analysis of variance table. The sum of squares column is£0) -£n for the various hypotheses, and £a for the residual. To simplify the interpretation of the significance of the various hypotheses, the last column gives the NUMBER 31 13 3 > z LL. o a: LU CO z z < 50 ~ 40 30 20 STATION 1 BOLIVINA STRIATULA 3<=33.94 S = 16.19 1969 N M M X O 1970 50 STATION 3 BOLIVINA STRIATULA ~X=27.17 S = 14.98 3 > o Z LL. o Z) Z z < 40 30 20 1969 NDJFMAMJ JA FICURE 9.—Monthly variations in density of Bolivina striatula. O 1970 TABLE 2.—Statistical analyses for Bolivina striatula Variability on account of df Sum of squares Mean square P(F) TT/3 interaction TT/6 interaction , TT/3 overall periodicity and interaction TT/6 overall periodicity and interaction , Environmental variables Station differences Residual , 4 12 1 74 0.01 0.16 0.51 3.04 4.76 0.10 18.55 0.004 0.08 0.13 0.76 0.40 0.10 0.25 0.01 0.32 0.51 3.03 1.58 0.38 0.99 0.73 0.73 0.02 0.11 0.54 TABLE 3.—Statistical analyses for Bolivina subexcavata Variability on account of df Sum of squares Mean square 1 P (F) TT/3 interaction 2 2 4 4 12 1 74 1 0 2 3 5 0 29 00 38 91 69 76 06 74 0 0 0 0 0 0 0 50 19 73 92 48 06 40 1 0 1 2 1 0 75 48 81 29 19 16 0 0 0 0 0 0 29 TT/6 interaction 62 TT/3 ar overall periodici id interaction ... ty 11 TT/6 overall periodic! and interaction ... ty 07 Envi Stat ronmental variabl sion differences , es ... 30 69 Res- TABLE 4.—Statistical analyses for Trifarina occidentalis Variability on account of df Sum of squares Mean square P(F) TT/3 interaction TT/6 interaction , TT/3 overall periodicity and interaction TT/6 overall periodicity and interaction , Environmental variables Station differences Residual 4 12 1 74 1.19 0.19 1.40 3.71 6.59 0.05 28.59 0.59 0.10 0.35 0.93 0.55 0.05 0.39 54 25 0.90 2.40 1.42 0.14 0.22 0.78 0.47 0.06 0.16 0.71 TABLE 5.—Statistical analyses for Ammonia beccarii Variability on account of df Sum of squares Mean square P(F) 7r/3 interaction TT/6 interaction TT/3 overall periodicity and interaction TT/6 overall periodicity and interaction Environmental variables Station differences Residual 2 1.36 0.68 1.11 0.33 2 1.78 0.89 1.47 0.24 4 1.61 0.40 0.66 0.62 4 5.11 1.28 2.10 0.09 12 3.47 0.29 0.48 0.93 1 0.29 0.29 0.48 0.49 74 45.00 0.61 -- -- NUMBER 31 15 probability of obtaining the F values shown. The hypothesis n/6 overall periodicity and interac- tion, having F = 3.03 and p (F) = 0.02, is the only one reaching statistical significance. We conclude that B. striatula has an overall periodicity and that it is the same at both stations. Bolivina subexcavata has a grand mean of 19.57. Figure 10 shows a maximum mean abundance for this species of 36.75 in May. A minimum mean value of 15.00 was observed in March. At station 3 the maximum mean, 31.25, was also observed in May, and a minimum, 4.25, was observed in October. Table 3 indicates the highest mean square value of B. subexcavata, having an F = 2.29 and p (F) = 0.07, is for the hypothesis n/6 overall periodic- ity and interaction. While falling short of signifi- cance at the 95% level, this relatively high mean square value reflects the monthly periodicities shown in Figure 10. Trifarina occidentalis has a grand mean of 19.18. Figure 11 shows a maximum mean value of 25.75 in May at station 1. Other monthly values at sta- tion 1 are all fairly close to the overall station mean. At station 3 a maximum of 43.50 also occurs in May. A minimum of 9.00 was recorded in January. Table 4 shows the highest mean square value for T. occidentalis is for the hypothesis IT/6 over- all periodicity and interaction. This hypothesis has a value of F = 2.40 with p(F) = 0.06. This hy- pothesis is nearly significant at the 95% level and reflects the observed monthly means plotted in Figure 11. As with the previous two species the other hypotheses have relatively low mean square values and do not approach statistical significance. Ammonia beccarii has a grand mean of 18.78. Figure 12 shows station 1 has high mean values of 33.50 and 42.25 in November and May. The low- est monthly means occur in January, February, March, July, and August. At station 3, high monthly means of 40.00, 44.00, and 48.25 occur in February, May, and June. Low mean values of 6.75, 7.75, and 7.50 occur in December, September, and October. Table 5 shows the highest mean square value is for the hypothesis II/6 overall periodicity and interaction. This hypothesis has an F = 2.10 with p(F) = 0.09. While not reaching significance at the 95% level, this hypothesis is the only one approach- ing any statistical significance, and reflects the im- portance of overall monthly periodicity at the two stations. Rosalina globularis has a grand mean of 11.01. Figure 13 shows station 1 has a maximum mean in April of 10.00. A minimum of 1.75 was observed in November. Station 3 has high means of 35.75 in May and 49.75 in June. A minimum of 4.00 oc- curs in October. Table 6 indicates that at the 95% level the hy- potheses n/3 interaction and n/6 overal periodic- ity and interaction are significant. In addition, LT/3 overall periodicity and interaction are signifi- cant at the 90% level. Curiously, although the mean values are much higher at station 3, the hy- pothesis for station differences is not statistically significant. The high mean square value for n/6 overall periodicity and interaction is probably due mainly to the overall periodicity exhibited by both stations with highs during spring and early sum- mer. The relatively high mean squares for n/3 interaction and n/3 overall periodicity and inter- action are apparently accounting for differences between the stations over the sampling period. This is the only species in the study to exhibit this phenomenon. Discorbis mira has a grand mean of 7.76. Figure 14 shows maxima mean values at station 1 of 4.00, and 3.75 occur in May and October. A minimum of 0 occurs in November. Station 3 has a maxi- mum of 29.00 individuals in May; however, in this case, the monthly mean was inflated by the occur- rence of 60 individuals in a single observation. A minimum mean value of 6.00 was observed in January. Table 7 shows a very large mean square value for the hypothesis station differences, F = 13.17 and p(F) = 0.0005. This is in keeping with the large differences in the number of individuals ob- served at the two stations discussed above and plotted in Figure 14. The hypotheses n/6 inter- action and n/6 overall periodicity and interaction have relatively high mean square values. The first, n/6 interaction, has an F = 2.68 with p (F) = 0.08, indicating the two stations differ with respect to their monthly periodicities. As Figure 14 shows, station 1 has relatively high values in January, September, and October, while these same months are relatively low for station 3. Station 1 exhibits an increase in early autumn and station 3 a de- 16 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY STATION 1 BOLIVINA SUBEXCAVATA 7=21 .96 S = 14.43 3 a > Z i z z < 30 20- 10 1969 N M M 1970 STATION 3 BOLIVINA SUBEXCAVATA 7= 17.19 S= 12.04 < D 9 > Q Z L1- O LU CD 3 Z z < 30 20 10 1969 N M M J O 1970 FIGURE 10.—Monthly variations of density of Bolivina subexcavata. NUMBER 31 17 STATION 1 TRIFARINA OCCIDENTALIS 7=15.90 S= 8.38 < 3 9 > Q Z u_ o i Z z < 30 20 10 1969 N M M -L O 1970 < 3 9 > Q Z LI_ o 3 z z < 40 30 20 10 STATION 3 TRIFARINA OCCIDENTALIS ~X = 22.46 S = 14.10 1969 NDJFMAMJ JA FIGURE 11.—Monthly variations in density of Trifarina occidentalis. -L. O 1970 18 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 40 < 3 Q > 30 z 20 3 Z Z < 10 STATION 1 AMMONIA BECCARII X=17.25 S=15.36 x O 1970 1969 N STATION 3 AMMONIA BECCAR ~X=20.31 S = 31 .40 40 < 9 30 > Q Z O 20 3 Z z < 10 1969 N J M M O 1970 FIGURE 12.—Monthly variations in density of Ammonia beccarii. NUMBER 31 19 40 STATION 1 ROSALINA GLOBULARIS "x=4.67 S = 4.27 O 1970 < 3 9 > Q Z LL. o CO 3 z z < < 3 9 > z o 3 z z < 30 20 10 JFMAMJJA FIGURE 13.—Monthly variations in density of Rosalina globularis. 20 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 5 15 9 > z STATION 1 DISCORBISMIRA X=l .44 S=2.35 O 10 :> 3 Z z < O 1970 1969 N 30 O 20 10 STATION 3 DISCORBISMIRA ~X=14.08 S=l 1 .07 1969 N M M J O 1970 FIGURE 14.—Monthly variations in density of Discorbis mira. NUMBER 31 21 TABLE 6.—Statistical analyses for Rosalina globularis Variability on account of df Sum of squares Mean square F P(F) TT/3 interaction 2 2 4 4 12 1 74 4.09 0.34 4.52 10.37 5.51 0.93 38.77 2.05 0.17 1.13 2.59 0.46 0.93 0.52 3.91 0.33 2.16 4.95 0.88 1.77 0.02 TT/6 interaction 0.72 TT/3 overall periodicity and interaction 0.08 TT/6 overall periodicity and interaction 0.001 Environmental variables ... 0.57 0.19 TABLE 7.—Statistical analyses for Discorbis mira Variability on account of df Sum of squares Mean square F P(FJ TT/3 interaction 2 2 4 4 12 1 74 1.37 2.95 2.49 4.27 7.44 7.24 40.65 0.68 1.47 0.62 1.07 0.62 7.24 0.55 1.25 2.68 1.13 1.94 1.13 13.17 0.29 TT/6 interaction 0.08 TT/3 overall periodicity and interaction 0.35 TT/6 overall periodicity and interaction 0.11 Environmental variables ... Res i dua1 0.33 0.0005 TABLE 8.—Statistical analyses for Rosalina subaraucana Variability on account of df Sum of squares Mean square Ll- P(F) TT/3 interaction 2 2 4 4 12 1 74 0.65 3.25 2.02 7.89 3.91 1.20 31.53 0.33 1.62 0.51 1.97 0.33 1.20 0.43 0.77 3.81 1.19 4.63 0.76 2.81 0.47 TT/6 interaction 0.03 TT/3 overall periodicity and interaction 0.32 TT/6 overall periodicity and interaction 0.002 Environmental variables ... Residual 0.69 0.10 crease. The hypothesis n/6 overall periodicity and interaction has an F = 1.94 with p(F) = 0.11 and reflects the overall monthly periodicity at the two stations. Rosalina subaraucana has a grand mean of 6.88. Figure 15 shows a maximum mean value of 7.75 at station 1 in October. May with a mean of 4.50 is also relatively high. A minimum mean value of 0.50 was recorded in December. At station 3, a maximum mean value of 22.50 was observed in May, a minimum of 2.75 was recorded in January. Table 8 shows the highest mean square values are for the hypotheses TJ/6 interaction and n/6 overall periodicity and interaction. The first has an F = 3.81 with p(F) = 0.03, and the second F = 4.63 with p(F) = 0.002. As Figure 15 shows, the stations do differ with respect to their periodic- ities, especially from December to January, March to April, and September to October. At the same time both show peaks in abundance in February 22 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY STATION 1 ROSALINA SUBARAUCANA X=2.77 S=2.60 <15 9 > z u_ O10 O 1970 3 z 1969 N ?5 STATION 3 ROSALINA SUBARAUCANA 7=10.98 S = 8.97 20r < 9 15 > Q z O £10 CO D z z < m c 1969 NDJFMAMJJAS FIGURE 15.—Monthly variations in density of Rosalina subaraucana. -L O 1970 NUMBER 31 23 and May. We conclude that R. subaracucana has significant monthly periodicities and that these periodicities differ at the two stations. Rosalina floridana has a grand mean of 6.71. Station 1 has high mean values of 17.25 and 18.50 in May and September (Figure 16). Low mean values of 4.75, 5.00, 4.75, and 5.25 occur from January through April. At station 3, a maximum mean value of 7.00 was recorded in September. Uniformly low values were recorded from February through April. Station 3 shows less variation in monthly means than station 1. Table 9 indicates that only the hypothesis for station differences, F = 5.14, p(F) = 0.03, is signifi- cant for R. floridana. The remainder of the mean square values is very low. Amphistegina gibbosa has a grand mean of 6.04. At station 1 little difference exists among monthly means (Figure 17). At station 3, a maximum mean of 33.75 occurs in June. In July, September, and October mean values of 18.50, 17.75, and 24.25, respectively, were recorded. A minimum mean value of 1.75 was recorded in December. In general, more individuals were observed at station 3 during the last half of the sampling period. Table 10 indicates that the hypothesis for station differences is highly significant with an F = 14.81 and p(F) = 0.0002. This is expected considering the large difference between mean values at the sta- tions cited above. The hypothesis n/6 interaction has an F = 2.48 with p (F) = 0.09, and n/6 over- all periodicity and interaction has an F = 1.95 with p(F) = 0.11. The relatively high mean squares for these hypotheses reflect the differences in mean monthly values between the stations (Figure 17). Cymbaloperetta squammosa has a grand mean of 5.05. Figure 18 shows maxima mean values of 14.75 and 17.00 in September and October for sta- tion 1. One sample in October contained 47 individuals and accounts for much of the high monthly mean. Minima of 0.025 occurred in No- vember and December. Station 3 has high mean values of 8.50 and 10.00 in May and June. A mini- mum of 0.75 occurs in October and low mean values of 1.75 occur in December, January, and March. TABLE 9.—Statistical analyses for Rosalina floridana Variability on account of df Sum of squares Mean square P(F; TT/3 interaction TT/6 interaction TT/3 overall periodicity and interaction TT/6 overall periodicity and interaction Environmental variables Station differences .... Residual 4 12 1 74 1.26 1.20 1.64 1.49 6.35 2.60 37.45 0.63 0.60 0.41 0.37 0.53 2.60 0.51 1.25 1.18 0.81 0.73 1.05 5.14 0.29 0.31 0.52 0.57 0.40 0.03 TABLE 10.—Statistical analyses for Amphistegina gibbosa Variability on account of df Sum of squares Mean square Li_ P LL. ir/3 interaction 2 2 4 4 12 1 74 0 1 0 3 3 5 29 33 95 61 05 77 81 05 0 0 0 0 0 5 0 17 97 15 76 31 81 39 0.42 2.48 0.39 1.95 0.80 14.81 0 0 0 0 0 0 66 TT/6 interaction 09 TT/3 overall periodic!* and int.prart-.inn ty 32 TT/6 ar overall periodici d interaction ... ty 11 Envi Srat ronmental variabl .ion differences . es ... 65 noo? Resi 24 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY STATION 1 ROSALINA FLORIDANA X*9.06 S-8.74 <15 9 > z O 10 :> z z < 1969 N M M O 1970 20 < 9 > z u_ O 15 10 STATION 3 ROSALINA FLORIDANA X*4.35 S = 3.74 3 z z < J. 1969 N J M M O 1970 FICURE 16.—Monthly variations in density of Rosalina floridana. NUMBER 31 25 < 3 9 > z u. 2 O STATION 1 AMPHISTEGINA GIBBOSA X-0.17 S= 0.48 3 z z i < 1969 N M M O 1970 < 3 9 > z 30- o 20L 3 z z < 10 - O 1970 1969 N STATION 3 AMPHISTEGINA GIBBOSA X~=l 1 .92 S= 13.44 FIGURE 17.—Monthly variations in density of Amphistegina gibbosa. 26 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 20 STATION 1 CYMBALOPORETTA SQUAMMOSA ~X = 6.40 5=8.13 1969 N < 3 9 > a z 10 O 2 3 z z < 15 20r < > 5 z O CtZ. 10 3 z 2 5 z < STATION 3 CYMBALOPORETTA SQUAMMOSA X=3.71 S = 4.08 1969 NDJ FMAMJ JA S FIGURE 18.—Monthly variations in density of Cymbaloporetta squammosa. O 1970 NUMBER 31 27 TABLE 11.—Statistical analyses for Cymbaloporetta squammosa Variability on account of df Sum of squares Mean square P (F) TT/3 interaction 2 2 4 4 12 1 74 6.89 1.98 7.16 4.44 6.09 0.06 51.95 3 0 1 1 0 0 0 44 99 79 11 51 06 70 4 1 2 1 0 0 91 41 55 58 72 08 0 0 0 0 0 0 01 TT/6 interaction 25 TT/3 overall periodic! and interaction ... ty D5 TT/6 overall periodici and interaction ... ty 19 Envi Sta1 ronmental variabl :ion differences es ... 73 7R RPST TABLE 12.—Statistical analyses for Cymbaloporetta tobagoensis Variability on account of df Sum of squares Mean square P(F) TT/3 interaction TT/6 interaction TT/3 overall periodicity and interaction TT/6 overall periodicity and interaction Environmental variables Station differences Residual 4 12 1 74 1.19 1.60 2.97 2.35 11.01 0.27 42.43 0.59 0.80 0.74 0.59 0.92 0.27 0.57 1.04 1.40 1.29 1.03 1.60 0.47 0.36 0.25 0.28 0.40 0.09 0.50 Table 11 shows the hypothesis n/3 interaction has a F = 4.91 with p(F) = 0.01, and n/3 overall periodicity and interaction has an F = 2.55 with p(F) = 0.05. The large value for the n/3 inter- action hypothesis is borne out by Figure 18, which shows large differences in monthly periodicities between the stations. Rosalina globularis is the only other species in this study having a significant in- teraction of the n/3 type. Cymbaloporetta tobagoensis has a grand mean of 4.27. Figure 19 shows that at station 1 maxima mean values of 5.00 and 6.25 occur in May and October; minima of 0.75 and 0.50 occur in April and July. In general, the monthly variation at sta- tion 1 is small. At station 3 a maximum mean value of 11.00 occurs in November. A mimimum value of 0.25 was recorded in January. Table 12 shows that none of the hypotheses tested are significant at the 95% level. The hy- pothesis for environment variables, however, has at F = 1.60 and p(F) = 0.09 and is significant at the 90% level. None of the other hypotheses ap- proach statistical significance. The fact that the environmental variables have a relatively high F value is somwhat puzzling because C. tobagoensis does exhibit periodicity of the n/3 or n/6 type, and station differences are not significant. Cymbaloporetta atlantica has a grand mean of 3.95. Figure 20 shows high mean values at station 1 of 6.50 and 7.50 in May and October. Minima of 0.00 occur in March, June, and September. Sta- tion 3 has a maximum mean value of 22.50 in May. Low mean values of 2.00, 2.25, and 2.50 occur in November, December, and January. February through June have the highest mean values at station 3. Table 13 indicates the hypotheses n/6 interac- tion and n/6 overall periodicity and interaction have values of F = 4.27 with p(F) = 0.02, and F = 6.06 with p (F) = 0.0003. Figure 20 shows the stations do differ with respect to their periodicity, especially in early autumn. At the same time, both have high mean values in spring, accounting for the highly significant value obtained for the hypotheses n/6 overall periodicity and interaction. The hy- pothesis station difference has an F = 5.58 and 28 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY O 1970 1969 N < 3 9 > 5 z u_ O 3 z 5 - z < 20 15 10 STATION 1 CYMBALOPORETTA TOBAGOENSIS X = 2.83 S = 2.88 20r §15 9 > Q Z 10 O 2 3 z 5- z < 1969 N STATION 3 CYMBALOPORETTA TOBAGOENSIS X=5.71 S = 4.55 DJFMAMJJAS FIGURE 19.—Monthly variations in density of Cymbaloporetta tobagoensis. O 1970 NUMBER 31 29 20 < 3 a > Q 15 STATION 1 CYMBALOPORETTA ATLANTICA Xs0.83 S=l .64 O O 1970 3 10 z 1969 N z < STATION 3 CYMBALOPORETTA ATLANTICA X"s7.06 S=8.39 20|- <15 9 > z LL. 3 z z < o 10 5 - _L -L ± 1969 NDJFMAMJJAS FIGURE 20.—Monthly variations in density of Cymbaloporetta atlantica. O 1970 30 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 13.—Statistical analyses for Cymbaloporetta atlantica Variability on account of df Sum of squares Mean square F P(F) TT/3 interaction 2 2 4 4 12 1 74 1.31 3.38 3.74 9.59 7.21 2.21 29.27 0.65 1.69 0.93 2.40 0.60 2.21 0.40 1.65 4.27 2.36 6.06 1.52 5.58 0.20 TT/6 interaction 0.02 TT/3 overall periodicity and interaction 0.06 tf/6 overall periodicity and interaction 0.0003 Environmental variables ... 0.12 0.02 p(F) = 0.02, indicating a significant statistical dif- ference between the overall means of the two stations. Asterigerina carinata has a grand mean of 3.82. Figure 21 shows that at station 1 maxima mean values of 10.75 and 14.25 occur in September and October. A minimum of 0.25 occurs in April. In general, station 1 shows an increase in the number of individuals observed in the last half of the sampling period. The monthly pattern at station 3 has maxima mean values of 6.00, 7.25, and 6.25 in November, February, and May. The minimum values of 1.50 occur in August and October. The overall pattern at station 3 is quite different from station 1. Table 14 indicates the hypotheses n/6 interac- tion and n/6 overall periodicity and interaction are highly significant. We conclude that A. carinata exhibits a monthly periodicity and that this pe- riodicity is different at the two stations (Figure 21). Bolivina doniezi has a grand mean of 3.73. This species exhibits very little variation from month to month. May, however, has the greatest number of individuals with means of 7.75 and 8.50 at sta- tions 1 and 3, respectively (Figure 22). Table 15 indicates that none of the hypotheses considered here are even close to being statistically significant for B. doniezi. Planorbulinella acervalis has a grand mean of 3.22. Figure 23 shows that at station 1 maximum mean values of 2.75 occur in May and July. A minimum of 0.25 was recorded in November. Sta- tion 1 exhibits an increase in mean monthly abun- dances during the sampling period. Station 3 has maxima mean values of 8.50 and 8.00 in November and May. Unlike station 1, no simple increase or decrease in mean abundances was observed during the sampling period. Table 16 indicates that none of the hypotheses considered are significant for P. acervalis. The mean square for station differences is the largest reflecting the differences in station means cited above. Nonionella auricula has a grand mean of 2.34. Figure 24 shows that at station 1 maxima mean values occur in November and October with values of 2.00 and 2.75. Minima mean values of 0.25 occur TABLE 14.—Statistical analyses for Asterigerina carinata Variability on account of TT/3 interaction TT/6 interaction 7r/3 overall periodicity and interaction TT/6 overall periodicity and interaction Environmental variables Station differences Residual df Sum of squares Mean square U- P(F) CM CM 1.02 6.58 0.51 3.29 0.91 5.86 0.41 0.004 4 1.63 0.41 0.73 0.57 4 12 1 74 12.03 8.69 1.10 41.53 3.01 0.72 1.10 0.56 5.36 1.29 1.95 0.0008 0.22 0.17 NUMBER 31 31 O 1970 1969 N 20 < 3 9 I- o 3 z z < 10 - STATION 1 ASTERIGERINA CARINATA X"=3.90 S = 5.29 20r- 5 15 9 > o O z 10- STATION 3 ASTERIGERINA CARINATA X"= 3.75 S= 3.42 :> 3 z z < I -L 1969 NDJ FMAMJ JAS FIGURE 21.—Monthly variations in density of Asterigerina carinata. ± O 1970 32 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY STATION 1 BOLIVINA DONIEZI 7=3.8 1 S = 3.29 Sis 9 > o z 10 5 3 z z < 1969 N M M ± O 1970 3 15 9 > o z 10 STATION 3 BOLIVINA DONIEZI x"=3.65 S =3.10 3 z z < 5 ± 1969 N M M -J O 1970 FIGURE 22.—Monthly variations of density of Bolivina doniezi. NUMBER 31 33 < 9 If > Q z STATION 1 PLANORBULINELLA ACERVALIS X= 1.35 S= 1 .56 3510 3 z z < LU 2 5 O 1970 1969 N 15 STATION 3 PLANORBULINELLA ACERVALIS "X= 5.08 S = 3.99 oio 3 z z < ± -L X J_ 1969 NDJ FMAMJJ AS FIGURE 23.—Monthly variations in density of Planorbulinella acervalis. O 1970 34 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY < 3 9 > 5 Sic u_ O STATION 1 NONIONELLA AURICULA X= 1 .33 S = 1 .37 3 z z < O 1970 1969 N < 3 9 z STATION 3 NONIONELLA AURICULLA X= 3.35 S= 2.79 O 3 z z < 1969 N M M O 1970 FIGURE 24.—Monthly variations in density of Nonionella auricula. NUMBER 31 35 TABLE 15.—Statistical analyses for Bolivina doniezi Variability on account of df Sum of squares Mean square F P(F) IT/3 interaction 2 2 4 4 12 1 74 0.01 0.12 0.23 0.69 4.04 0.67 31.50 0.01 0.06 0.06 0.17 0.34 0.67 0.42 0.01 0.15 0.14 0.41 0.79 1.58 0.99 TT/6 interaction 0.86 TT/3 overall periodicity TT/6 overall periodicity and interaction 0.97 0.80 Environmental variables ... 0.66 0.21 TABLE 16.—Statistical analyses for Planorbulinella acervalis Variability on account of df Sum of squares Mean square F P :F) TT/3 interaction 2 2 4 4 12 1 74 1 0 2. 0 6. 1 35. 48 76 15 78 10 16 16 0 0 0 0 0 1 0 74 38 54 19 51 16 46 1 0 1 0 1 2 56 79 13 41 05 43 0 0 0 0 0 0 ?? TT/6 interaction 46 TT/3 overall periodici and interaction ... ty ?5 TT/6 overall periodici and interaction ty 80 Envi Sta1 ronmental variabl :ion differences . es ... 40 12 Res- TABLE 17.—Statistical analyses for Nonionella auricula Variability on account of df Sum of squares Mean square F P(F) TT/3 interaction 2 2 4 4 12 1 74 0.79 0.17 0.89 1.69 5.35 1.37 31.75 0.39 0.08 0.22 0.42 0.45 1.37 0.43 0.92 0.20 0.52 0.99 1.04 3.20 0.40 TT/6 interaction 0.82 TT/3 overall periodicity 0.72 TT/6 overall periodicity and interaction 0.42 Environmental variables ... 0.41 0.08 in February and June. At station 3, high means of 6.00 and 5.25 occur in August and October. A minimum of 2.00 occurs in December. Table 17 indicates that all the mean square values, except for station differences, are relatively low. The hypothesis, station differences, has F = 3.20 and p(F) = 0.08, indicating a significant differ- ence between station means at the 90% level. Cyclogyra planorbis has a grand mean of 2.33. Figure 25 shows that a high mean value of 22.50 occurs at station 1 in March, while a high of 3.25 occurs at station 3 in February. All other monthly values are relatively low. At the 95% level of significance none of the hypotheses considered are significant (Table 18). At the 90% level, however, n/6 overall periodicity and interaction is significant, reflecting the high values (especially at station 1) discussed above. Interestingly, although the overall means at the two stations differ considerably, the statistical anal- 36 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 1970 STATION 1 CYCLOGYRA PLANORBIS X~=3.85 S = 7.56 20- < 3 Q >15 D Z o 10 3 z z < 1969 N STATION 3 CYCLOGYRA PLANORBIS X"= 0 .8 1 S=l .48 < 3 9 > Q 2io LL. o 2 3 z z < 5- 1969 NDJFMAMJJA FIGURE 25.—Monthly variations in density of Cyclogyra planorbis. O 1970 NUMBER 31 37 TABLE 18.—Statistical analyses for Cyclogyra planorbis Variability on account of df Sum of squares Mean square F P(F) TT/3 interaction 2 2 4 4 12 1 74 0.97 2.07 1.12 4.36 7.33 0.86 39.89 0.48 1.03 0.28 1.09 0.61 0.86 0.54 0.90 1.92 0.52 2.02 1.13 1.59 0.41 TT/6 interaction 0.15 TT/3 overall periodicity and interaction 0.72 TT/6 overall periodicity and interaction 0.10 Environmental variables ... Station differences Residual 0.34 0.21 TABLE 19.—Statistical analyses for Discorbis murrayi Variability on account of TT/3 interaction TT/6 interaction TT/3 overall periodicity and interaction TT/6 overall periodicity and interaction Environmental variables Station differences Residual Sum of Mean df squares square LL. P(F) 2 0.94 0.47 1.05 0.36 2 0.02 0.01 0.02 0.98 4 0.98 0.24 0.55 0.70 4 0.80 0.20 0.45 0.77 12 7.17 0.60 1.33 0.20 1 1.95 1.95 4.35 0.04 74 33.16 0.45 — — yses indicate no significant difference between sta- tions. This is probably so because the higher over- all mean value recorded at station 1 is due in large degree to the relatively great number of individuals observed at that station in March. Discorbis murrayi has a grand mean of 2.08. At station 1 a maximum mean value of 5.25 was ob- served in May (Figure 26). At station 3, a maximum of 7.75 was observed in November. High mean values of 6.50 and 5.75 were observed in February and April. A low mean value of 0.00 was observed in October. During the sampling year a decrease in the abundance of D. murrayi was observed at station 3. Table 19 indicates that only the hypothesis for station differences is significant for D. murrayi, F = 4.35, p(F) = 0.04. The significant difference observed between stations is mainly due to the higher values observed at station 3 during the first 6 months of sampling time (Figure 26). Fursenkoina pontoni has a grand mean of 2.02. This species is the only one among the abundant species that does not occur at all at one of the stations. At station 3 (Figure 27) F. pontoni has a high mean of 21.50 in August. This high value is due almost entirely to a single sample in which 51 individuals were observed. Table 20 indicates that at the 95% level the hypotheses for n/3 overall periodicity and inter- action and station differences are significant. As stated above, the high observed in August is due mainly to a single observation and, consequently, the significance of the periodicity hypothesis may be spurious. The significant value for station differences, however, is clearly in line with the observations. SUMMARY OF SPECIES ANALYSES.—Table 21 shows p(F) values for the various hypotheses for each species. The species are arranged in decreasing order of abundance. Of the 19 abundant species (mean greater than 2.00), 8 have no significant hypothesis at the 95% level. Three of these are among the most abundant species, and five are among the rarer species. At the 90% level only two species, 38 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 3,J 9 > 5 z OlO STATION 1 DISCORBIS MURRAYI x"=1 .40 S= 1.84 O 1970 3 z z < 5 1969 N 2 5 15 9 > Q Z STATION 3 DISCORBIS MURRAYI X=2.77 S-3.23 O 10 z < 5 :> 1969 N DJFMAMJJAS FICURE 26.—Monthly variations in density of Discorbis murrayi. O 1970 NUMBER 31 39 < 3 9 > z O STATION 1 FURSENKOINA PONTONI* "xso.oo S=0.00 2 3 z z < 1969 N M M O 1970 20 STATION 3 FURSENKOINA PONTONI X=4.04 Ss 7.99 < 3 9 15 > a z O 10 LU CO 3 Z Z < LU 2 1969 N DJ FMAMJ J AS FIGURE 27.—Monthly variations in density of Fursenkoina pontoni. O 1970 40 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 20.—Statistical analyses for Fursenkoina pontoni Variability on account of df Sum of squares Mean square F P LL. TT/3 interaction 2 2 4 4 • 12 1 74 1 1 3 1 2 1 26 81 34 58 85 82 91 13 0 0. 0 0 0 1 0 90 67 90 46 24 91 35 2 1 2 1 0 5 56 90 53 31 67 40 0 0 0 0 0 0 08 TT/6 interaction 16 TT/3 ar overall periodici id interaction ... ty 05 Tl/6 ar overall periodici id interaction ... ty ?7 Env Sta1 ronmental variabl tion differences . es ... 78 0? Res- TABLE 21.—Summary of p(F) values for 19 abundant species Species TT/3 inter TT/6 inter TT/3 ovrin n/6 ovrin Envir var Sta diff Bolivina striatula 0.99 0.29 0.22 0.33 0.02 0.29 0.47 0.29 0.66 0.01 0.36 0.20 0.41 0.99 0.22 0.40 0.41 0.36 0.08 2 0.73 0.62 0.78 0.24 0.72 0.08 0.03 0.31 0.09 0.25 0.25 0.02 0.004 0.86 0.46 0.82 0.15 0.98 0.16 3 0.73 0.14 0.47 0.62 0.08 0.35 0.32 0.52 0.82 0.05 0.28 0.06 0.57 0.97 0.35 0.72 1.72 0.70 0.05 2 0.02 0.07 0.06 0.09 0.001 0.11 0.002 0.57 0.11 0.19 0.40 0.0003 0.0008 0.80 0.80 0.42 0.10 0.77 0.27 5 0.11 0.30 0.16 0.93 0.57 0.33 0.69 0.40 0.65 0.73 0.09 0.12 0.22 0.66 0.40 0.41 0.34 0.20 0.78 0 0.54 Ammonia beccarii 0.69 0.71 0.49 Rosalina globularis 0.19 Discorbis mira 0.0005 Rosalina subaraucana 0.10 Rosalina floridana 0.03 Amphisteqina qibbosa 0 0002 Cymbaloporetta sguammosa Cymbaloporella tobagoensis ... Cymbaloporetta atlantica Bolivina doniezi 0.78 0.50 0.02 0.70 0 21 Planorbulinella acervalis Nonionella auricula Cyclogyra planorbi s 0.12 0.08 0 21 Discorbis murrayi Fursenkoina pontoni 0.04 0 02 TOTALS 6 Bolivina doniezi and Planorbulinella acervalis, have no significant hypothesis. Both of these are rela- tively rare. Most of the species analyzed indi- vidually, then, have one or more significant hypotheses. The four most abundant species, Bolivina stria- tula, B. subexcavata, Trifarina occidentalis, and Ammonia beccarii all have the hypothesis n/6 overall periodicity and interaction significant at the 90% level. Figures 9 through 11 show that the monthly periodicities for these species are quite similar. All have high densities in May with smaller peaks in November, February, August, or Septem- ber. The remaining 15 rarer species also show simi- lar periodicity, although less uniform and more often differing between stations. Figures 12 through 27 show that for most of these species, May is the month with high densities. Often November, Feb- ruary, August, September, or October exhibit high densities. At the 95% level 7 of the 19 abundant species have one or more significant periodicity hypotheses, at the 90% level, 13. We conclude that monthly periodicity is important for most of the species analyzed. At the 95% level six species have a significant station difference hypothesis. At the same level two of these also have significant periodicity hypotheses. At the 90% level eight species have a significant NUMBER 31 41 station difference hypothesis; of these, five also have significant periodicity hypothesis. For Discorbis mira, Amphistegina gibbosa, Cymbaloporetta atlan- tica, and Fursenkoina pontoni the differences in densities between stations is substantial. The latter is the only species among the 19 abundant ones that does not occur at all at one of the stations (sta- tion 1). Table 21 shows that most of the species studied have one or more significant hypotheses. Monthly periodicity and station differences are most impor- tant for these species. At the same time, Table 21 shows that the set of 12 environmental variables is not significant for any of the species at the 95% level. STATISTICAL ANALYSES OF GENERA.—Those genera that contain a number of abundant species were also analyzed as single variates. Table 22 shows the analysis for Bolivina. The species added together and treated as a single variate for the analysis are Bolivina doniezi, B. paula, B. subexcavata, B. stria- tula, B. cf. B. compacta, B. lowmani, B. rhomboid- alis, B. cf. B. subexcavata, and B. sp. The mean number of Bolivina at station 1 is 62.06, and at sta- tion 3, 51.67. Table 22 indicates the only hypothesis with a large mean square is n/6 overall periodicity and interaction, which has an F = 2.27 and p(F) = 0.07. Thus the analysis for the genus Bolivina closely reflects the analysis for its two most abun- dant species, B. striatula and B. subexcavata (Tables 2, 3). These species have grand means of 30.55 and 19.57 respectively, while the next most abundant species, B. doniezi, has a grand mean of 3.73. The remaining six species all have means of less than 1. The genus Elphidium has seven rare species. Taken together these species, E. advenum, E. nor- vangi, E. rugulosum, E. excavatum, E. gunteri, E. kugleri, and E. sp., have a mean of 3.33 at station 1 and 3.64 at station 3. Table 23 indicates an F = 3.39 and p(F) = 0.01 for the hypothesis n/6 overall periodicity and interaction. The next largest mean square is for n/6 interaction, which has F = 2.90 and p(F) = 0.06. The genus behaves like many of the abundant species of other genera. TABLE 22.—Statistical analyses for Bolivina Variability on account of TT/3 interaction , IT/6 interaction , TT/3 overall periodicity and interaction TT/6 overall periodicity and interaction Environmental variables Station differences Residual df Sum of squares Mean square F P(F) 2 2 0.46 0.31 0.23 0.16 0.62 0.42 0.54 0.66 4 1.49 0.37 1.00 0.41 4 12 1 74 3.37 5.55 0.00 27.50 0.84 0.46 0.00 0.37 2.27 1.24 0.01 0.07 0.27 0.92 TABLE 28.—Statistical analyses for Elphidium Variability on account of TT/3 interaction IT/6 interaction TT/3 overall periodicity and interaction TT/6 overall periodicity and interaction Environmental variables Station differences Residual df Sum of squares Mean square F P(F) ro ro 1.70 2.35 0.85 1.18 2.10 2.90 0.13 0.06 4 1.97 0.49 1.22 0.31 4 12 1 74 5.49 6.51 0.03 30.03 1.37 0.54 0.03 0.41 3.39 1.34 0.07 0.01 0.22 0.79 42 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 24.—Statistical analyses for Cymbaloporetta Variability on account of df Sum of squares Mean square F P(F) TT/3 interaction 2 2 4 4 12 1 74 2.05 4.99 4.47 9.18 6.83 0.47 34.54 1.02 2.50 1.12 2.30 0.57 0.47 0.47 2.19 5.35 2.39 4.92 1.22 1.00 0.12 TT/6 interaction 0.01 IT/3 overall periodicity and interaction 0.06 TT/6 overall periodicity and interaction 0.001 Environmental variables ... 0.29 0.32 The genus Cymbaloporetta is represented by C. atlantica and C. squammosa. This genus has a mean of 7.23 at station 1 and 10.77 at station 3. Table 24 indicates high mean square values for the hypotheses n/6 interaction and n/6 overall perio- dicity and interaction. The former has an F = 5.35 and p(F) = 0.01, while the latter has F = 4.92 and p(F) = 0.001. The values for n/3 interaction and n/3 overall periodicity are also relatively large, the latter reaching significance at 90% level. This is in keeping with the individual species analysis because C. sqaummosa has significant values of the hypotheses n/3 interaction and n/3 overall perio- dicity and interaction, while C. atlantica has signifi- cant n/6 overall periodicity and interaction. The significant difference between stations noted for C. atlantica is ameliorated when the two are added together and no significant difference between sta- tions exists for the genus. The genus Discorbis is represented by D. mira, D. murrayi, and D. granulosa. The mean at station 1 is 3.34 and at station 3, 16.91. Table 25 indicates a very large mean square for station differences and correspondingly high value of F = 12.51 with p(F) = 0.0007. A relatively high mean square with F = 2.89 and p(F) = 0.03 is also present for the hypothesis n/6 overall periodicity and interaction. The hypothesis for environmental variables also has a relatively high mean square value with an F = 1.80 and p(F) = 0.06. This is a little surprising because the values for this hypothesis are relatively small for the species D. mira and D. murrayi (Tables 7, 19). The genus Rosalina is represented by R. con- cinna, R. floridana, R. globularis, R. subaraucana, R. bulbosa, R. candeiana, R. sp. a, and R. sp. b. The most abundant species are R. globularis, R. subaraucana, and R. floridiana. The mean num- ber of Rosalina at station 1 is 17.38 and at sta- tion 3, 36.03. Table 26 indicates a high F = 3.85 and p(F) =0.01 for the hypothesis n/6 overall periodicity and interaction. The hypotheses n/6 interaction and n/3 interaction also have rela- tively high values. This is in keeping with the analyses for R. globularis and R. subaraucana. The significant differences between stations observed lot TABLE 25.—Statistical analyses for Discorbis Variability on account of df Sum of squares Mean square F P F) TT/3 interaction TT/6 interaction 2 2 4 4 12 1 74 1 1 1 4 8 5 29 48 41 68 62 62 01 60 0 0 0 1 0 5 0 74 70 42 16 72 01 40 1 1 1 2 1 12 85 76 05 89 80 51 0 0 0 0 0 0 16 18 TT/3 overall periodici and interaction ... ty W TT/6 overall periodici and interaction ... ty 03 Environmental variabl Station differences . es ... 06 0007 Residual NUMBER 31 43 TABLE 26.—Statistical analyses for Rosalina Variability on account of df Sum of squares Mean square F P IF) TT/3 interaction 2 2 4 4 12 1 74 1 1 1 4 4 0 23 52 74 60 97 10 03 91 0 0 0 1 0 0 0 76 87 40 24 34 03 32 2 2 1 3 1 0 36 69 24 85 06 08 0 0 0 0 0 0 in Tf/6 interaction 07 TT/3 overall periodici and interaction ... ty 30 TT/6 overall periodici and interaction ... ty ni Envi Sta1 ronmental variabl :ion differences . es ... 41 78 Res1 R. subaraucana and R. floridana, however, do not appear when the species are added together. In general, we may conclude that the analyses of genera follow closely the analyses of their most abundant species. The rare species contribute little to the statistical analysis. Consequently, when analyzing a genus, only the more abundant species need be considered for ecological work. Similarly, the inclusion of rare species in the counts of the more abundant species due to misidentification, etc., will probably not affect the outcome. The total number of living individuals observed (standing crop) in each replicate was also analyzed as a single variate. The mean number of individuals at station 1 is 161.23, and at station 3, 227.54. A plot of the mean number of individuals per month at stations 1 and 3 is shown in Figure 28. The pattern shown in Figure 28 clearly reflects the pattern of the most abundant species (Figures 9-12). At station 1, the five most abundant species make up 58% of the total, and at station 3, 46%. Table 27 indicates that the hypothesis n/6 over- all periodicity and interaction with an F of 4.83 and p(F) = 0.002 is highly significant. This again reflects the influence of the most abundant species for whom this hypothesis has the highest mean square values. The significant station differences observed for the less abundant species are of no importance in the analysis of the total living indi- viduals. Surprisingly, the hypothesis for environ- mental variables has an F = 1.76 and p(F) = 0.07, which is significant at the 90% level. This is not in keeping with individual analysis of the significance of environmental variables shown in Table 21. Multivariate Analyses In the earlier section, we analyzed each of the 19 most abundant species by a general linear model. Now we analyze the entire ensemble of 19 species simultaneously by the same general linear model. In matrix notation, CI model is written as Cl: X = Z' B + E (6) (96 x 19) (96 x 22) (22 x 19) (96 x 19) where X is the matrix of 96 observed abundances, TABLE 27.—Statistical analyses for total live individuals Variability on account of df Sum of squares Mean square F P F) TT/3 interaction 2 2 4 4 12 1 74 0 0 0 3 3 0 12 49 67 50 27 57 00 53 0 0 0 0 0 0 0 25 33 12 82 30 00 17 1 1 0 4 1 0 45 97 73 83 76 02 0 0 0 0 0 0 24 TT/6 interaction 15 TT/3 overall periodici and interaction ... ty 57 TT/6 overall periodici and interaction ... ty on? Environmental variabl Station differences . es ... 07 89 44 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 400 z o i— 5 O 300 Q_ Li- o 2 200 z> z 100 z < STATION 1 TOTAL LIVE POPULATION X=161.23 S = 76.55 1969 N M M O 1970 400 Z g i— 300 o_ O O I 200 z z < STATION 3 TOTAL LIVE POPULATION "X= 227.50 S= 110.48 100 J. 1969 NDJ FMAMJ J ASO 1970 FICURE 28.—Monthly variations in density of total live population (excluding miliolids). NUMBER 31 45 Z' the same as in the univariate case, B a matrix of betas (the multivariate analog of the regression coefficients of the univariate model), and E a matrix of residuals assumed to be distributed as N(0,2). As in the univarite case the model is solved by matrix transposition, multiplication, and inversion. The solution for B is B = (Z Z')-1 Z X (7) and E' E = (N-q) %a = X' X - B' ZZ' B. (8) The sum of squares of the residual or the variance- covariance matrix so obtained, (N-q) %a, is anal- ogous to ee' = £a = (N-q) o-2o of the univariate model. As in the univarite case six models are hypothe- sized by deleting columns of the matrix Z'. Their respective variance-covariance matrices (N-s) %w, are estimated in the same way as for CI. A comparison of the O model with an q_s, N-, - |N-S %w\ ' Seal (1964) indicates that an approximate test for the significance of U is given by X% (q_B) ~ [N-q - 0.5 (p-q4-s+ 1)] In Up,,.^. (10) Table 28 lists the hypotheses and determinants for each of them. In the case of the i = 1 P = _^~~ i = 1 and an estimate of the variance of p is obtained from E m2 (Pi " P)2 (15) i = 1 n m2 (n — 1) (see Cochran, 1963). To illustrate the differences obtained with the two methods of calculating tfp, we use the data for November at station 1 for Bolivina striatula. In November the counts for B. striatula were 35, 51, 53, and 89, while the total number of individuals were 132, 223, 157, and 275, respectively. The standard error of p using the binomial formula is crp = 0.016 and using the formula for cluster sampling is op = 0.027. While the differences are relatively small, the confidence limits for the binomial formula are about ±3% and for the cluster formula ±5%. Thus we see the confidence placed in estimates of p is not as great as we might suppose from the binomial distribution. Table 30 gives the percent of the five most abundant species at stations 1 and 3 by months. Recall that for most species the months February, May, and August or September were times of maxima density. For the species Bolivina subex- cavata and Trifarina occidentalis, changes from month to month are not larger than we might reasonably expect from consideration of standard errors. Bolivina striatula, Ammonia beccarii, and Rosalina globularis, however, do have larger fluctuations. Figure 29 plots the percent of Bolivina striatula by month at stations 1 and 3. Notice that the per- cent pattern for B. striatula does not resemble the density pattern (Figure 9). This is also true for most of the other abundant species. As we have already shown, most species exhibit an overall periodicity that is similar, and, therefore, it would be impossible for individual species percents to resemble the actual changes in densities. For the percent pattern to resemble the actual species densities, the species must behave somewhat inde- pendently of the others. Rosalina globularis plotted in Figure 30 does so, and the percent pattern shown in Figure 30 closely resembles the density plot shown in Figure 13. As we have shown in the analyses of individual species, R. globularis and Cymbalopor- etta squammosa are unusual in having significant n/3 periodicities and having monthly patterns rather different from most of the other species. In general, we conclude that the percent pattern is not similar to the density pattern. This is so because, while the abundant species increase their densities during times of overall maxima, such as May, their species proportions are slightly decreased due to the addition of rarer species. TABLE 30.—Percent of abundant species at stations 1 and 3 by months B. striatula B. subexcavata T. occi dental is A. bee carii R. glob jlaris Month 1 3 1 3 1 3 1 3 1 3 28.97 9.86 13.09 8.16 7.50 11.24 17.03 6.89 0.64 3.71 Dec 31.17 10.45 10.58 9.40 11.50 10.75 13.17 4.03 1.30 8.06 26.91 12.05 16.14 8.84 11.21 7.23 7.40 8.84 1.12 6.83 Feb 22.58 11.45 19.03 7.16 9.52 9.30 6.45 14.31 4.36 5.55 16.82 11.10 11.09 7.11 9.98 11.97 7.21 8.10 4.44 7.98 Apr 18.47 13.03 14.65 6.92 9.34 12.10 10.83 8.91 8.49 6.52 16.95 10.19 12.52 8.01 8.77 11.15 14.40 11.28 2.38 9.17 21.48 8.18 14.80 7.55 11.19 6.77 11.91 13.61 2.53 14.03 Jul 20.12 14.45 15.54 8.70 11.16 8.34 8.76 7.76 2.79 9.17 Aug 20.20 18.01 13.86 8.26 12.67 12.74 9.11 5.08 3.76 5.57 17.96 13.96 13.44 6.29 10.21 7.11 8.92 4.24 1.81 5.88 Oct 16.42 15.43 11.44 3.08 8.64 9.44 8.03 5.44 3.28 2.90 48 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY -•5? I— z LU u 30 20 STATION 1 BOLIVINA STRIATULA % OF TOTAL LIVE POPULATION 10 1969 N M M O 1970 30 STATION 3 BOLIVINA STRIATULA % OF TOTAL LIVE POPULATION 20 10 1969 NDj FMAMJ JASO 1970 FIGURE 29.—Monthly variations of Bolivina striatula in percent of total live population. NUMBER 31 49 STATION 1 ROSALINA GLOBULARIS % OF TOTAL LIVE POPULATION 10 O 1970 z LU 1969 N u z LU 10 STATION 3 ROSALINA GLOBULARIS % OF TOTAL LIVE POPULATION JL 1969 NDJ FMAMJ JASO 1970 FIGURE 30.—Monthly variations of Rosalina globularis in percent of total live population. 50 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY DISTRIBUTION OF SPECIES ABUNDANCE.—Naturalists have long been aware that the number of indivi- duals is not evenly distributed among species. Usually a few species are abundant and the re- mainder very rare. In the present study, at both stations 1 and 3, 95% of the individuals observed belong to a third of the total species observed. In other words two-thirds of the species observed account for only 5% of the individuals. At station 1, 75% of the individuals are accounted for by 10 species, at station 3 by 14 species. R. A. Fisher (Fisher, Corbet, and Williams, 1943) reasoned that if species had different means, and if the means were distributed as chi-square, the result- ing distribution would be the negative binomial. The distribution is completely defined by the mean and k, a positive exponent. When the variance of the negative binomial approaches the mean k —+• oo, whereas with great heterogeneity k —> o. Fisher observed that on a series of Malayan butterflies the negative binomial fit the data well and the value of k was very small. Fisher, therefore, derived an expression for the expected number of species hav- ing n individuals as _eLxn, (16) n where a is a constant and x a positive number less than 1. The total number of species expected is xn S = 2 -2— = - in (1-x). (17) For a given N, total number of individuals, and S, number of species, Williams (1964, Appendix A) has tabled values of x. After obtaining an approxi- mate value of x, the expression "IT = ^ [-ln(l-x)] (18) can be used to calculate a more accurate value by successive approximations. Once x is calculated a can be calculated from the expression = N(l-x) (19) a x Because x is near unity, a (the constant in 16, 17) is a number very close to nu the expected number of species containing one individual, and has been used extensively as a measure of species diversity (Murray, 1973). At station 1, S = 115, and N = 7,745. Using (18) successive approximations yield a value of x = 0.9976, a = 18.63, and the expected number of species S = 112.38. The expected number of species, 112, is very close to the observed number 115. Similarly nu the expected number of species with one individual, is 18.59, while 23 were observed. At station 3, the total living population contains N = 10,899 individuals and S = 117 species. The calculated value of x = 0.9983, a = 18.56, and the expected number of species S = 118.36, nx = 18.53, while the observed number of species at station 3 with one individual is 27. Preston (1948, 1962) presented evidence suggesting that the distribution of individuals among species is best described by the log-normal distribution. Although Preston used log2 in his calculations and illustrations, the more conventional ln is used here. After taking the ln of the number of individ- uals observed per species, the logarithmic mean, \i, and cr2 variance were calculated. These parameters were then used to calculate the log-normal proba- bility density function by - 1 1 2o* -(lnx-u)2 (20) f(x) = x (2n a2)* The expected values of x were calculated up to a value of 121, the remainder were estimated by calculating f(x) at intervals. The error so introduced is small because as x becomes large, f(x) changes little. Tables 31 and 32 list the observations and ex- pected values for the log series and log-normal distribution. The class intervals chosen are the same as used by Williams (1964). The observations and the expected log series and log-normal curves for station 1 and 3 are plotted in Figure 31. Tables 31 and 32 and Figure 31 indicate the log series gives a better "fit" to the observations than the log normal. Both distributions underestimate the number of rare species observed but the log normal gives a poorer estimate. To evaluate "goodness of fit," a chi-square evaluation was made, grouping the last two classes in each case because of the low expected values. At station 1, the value of chi- square for the log series is 6.62 and for the log normal 18.41. At station 3, chi-square for the log series is 5.72 and for the log normal 29.29. The value of X2 (6,95) is 12.59. The chi-square evaluation is, then, in agreement with our visual inspection, and we conclude that Fisher's log series fits the data remarkably well. NUMBER 31 51 TABLE 31.—Number of species observed, predicted log series, and log-normal distribution at station 1 Individuals Predicted Predicted per species Observed log series log normal 1 23 18.59 11.81 2-4 27 20.04 20.89 5-13 18 20.04 26.42 14-40 21 19.29 23.80 41-121 14 17.17 14.97 122-364 7 12.20 6.92 365-1093 4 4.66 2.71 1094-3280 1 0.39 1.07 TABLE 32.—Number of species observed, predicted log series, and Individuals Predicted Predicted per species Observed log series log normal 1 27 18.53 10.71 2-4 19 20.01 18.32 5-13 19- 20.07 23.73 14-40 15 19.56 23.31 41-121 18 18.00 17.14 122-364 11 14.09 9.94 365-1093 7 7.00 4.81 1094-3280 1 1.10 1.54 SPECIES DIVERSITY.—The simplest measure of species diversity is S, the number of species observed. Another popular measure of species diversity is the Shannon-Wiener information function H = -2 Pi ln Pl, (21) where pi is the proportion of the ith species (MacArthur and MacArthur, 1961; Pielou, 1966; and Gibson and Buzas, 1973). The information function has the advantage of not placing much weight on the rarer species, so the chance occurrence of a few individuals representing species not pre- viously collected will not alter the function by much. For example, a species with p = 0.20 has a value of p In p of — 0.32, whereas a species with p = 0.01 has a value of — 0.05. The rarer species contribute an order of magnitude less to the func- tion than the more abundant ones. While the addition of a few rare species alters the function little, the evenness with which the species are distributed does change the value greatly. For example, five species each with a proportion of 0.20 would have an H value of 1.60, while five species with proportions of 0.90, 0.04, 0.03, 0.02, and 0.01 would have an H = 0.45. Consequently, in evaluating values of H, some measure of evenness or equitability is desirable. The maximum value of H occurs when all species are equally distributed and in this case H = ln S. Therefore, from the definition of logarithms, in this situation eH = S, and eH/S = 1. Any departure from complete even- ness will have a value of less than 1 and can be considered a measure of evenness. Buzas and Gibson (1969) suggested the ratio be used for this purpose, and Sheldon (1969) called it E. Other measures have been used (see Pielou, 1966), but for reasons given by Hill (1973), F _ e« (22) appears most appropriate. To facilitate examination of species diversity, a computer program was written that ranks the species by abundance and calculates the species proportions and cumulative proportions: N = the total number of individuals, S = the number of species, H = the information function, and E = the equitability function. Computer runs were made 52 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY A OBSERVED O LOG-SERIES • LOG-NORMAL STATION 1 13 40 121 INDIVIDUALS PER SPECIES A OBSERVED O LOG-SERIES • LOG-NORMAL STATION 3 364 13 40 121 INDIVIDUALS PER SPECIES 1093 3280 FIGURE 31.— Distribution of numbers of individuals per species. NUMBER 31 53 for each biological sample (20 ml of sediment), each month at each station, and the cumulative individ- ual samples were taken sequentially for the entire year at each station. The results for S, H, and E by month are plotted in Figure 32. The range of S at station 1 is from 45 to 73 and at station 3 from 49 to 54. Except for November and May, the number of species is always greater at station 3. The value of H at station 3 is always greater than at station 1. The mean at station 1 is 2.84 and at station 3, 3.21. Figure 32 shows that the value of H at the respective stations varies little from month to month. Except for the month of August, the value of E is always greater at station 3. The average monthly value for E at station 1 is 0.33 and at station 3, 0.40. The consistently higher value of E at station 3 insures that the value of H at station 3 remains above that of station 1. Thus, in November and May when S is greater at station 1, the value of H remains higher at station 3. E like S shows considerable variation from month to month. The value of S observed from month to month is dependent on N, the number of individuals ob- served each month. Figure 33 plots S against N (on a log scale) for each of the 48 biological samples at stations 1 and 3. Spearman's rank correlation co- efficient indicates a highly significant positive cor- relation for N and S at both stations. Note that the lines fitted to the points do not have the same slope. While we expect station 1 to have fewer species when N is less than about 800, when N is greater than 800, station 1 will have more species. The figure illustrates well Hurlbert's (1971) sug- gestion that values of S should be compared only when values of N are nearly equal. At station 3, the expected S for N = 100 is 24, and for N = 300, 38. In other words a difference of 14 species can be accounted for by the different values of N. Figure 34 plots H against N (on a log scale). At station 1 Spearman's rank correlation coefficient yields a significant positive value. Note, however, that the slope of the line is small. For N = 100, the expected H is 2.54, while for N = 300, H is 2.75. Station 3 shows no significant increase in H with N. Figure 35 plots individual sample values of E against N (on a log scale). At both stations Spear- man's rank correlation coefficient indicates a significant negative relationship. The slope of the two lines is nearly equal. Figures 33, 34, and 35 indicate that for most values of N, S is greater at station 3, but S increases at a greater rate at station 1. Consequently, because the negative slopes of E at both stations are nearly equal, Hx increases with increasing H, while H3 does not because the possible increase of S with N is nullified by the decrease in E. A common way of looking at the relationship of S with increasing N is by looking at cumulative curves. Plant ecologists have long looked at the distribution of S with increasing area. Arrhenius (1921) suggested that species and area are related by a simple power series so that when plotted on a log-log basis, the points fall in a straight line. Gleason (1922) indicated the relationship was semi- log, S plotted arithmetically and A on a log scale. Note that this is not the same as looking at individual random samples. Instead, we take the first sample, add the second to it, and so on. This is equivalent to starting at some point and measur- ing outward—clearly not a random procedure. In the present study, the data are further complicated because each set of four biological samples is taken at a different month. Nevertheless to look at the relationship of S and N over a wide range seems worthwhile. Figure 36 plots cumulative S against N (on a log scale). To simplify the presentation, only the values for every fourth (end of each month) sample are plotted. The lines, however, were calculated from the entire set of 48. As Figure 36 shows, the calcu- lated equations fit the data very well. A simple power series was also calculated for the two stations, Sx = 13.44 Nx 24 and S3 = 14.37 N3-23. These lines fit the data equally well. This phenomenon was pointed out by Preston (1962), who indicated that both curves will plot as a straight line unless the range of the abscissa is very large. Although 7,735 individuals were observed at station 1 and 10,899 at station 3, there is no indication that we have found the total number of species present in the area. The rare species are often observed only once; at station 1, 23 species are represented by one individual and at station 3, 27. Obviously, continued collecting will keep turning up rare species until the fatigue of the researcher puts an end to it. Figure 37 plots cumulative H against N. Station 1 has a slight increase with N, while station 3 is 54 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY • STATION 1 O STATION 3 .50 E .40 .30 H3.00 70 S60 50 -L -L -L M N O A M J J A S MONTH FIGURE 32.—Monthly variations in equitability, information function, and number of species. NUMBER 31 55 50 40 30 20 • STATION 1 S=-17.16+10.07lnN O STATION 3 S=-32.01+12.27lnN 20 J I I L 40 60 80 100 200 NUMBER OF INDIVIDUALS FIGURE 33.—Semilog plot of numbers of individuals vs. species 300 400 600 800 nearly horizontal. The values of H remain remark- ably stable over the entire observed range. If a single measure of species diversity must be used for data with differing values of N, H is well suited. Figure 38 plots cumulative E against N. For greater than 1000 individuals E1 is nearly horizontal. On the other hand, E3 shows a sharp negative relationship with increasing N, perhaps reaching stabilization at about 8000 individuals. Considera- tion of Figures 36, 37, and 38 shows that Sa and S3 increase continually with N; because Ex is nearly stable Hx must increase with N from a considera- tion of (22). At the same time, if H3 is stable, and S3 is increasing,* E3 must decline. An illustration of how this is accomplished will make the relationship clearer. Table 33 shows cumulative samples for the end of February and July at station 1. The table shows the value of H and E for each class of 10 species. The final value of E for each sample is 0.19. Notice that the value of H at the end of class intervals in July is always higher, most notably in the classes 0-10, 10-20, 20-30, 30-40, and 40-50. Correspondingly, at the end of any interval the value of E is higher in July. The addition of another 20 species in the July sample, however, just offsets the increase in H, making the ratio eH/S the same in both samples. At station 3, the situation is quite different. Table 34 shows that after the first class of 10 species until about 70 species, the cumulative sample for July does not increase as much in its H values as February. Consequently, at 90 species, February has a value of 3.42 and July 3.29. The addition of another 21 species in the July cumulative sample increases the value of H by only 0.02 and the ratio eH/S is much smaller in the July sample. Examination of computer print- 56 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY • STATION 1 H=1.65+.19lnN O STATION 3 H=2.71 +.05lnN 3.5 H 3.C 2.5" O o o ,° o %cP ° oo0°° • 2.0 20 ' ' I L—L 40 60 80 100 200 NUMBER OF INDIVIDUALS 300 J L 400 600 J L 800 FIGURE 34.—Semilog plot of numbers of individuals vs. information function. outs indicates that this pattern is a general trend for stations 1 and 3. The analyses presented here indicate how difficult it is to ask a simple (but imprecise) question. Is species diversity higher at station 1 or 3? No simple answer is possible. The question so stated is too vague. For values of N (about 100 to 400) occurring in biological samples of 20 ml, usually S3 > Slf H3 > Hj, and E3 > Ej (Figures 33, 34, 35). Considera- tion of cumulative samples indicates that above 800 or 1000 individuals for any given N, we expect Sx > S3, H3 > Hlf and E3 >EX (Figures 36, 37, 38). In addition we observe that the values of Ex and E3 indicate the species are not cumulating in the same way at the two stations (Figure 38, Tables 33, 34). If we ignore the warnings of Hurlbut (1969) concerning N and look at the totals for the two stations, we find Sx = 115, and S3 = 117 (for all practical purposes equal). Similarly, Fisher's alpha is at = 18.63 and a2 = 18.56. We must keep in mind, however, that N2 = 7,745 and N3 = 10,899, and for a given N above about 800, Figure 36 indi- cates Si > S3. Biologically, the individual 20 ml samples or perhaps each set of four replicates is most meaning- ful. They represent random samples of individuals living together at the same time and place. For these samples, S, H, and E are usually higher at station 3, where environmental stress is the lowest. Discussion Explaining observed patterns of densities by environmental variables is much more difficult than their recognition. In an earlier study Buzas (1969) demonstrated a highly significant statistical relation- NUMBER 31 57 • STATION 1 O STATION 3 E=10.2-.lllnN E=1.08-.10lnN .60 .50 .40" .30 20 J L J I I L 300 60 40 80 100 200 NUMBER OF INDIVIDUALS FIGURE 35.—Semilog plot of numbers of individuals vs. equitability. 400 J I L_L 800 600 ship between a set of environmental variables and species densities in a northern temperate estuary. Although the set of environmental variables was highly significant, no single variable could be isolated because the variables are highly correlated and the effect of one cannot be isolated. Lee (1974) made the same point in discussing experiments conducted by Muller (1975). Similarly, Lankford and Phleger (1973) indicated that the zoogeographic distribution of nearshore foraminifera of western North America could not be related to a single environmental variable, but only to properties of coastal-water masses. In the present study, no significant statistical relationship was found between the densities of the 19 abundant species and the set of environmental variables (Table 21). Even when the 19 species were analyzed simultaneously (multivariate population), no statistical significance was found (Table 29). Most of the abiotic variables thought to be important in controlling the density of foraminifera were measured (temperature, salinity, oxygen, tur- bidity, pH, and substrate). It is, of course, likely that some extremes of these variables were not observed, especially at station 1 (back-reef flat). Such extremes, however, would probably affect the entire fauna or most of it, and we see no evidence of such changes. Consequently, we believe abiotic variables do not regulate species densities in these Thalassia habitats. Unlike more northern areas, the physical environ- ment is sufficently benign so that biotic variables are most important. This conclusion is in agreement with Sanders (1969), Slobodkin and Sanders (1969), Dayton and Hessler (1972), and Grassle and Sanders (1973), who indicated that tropical and deep-sea environments are similar in that both have high species diversity, environmental stability predict- ability, and biotic regulation. 58 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 120 110 100 90 UJ u 80 UJ a. p. 0-1 Three specimens from station 3 and two from station 1 (represented by hypotypes USNM 211257, 211258) are so referred. They conclude with at least three series of biserial chambers occupying two-thirds to three-fourths of the test. The early portion may not be biserial, however. Tests are somewhat compressed, with rounded margins. The early portions appear unusually thick for a biserial form and the median sutures run slightly diagonally upward, suggesting a triserial or even coiled base. It has been impossible to determine the form of this portion with certainty, however. The apertures rise as wide slits from the bases nearly to the center of the final chambers and have tooth plates. The walls are fairly thin, trans- lucent, smooth, and shiny. The pores are medium sized and of fairly even distribution. The hypotype is not as twisted and is more transparent and smooth than the others and its wall can be seen clearly to be radial. A single specimen from station 1 (hypotype USNM 211259) is tentatively assigned to Bolivina. Apparently juvenile, it is very small and has seven or eight chambers. The final four chambers form two biserial pairs, with an open, slitlike aperture rising from the base a short distance up the final chamber. The growth pattern of the initial part is, however, difficult to determine. There is either a bulbous proloculus followed by two similar cham- bers in biserial arrangement and all tilted "forward" and upward slightly, or there is a very small 80 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY proloculus followed by three chambers in trocho- spiral arrangement. The wall is opaque and rather granular appearing, not smooth, with small, regularly spaced pores. The test is only slightly compressed, has rounded margins, narrow and depressed sutures, and rather inflated chambers. Hypotype: USNM 211257, 211258, 211259. standard mean deviation range station 1 0.10 0.31 0-1 station 3 0.00 0.00 0 Genus Rectobolivina Cushman, 1927 Rectobolivina raphana? (Parker and Jones) PLATE 3: FICURES 1, 2 fSiphogenerina raphanus (Parker and Jones).—Cushman, 1922a:35, pl. 5: fig. 5. Rectobolivina raphana (Parker and Jones).—Loeblich and Tappan, 1964:C553, fig. 438: 9-11. Rare live individuals occur in the present mate- rial, though empty tests also are found in sufficient numbers to suggest that they live nearby and are washed in, as with Siphogenerina costata. As to the taxonomic relationships and placement of S. costata and Rectobolivina raphana, see the remarks under the former. (The question here is not of the specific identity relating to other specimens but of the validity of the generic and specific breakdown.) Cushman's figured specimen, cited above, appears to actually belong to R. raphana, though some question exists between the two species. Figured hypotype: USNM 211214. standard mean deviation range station 1 0.02 0.14 0-1 station 3 0.08 0.28 0-1 Family BULIMINIDAE Jones, 1875 Subfamily PAVONININAE Eimer and Fickert, 1899 Genus Reussella Galloway, 1933 Reussella atlantica Cushman Verneuilina spinulosa Reuss.—Cushman, 1921:51; 1922a:28, pl. 3: fig. 11. Reussella spinulosa (Reuss) var. atlantica Cushman, 1947:91, pl. 20: figs. 6, 7. Reussella atlantica Cushman.—Phleger and Parker, 1951:18, pl. 8: figs. 8, 9.—Parker, Phleger, and Peirson, 1953:13, pl. 4: figs. 12, 13.—Parker, 1954:519, pl. 7: fig. 28—Bandy, 1954:138, pl. 31: fig. 7; 1956:197.—Lankford, 1959, pl. 3: fig. 9. The holotype and seven paratypes from the Recent off Canaveral, Florida and abundant un- figured "topotypic" specimens from off the southeast coast of the United States also in the USNM collec- tions have been seen. The present specimens seem to be a bit smaller and more delicate and have slightly better developed spines than the primary types, but too few Discovery Bay specimens occur to be certain. As Hofker him- self has stated (1956:52: figs. 1, 4-7), his Reussella mortenseni, from the southern Caribbean, may be conspecific with Reussella atlantica, but we have seen no specimens so attributed and hesitate to synonymize. Hypotype: USNM 211322. standard mean deviation range station 1 0.02 0.14 0-1 station 3 0.52 0.80 (M Genus Chrysalidinella Schubert, 1908 Chrysalidinella aff. C. miocenica Cushman A single specimen is so referred. It resembles the holotype of Chrysalidinella miocenica and some of the specimens in the USNM collections so referred. The holotype of C. miocenica is, however, shorter and wider. The present specimen also resembles two paratypes of Chrysalidinella popei Andersen. Those specimens, from the Recent of South Pass, have, however, sharper keels that rise on each chamber from raised limbate sutures. The Discovery Bay specimen has a keel arising in the same fashion, but the sutures and keel are not as protruding and the keel does not become as sharp. The specimens in the collections (USNM) from off Onslow Bay, North Carolina referred to C. miocenica appear to be conspecific with C. popei paratypes. The scope of the present study does not allow solution of whether or not C. miocenica and C. popei are distinct species or of the problem that some speci- mens in the collections referred to Chrysalidinella NUMBER 31 81 dimorpha (H. B. Brady) are very similar to the present specimen. Hypotype: USNM 211342. standard mean deviation range station 1 0.00 0.00 0 station 3 0.02 0.14 0-1 Subfamily BULIMININAE Jones, 1875 Genus Sagrina d'Orbigny, 1839 Sagrina pulchella d'Orbigny in de la Sagra PLATE 3: FIGURES 3-6 Sagrina pulchella d'Orbigny, 1839:150, pl. 1: figs. 23, 24. Bolivina pulchella (d'Orbigny).—Cushman, 1922d:41, pl. 7: fig. 4; 1922a:25, pl. 1: figs. 8, 9; 1937:151, pl. 15: figs. 10, 11; 1941:10.—Cushman and Parker, 1931:15, pl. 3: fig. 20. Hofker (1956:49, pl. 4: figs. 21-36) reports this form as Bitubulogenerina pulchella from the southern Caribbean. As Bolivina pulchella, as well as the above-cited reference, unfigured conspecific specimens in the USNM collections come from Montego Bay, Jamaica, and (abundant specimens) from off the southeast coast of the United States. Figured hypotypes: USNM 211178, 211179. standard mean deviation range station 1 0.85 0.92 0-3 station 3 1.35 1.59 0-6 Sagrina pulchella d'Orbigny in de la Sagra subsp. primativa (Cushman) Bolivina pulchella (d'Orbigny) var. primativa Cushman, 1930: 47, pl. 8: fig. 12; 1937:90, pl. 12: fig. 6.—Phleger and Parker, 1951:14, pl. 7: fig. 3.—Bandy, 1956:193.—Todd and Bronnimann, 1957:34, pl. 8: figs. 9, 10. Bolivina pulchella primativa Cushman.—Parker, Phleger, and Peirson, 1953:6, pl. 4: figs. 2, 3.—Parker, 1954:516, pl. 7: fig. 36.—Lafikford, 1959:3: fig. 7. Some specimens from the Discovery Bay material are placed in this subspecies instead of the species, sensu stricto. (The holotype has been examined.) Whether or not two subspecifically different popu- lations are represented, however, is a question. A transition appears to occur and some specimens have been arbitrarily designated. A larger popula- tion would be needed in the present material to resolve this question. The large unfigured Recent USNM collection from off the southeast coast of the United States re- ferred to the species, sensu stricto, shows a complete and common range of variation in the character (length of the biserial part) given by Cushman to differentiate the subspecies. Probably this would be the case for at least some other Recent (at least) populations too, but in most cases available for some observation, there are only one to a few "char- acteristic" specimens from given locations included in synonymy and remarks. Therefore, though quite possibly all specimens could be included in the species, sensu stricto, a morphologic breakdown into species and subspecies has been retained herein. standard mean deviation range station 1 0.23 0.59 0-3 station 3 0.52 0.74 0-3 Family UVIGERINIDAE Haeckel, 1894 Genus Siphogenerina Schlumberger, 1882 Siphogenerina costata Schlumberger Siphogenerina cf. raphanus (Parker and Jones).—Cushman and Parker, 1931:17, pl. 3: figs. 25, 26. Siphogenerina raphana (Parker and Jones).—Todd and Bronnimann, 1957:36, pl. 9: fig. 7. Siphogenerina costata Schlumberger.—Loeblich and Tappan, 1964:C569, fig. 449: 1-4. Following Loeblich and Tappan (1964), these specimens are placed in Siphogenerina costata rather than "Siphogenerina" raphana, though they are conspecific with specimens referred to "S." raphana. Much taxonomic confusion arose regard- ing these two species, which now appears to be resolved by Loeblich and Tappan (1964:C553, C569). It does seem possible, however, as has been argued, that S. costata and Rectobolivina raphana are not only congeneric but conspecific, as some appear very similar. The specimen referred to "Siphogenerina raphanus (Parker and Jones)" by Cushman (1922a: 35, pl. 5: fig. 5) and herein ques- tionably referred to Rectobolivina raphana is closer to R. raphana than Siphogenerina costata, as typi- fied, but suggests a transition between the two forms. 82 Other unfigured specimens in the USNM collec- tions from Rio de Janeiro Harbor referred to "Siphogenerina raphanus (Parker and Jones)" are Siphogenerina costata of the present usage, and those figured and unfigured from Rio de Janeiro Harbor and the Gulf of Paria clearly are conspe- cific. Eight unfigured "Siphogenerina raphanus" from Puerto Rico may be Siphogenerina costata or Rectobolivina raphana, or may further indicate by their transitional appearance that one species only is represented. Examination of larger populations probably could resolve this question. Both forms occur in small numbers in the Discovery Bay sam- ples. Holotype: USNM 211339. standard mean deviation range station 1 0.56 0.85 0-3 station 3 0.25 0.44 0-1 Genus Trifarina Cushman, 1923 Trijarina occidentalis (Cushman) PLATE 3: FIGURES 7-10 Uvigerina angulosa Cushman [not Williamson], 1922a:34, pl. 5: figs. 3, 4. Angulogerina occidentalis (Cushman).—Cushman and Parker, 1931:17.—Todd and Bronnimann, 1957:36, pl. 9: figs. 5, 6. Numerous unfigured specimens in the USNM collections from off the southeast coast of the United States are assigned to "Angulogerina oc- cidentalis (Cushman)" and are conspecific with the present specimens. The figured specimen ascribed to "Angulogerina jamaicensis Cushman and Todd" by Parker (1954: 521, pl. 8: fig. 8) from the northeast Gulf of Mexico is very similar to Trifarina occidentalis, but shows the spines of T. jamaicensis on the early chambers. Considerable intraspecific variation exists in this species in the number of costae and angularity of test, especially in the upper part, where it may be typically triangular or rounded with lobate cham- bers. A large population is represented in the Dis- covery Bay samples. Figured hypotypes: USNM 211174, 211175. Hypotype: USNM 211223. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Superfamily DISCORBACEA Ehrenberg, 1838 Family DISCORBIDAE Ehrenberg, 1838 Subfamily DISCORBINAE Ehrenberg, 1838 Genus Discorbis Lamarck, 1804 Discorbis granulosa (Heron-Allen and Earland) PLATE 3: FIGURES 11, 12, 13 Discorbina valvulata (d'Orbigny), var. granulosa Heron-Allen and Earland, 1915:695, pl. 52: figs. 1-6. Discorbis valvulata (d'Orbigny) var. granulosa (Heron-Allen and Earland).—Todd, 1957, pl. 90: fig. 6. This distinctive species has a thick test wall that is coarsely perforate on the spiral side, imperforate on the umbilical side. The periphery is broadly rounded, lobate. The chambers are inflated, ar- ranged in two to two and a half whorls, with seven to eight chambers in the peripheral whorl. The sutures are radial, limbate, and deeply incised. The umbilicus is covered by overlapping imperforate triangular flaps extending from the bases of the chambers. The aperture is a narrow opening be- neath the chamber flap, remaining open in previous chambers. This species easily recognized by its extremely thick wall, the very coarse pores on the spiral side, and globose outline. Figured hypotype: USNM 211196. standard mean deviation range station 1 0.50 1.27 0-8 station 3 0.06 0.32 0-2 Discorbis mira Cushman PLATE 3: FIGURES 14, 15, 16 Discorbis mira Cushman, 1922a:39, pl. 6: figs. 10, 11; 1931:25, pl. 5: figs. 5, 6; 1941:11, pl. 2: figs. 4-6. This species is common throughout the present samples. It may be recognized by its biconvex test, and acute, entire periphery which bears a narrow imperforate keel. Figured hypotype: USNM 211195. standard standard mean deviation range mean deviation range station 1 15.90 8.38 2-37 station 1 1.44 2.35 0-13 station 3 22.46 14.10 0-72 station 3 14.08 11.06 0-60 NUMBER 31 83 Discorbis murrayi (Heron-Allen and Earland) PLATE 3: FIGURES 17, 18, 19 Rotalia murrayi Heron-Allen and Earland, 1915:722, pl. 53: figs. 27-34. This small species has an unequally biconvex test with a finely imperforate, hispid wall. The periph- ery is subangular, lobate. The chambers are glo- bose, arranged in one and one-half to two whorls, with five to seven chambers in the peripheral whorl. All chambers are visible on the flattened spiral side; only those of the peripheral whorl are visible on the strongly convex umbilical side. The sutures are radial and deeply depressed. The umbilicus is filled by an umbilical plug that is in turn covered by umbilical flaps extended from the bases of the chambers. The aperture is a narrow slit beneath the umbilical flap of the final chamber. The hispid surface of this species is quite dis- tinctive, somewhat resembling that of some plank- tonic species. This surface texture is sufficiently characteristic to allow identification of pathologic specimens on this basis alone. Figured hypotype: USNM 211190. side becomes strongly convex and the umbilical side less so. There are about two and one-half whorls visible on the spiral side of mature specimens, with eight to nine chambers in the peripheral whorl. The sutures are tangential, flush to raised on the spiral side, depressed on the umbilical side. The umbilicus is filled by a large umbilical plug. The umbilical ends of the chambers are extended into short, flat umbilical flaps. The aperture is an inte- riomarginal, umbilical-extraumbilical slit that is open under the chamber flaps, and open in previ- ous chambers. Genus Bronnimannia Bermudez, 1952 Bronnimannia caribaea (Cushman) Planulina caribaea Cushman, 1931:112, pl. 20: fig. 1. Two living specimens of Bronnimannia caribaea were found in the present material. This com- pressed species was originally described from Mon- tego Bay, Jamaica. Hypotype: USNM 211368. standard mean standard deviation range station 1 mean 0.04 deviation 0.20 range 0-1 station 1 1.40 1.84 0-8 station 3 0.00 0.00 0 station S 2.77 353 0-12 Discorbis rosea (d'Orbigny) Rotalia rosea d'Orbigny, 1826:272, Modele no. 35. Rotalina (Rotalina) rosea (d'Orbigny).—d'Orbigny, 1839:72, pl. 3, figs. 9-11. Truncatulina rosea (d'Orbigny).—Cushman, 1921:56, pl. 13: figs. 2, 3; 1922a:6, pl. 14: figs. 3-5. This species has not been included in the statis- tics of this study because it has a thick rose-colored test that matches the color of the stain (Rose Ben- gal) used in sample preparation to determine living protoplasm, and therefore the number of living specimens cannot be determined. The test is large, coarsely perforate, unequally biconvex. The periph- ery is acute, lobate in juveniles, entire in mature specimens, with a keel consisting of a row of closely spaced short spines. The spiral side is covered by a scattering of short, stubby spines, which are not present on the umbilical side. Juvenile specimens have a strongly convex umbilical side and a flat- tened spiral side; during development the spiral Genus Discorbinella Cushman and Martin, 1935 Discorbinella minuta, new species PLATE 3: FIGURES 20-25 Test small, trochoid, strongly compressed, con- cavoconvex; slightly convex spiral side moderately perforate, slightly concave umbilical side very finely perforate; periphery entire to somewhat lobate, with thick imperforate keel; chambers falciform, arranged in one and one-half to two whorls, seven to eight chambers in peripheral whorl, gradually increasing in size as added; chambers partially involute on both sides, only chamber flaps of pre- vious whorl visible on umbilical side; sutures lim- bate, strongly curved, raised to slightly depressed on spiral side, depressed on umbilical side; prox- imal ends of chambers on umbilical side bearing short imperforate flaps that approach small central umbilical plug; aperture narrow umbilical slit 84 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY station 1 station 3 between chamber flaps and plug, remaining open in previous chambers. This small species is a fairly common component of the population at station 3, but only a few specimens occur at station 1. It is somewhat difficult to separate from juvenile specimens of Cymbalopo- retta atlantica, due principally to the small size and flattened conical shape of both. Upon close exami- nation, however, the strongly curved sutures that are depressed on the umbilical side and the finely punctate umbilical-side wall distinguish this species. Descriptive statistics of the maximum diameter, minimum diameter, and thickness measured in mm for the holotype and three paratypes are: Genus Eoeponidella Wickenden, 1949 Eoeponidella pacifica (Uchio) Asterigerinata pacifica Uchio, 1960:67, pl. 10: figs. 26-31. Eoeponidella delicatula Seiglie, 1965:511, pl. 5, fig. 11. mean 0.08 0.13 This tiny species occurs infrequently throughout the present suite of samples. The present specimens have been compared with the type material and appear to be conspecific. standard deviation range 0.28 0-1 0.33 0-1 range 0.150-0.200 0.130-0.170 0.040-0.070 range 0-1 0-7 standard mean 0.173 0.145 0.051 deviation maximum diameter minimum diameter thickness Figured holotype: Figured paratype: 0.022 0.019 0.013 USNM 211163. USNM 211167. mean 0.06 0.92 standard deviation station 1 station 3 0.24 1.27 Discorbinella sp. A single broken specimen of Discorbinella is insufficiently diagnostic to allow further identifica- tion, although it exhibits sufficient characteristics to place it in Discorbinella. The test is concavo- convex. The low convex spiral side is thickened and finely perforate; the concave umbilical side is thin walled and coarsely perforate. The periphery is acute, entire, with a thin keel. The chambers are falciform, not inflated, arranged in one and one- half whorls, with six chambers in the peripheral whorl, chambers increasing rapidly in size as added. The sutures are arcuate, flush on spiral side, de- pressed on umbilical side. Although the final cham- bers are not preserved, previous chambers possess umbilical interiomarginal slit apertures. Hypotype: USNM 211267. standard mean deviation range station 1 0.00 0.00 0 station 3 0.02 0.14 0-1 Genus Epistominella Husezima and Maruhasi, 1944 Epistominella? sp. A single broken specimen questionably referable to Epistominella was found at station 3, but was lost during study. It was minute, bioconvex, and with about eight chambers in the peripheral whorl. The sutures were somewhat depressed on the um- bilical side. Since the final chamber was broken, the character of the aperture could not be deter- mined. The periphery was rounded and lobate. Genus Helenina Saunders, 1961 Helenina anderseni (Warren) Pseudoeponides anderseni Warren, 1957:39, pl. 4: figs. 12- 15.—Parker and Athern, 1959:341, pl. 50: figs. 28-31. Helenina anderseni (Warren).—Saunders, 1957:374, figs. 1, 2.—Todd and Low, 1971:18, fig. 2. Helenina anderseni (Warren).—Todd and Low, 1971:15, pl. 3: fig. 1. This species occurs often at station 1 but only once at station 3. Hypotype: USNM 211357. standard station 1 mean 0.46 deviation 1.13 range 0-7 station 3 0.02 0.14 0-1 Genus Neoconorbina Hofker, 1951 Neoconorbina terquemi (Rzehak) Rosalina orbicularis Terquem [not d'Orbigny, 1850], 1875:75, pl. 9: fig. 4. NUMBER 31 85 Discorbina terquemi Rzehak, 1888:228. Neoconorbina terquemi (Rzehak).—Parker, 1958:267, pl. 3: figs. 26, 27. Discorbis orbicularis (Terquem).—Cushman, 1922a:38, pl. 5: fig. 10; 1931:27, pl. 6: fig. 3. Conorbina orbicularis (Terquem).—Parker, 1954:522, pl. 8: figs. 13, 14. This species occurs in small numbers in several of the present samples. Hypotype: USNM 211372. standard mean deviation range station 1 0.23 0.66 0-3 station 3 0.44 0.99 0-4 Genus Patellinella Cushman, 1928 Patellinella sp. This form, with a low conical spire, superficially closely resembles Patellina corrugata Williamson and may have been mistaken for it in some in- stances. It has, however, the generic characters of Patellinella. Only one specimen was found at sta- tion 1, but, unfortunately, we lost it. Genus Rosalina d'Orbigny, 1826 Rosalina bulbosa (Parker) "Discorbis" bulbosa Parker, 1954:523, pl. 8: figs. 10-12. This species appears in small numbers in many of the present samples. These specimens are iden- tical to the holotype, but appear quite different from the illustration of "Rosalina bulbosa (Parker)" given by Akers and Dorman (1964:51, pl. 9: figs. 14, 15), which they collected from the Gulf Coast Pleistocene. This species has all the characteristics imputed to Rosalina (Loeblich and Tappan, 1964:C584), with the exception of the umbilical chamber flaps. The aperture of the present species is an interiomarginal arch extending from the large, open umbilicus to the rounded periphery. A narrow apertural lip is present. Although this species lacks the rosalinid chamber flap, it seems unwarranted to erect yet another genus to include only this species. Hypotype: USNM 211358. standard mean deviation range station 1 0.04 0.20 0-1 station 3 0.54 0.85 0-4 Rosalina candeiana d'Orbigny PLATE 4: FIGURES 1-3 Rosalina candeiana d'Orbigny, 1839:97, pl. 8: figs. 2-4. Truncatulina candeiana (d'Orbigny).—Cushman, 1921:57, pl. 13: fig. 5; 1922a:47, pl. 6: figs. 7-9. Truncatulina cora (d'Orbigny).—Cushman, 1922a:48, pl. 7: figs. 3-5. Specimens from the present samples appear con- specific with d'Orbigny's figured specimens from beach sands of Cuba. The test is planoconvex, low trochospiral, with a coarsely perforate wall. The periphery is subacute and lobate. The chambers are inflated on the spiral side, flattened on the umbil- ical side, arranged in about three whorls, with six chambers in the peripheral whorl. The sutures are straight, radial, and depressed. The umbilicus is rather small and open, partially covered by wide, imperforate, overlapping umbilical flaps extending from the base of each chamber, leaving a stellate umbilical opening as illustrated by d'Orbigny. The aperture is a low interiomarginal extraumbilical- umbilical arch bordered above by a prominent apertural lip, remaining open in earlier formed chambers. Figured hypotype: USNM 211186. Hypotype: USNM 211359. standard mean deviation range station 1 0.17 0.56 0-3 station 3 0.06 0.24 0-1 Rosalina concinna (Brady) PLATE 4: FIGURES 4-6 Discorbina concinna Brady, 1884:646, pl. 90, figs. 7, 8. Discorbis concinnus (Brady).—Bandy, 1956:193, pl. 31: fig. 4. This species is fairly abundant in the present suite of samples, especially at station 3. On casual observation, it may be confused with large flattened specimens of Rosalina globularis, from which it differs in having flush sutures and a thickening of the test over the earlier whorls on the spiral side, and in lacking the pronounced chamber flaps on 86 the umbilical side. It differs from Neoconorbina terquemi in having a rounded, nonkeeled periphery, in having an extremely finely perforate to imper- forate umbilical side, and in lacking the pro- nounced chamber flaps of N. terquemi, the chamber flap of this species being a wide, short, rounded extension of the base of the chamber. Figured hypotype: USNM 211187. standard mean deviation range station 1 0.65 1.06 0-4 station 3 2.69 2.66 0-8 Rosalina floridana (Cushman) PLATE 4: FIGURES 7-9 Discorbis floridana Cushman, 1922a:39, pl. 5: figs. 11, 12; 1931:21, pl. 4: figs. 7, 8.—Cushman and Parker, 1931:18, pl. 4: fig. 5. Discorbis micens Cushman.—Todd, 1957:290, pl. 90: fig. 7. This species appears in considerable numbers throughout the present samples. Study of the types and of the present specimens shows that this is a variable species, but it exhibits definite charac- teristics. The test is a low trochospiral coil, with a wall that is moderately perforate on the spiral side, finely perforate to imperforate on the umbilical side. The area above the umbilicus is imperforate in most specimens. The periphery is broadly rounded, lobate. The chambers are inflated, ar- ranged in two to two and a half whorls, with five to seven chambers in the peripheral whorl, the chambers increasing rather rapidly in size as added. Early sutures are limbate and flush, becoming sim- ple and somewhat depressed on the spiral side; all sutures are deeply incised on the umbilical side. The umbilicus is large and partially covered by very large triangular chamber flaps. Characteristic of the species is a deep furrow at the proximal edge of the flap on each chamber. The aperture is a low interiomarginal extraumbilical-umbilical arch with lip. Many different forms have been referred to this species, probably because of the poor and mis- leading original illustration, and because the final chamber of the holotype is broken. The paratype and topotypic specimens, however, give a clear picture of the characters and variation of this species. Figured hyptoype: USNM 211185. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Hypotype: USNM 211360. standard mean deviation range station 1 9.06 8.74 0-52 station 3 4.35 3.74 0-17 Rosalina globularis d'Orbigny PLATE 4: FIGURES 10-12 Rosalina globularis d'Orbigny, 1826:271, pl. 13: figs. 1-4. Discorbis columbiensis Cushman, 1925:43, pl. 6: fig. 13. Tretomphalus bulloides (d'Orbigny).—Cushman, 1934:86, pl. 11: fig. 2. Tretomphalus myersi Cushman, 1943:26, pl. 6: figs. 4-6. This, the type-species of Rosalina, is probably the most variable and misidentified species of the genus. The diagnostic characteristics of R. globu- laris have been clearly set forth by Douglas and Sliter (1965). They found considerable variation in laboratory-cultured specimens, including specimens of the type-species of Tretomphalus, T. bulloides (d'Orbigny), thus placing that genus in synonymy with Rosalina, as has been discussed under Cym- baloporetta atlantica. Many specimens throughout the present suite of samples fall within the range of variation reported by Douglas and Sliter, and have been so identified. Figured hypotype: USNM 211184. Hypotype: USNM 211361. standard mean deviation range station 1 4.67 4.27 0-15 station 3 17.35 15.90 2-83 Rosalina subaraucana (Cushman) PLATE 4: FIGURES 13-15 Discorbis subaraucana Cushman, 1922a:41, pl. 7: figs. 1, 2; 1931:32, pl. 7: fig. 2. Discorbis floridana Cushman.—Phleger and Parker, 1951:20, pl. 10: fig. 4.—Parker, Phleger, and Peirson, 1953:7, pl. 4: figs. 18, 19. Discorbis floridanus Cushman.—Bandy, 1954:136, pl. 31, fig. 1. This species occurs in considerable numbers in samples collected from this study. It is readily identified by its low trochospiral, nearly plano- convex test that is finely perforate on both sides, the fairly rapid increase in chamber size as added, the limbate, curved flush to slightly depressed sutures, and the small umbilicus and corres- pondingly small chamber flaps. NUMBER 31 Figured hypotype: USNM 211173. mean 2.77 10.98 standard deviation station 1 station 3 2.60 8.97 range 0-14 1-38 87 Parker's types. Parker (1954) stated that occasionally chambers of the second whorl are visible on the dorsal side. None of our specimens exhibit this feature, but otherwise are identical. Hypotype: USNM 211297. station 1 station 3 range 0 0-1 Rosalina sp. Test biconvex, low trochospiral coil, globose; wall moderately to coarsely perforate; periphery rounded, lobate; chambers inflated, arranged in about three whorls, with eight chambers in the peripheral whorl; sutures strongly curved, nearly tangential, deeply incised; umbilicus large, open, partially covered by overlapping triangular chamber flaps; aperture an interiomarginal extraumbilical-umbilical slit, remaining open in chambers previous to the final. Three specimens of this species were found in the present samples, all from station 3. Because of the scarcity of specimens, no study of the wall structure has been undertaken, but the appearance of the wall, die open umbilicus, and the chamber flaps all suggest placement in Rosalina. Hypotype: USNM 211363. mean 0.00 0.06 station 1 station 3 standard deviation 0.00 0.24 Rosalina? sp. A single broken and deformed specimen from station 1 is tentatively classified as a Rosalina. The test is a low trochoid coil. The chambers are strongly inflated, arranged in about two and one- half whorls, with six chambers in the final whorl. The last few chambers are eccentric, unrolling away from the axis of coiling. The periphery is rounded, lobate. The sutures are radial, depressed. The um- bilicus is rather large, partially closed by chamber flaps extending from the base of each chamber. The nature of the aperture is uncertain because of the apertural face of the final chamber is broken. Genus Stetsonia Parker, 1954 Stetsonia minuta Parker Stetsonia minuta Parker, 1954:534, pl. 10: figs. 27, 29 The specimens from Jamaica closely resemble standard mean deviation range 0.17 0.48 0-2 0.06 0.24 0-1 Subfamily BAGGININAE Cushman, 1927 Genus Baggina Cushman, 1926 Baggina aff. B. philippinensis (Cushman) PLATE 4: FIGURES 16-18 Occasional specimens comparable to Baggina philippinensis (Cushman) occur in samples from station 3. These specimens have inflated, coarsely perforate tests, only the area over the aperture lacking pores. The chambers are arranged in one and one-half to two whorls, with five chambers in the peripheral whorl, increasing rapidly in size as added. Sutures are slightly curved, simple, strongly depressed on the umbilical side, and slightly de- pressed on the spiral side. The aperture is a low interiomarginal extraumbilical-umbilical arch. These specimens are smaller than the holotype of "Pulvinulina philippinensis," and differ some- what from that specimen in the shape of the cham- bers and the size of the aperture. They do, however, appear conspecific with small specimens referred to Baggina philippinensis in the USNM reference collection. Figured hypotype: USNM 211172. Hypotype: USNM 211354. standard station 1 mean 0.00 deviation 0.00 range 0 station 3 0.13 0.39 0-2 Genus Cancris de Montfort, 1808 Cancris sagra (d'Orbigny) Rotalina (Rotalina) sagra d'Orbigny, 1839:77, pl. 5: figs. 13- 18. Cancris sagra (d'Orbigny).—Cushman, 1931:74, pl. 15: fig. 2 — Cushman and Todd, 1942:77, pi. 19: figs. 4-7.—Bandy, 88 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 1954:136, pl. 30: fig. 9.—Parker, 1954:532, pl. 10: figs. 15, 21.—Todd and Bronnimann, 1957:37, pl. 11: fig. 4. Living specimens of this species are quite rare in the present material. These specimens vary some- what in biconvexity and chamber shape, but fall within the range of variation reported for this species by Cushman and Todd (1942), and ex- hibited by specimens of Cancris sagra in the USNM reference collection. Hypotype: USNM 211369. standard mean deviation range station 1 0.06 0.24 0-1 station 3 0.06 0.24 0-1 Family GLABRATELLIDAE Loeblich and Tappan, 1964 Genus Glabratella Dorreen, 1948 Glabratella altispira, new species PLATE 4: FIGURES 19-24 Test minute, high trochospiral, about as high as wide, finely perforate on spiral side, imperforate on umbilical side, periphery rounded, lobate, cham- bers globose, about one and one-half times longer than wide, arranged in three to four whorls, with four chambers in each whorl, chambers increase gradually in size as added; sutures deeply incised, somewhat curved on spiral side, radial on umbilical side; umbilicus large, open; aperture arched um- bilical opening, remaining open in previous cham- bers, umbilical striae indistinct or lacking. This species is fairly common in the present sam- ples, occurring more often at station 1 than at sta- tion 3. Of the previously described species of Glabratella, it seems most like G. erecta (Sidebot- tom) in general shape, differing in being smooth rather than coarsely rugose on the spiral side, and in having less oblique sutures than G. erecta. Descriptive statistics of the maximum diameter, minimum diameter, and thickness measured in mm for the holotype and three apartypes are: standard mean deviation range maximum diameter 0.095 0.041 0.090-0.100 minimum diameter 0.089 0.002 0.085-0.090 thickness 0.109 0.002 0.105-0.110 Figured holotype: USNM 211165. Figured paratype: USNM 211160. Paratypes: USNM 211276, 211277. standard mean deviation range station 1 0.50 1.38 0-8 station 3 0.17 0.52 0-2 Glabratella braziliensis Boltovskoy Glabratella braziliensis Boltovskoy, 1959:90, pl. 14: figs. 8-11. This rather large species of Glabratella is found in several samples in this study. It is readily iden- tified by its hemispherical shape, roughened convex spiral side with tangential, flush sutures, and de- pressed umbilical side with radiating rows of granules and depressed radial sutures. Hypotype: USNM 239905. standard mean deviation range station 1 0.08 0.35 0-2 station 3 0.02 0.14 0-1 Glabratella cf. G. carinata Seiglie and Bermudez PLATE 5: FIGURES 1-3 A few specimens of Glabratella have been found at station 3 that are referred to Glabratella cf. G. carinata Seiglie and Bermudez. These specimens differ from the illustrations and description of that species in having an open umbilicus rather than having the umbilicus filled with granular material. The aperture appears to open into the rather large umbilicus. Because the amount of variation within species of Glabratella is not yet understood, it seems preferable to compare these specimens with G. carinata rather than erect a new species. Figured hypotype: USNM 211193. standard station 1 mean 0.00 deviation 0.00 range 0 station 3 0.08 0.35 0-2 Glabratella compressa, new species PLATE 5: FIGURES 4-6, 10-12 Test minute, compressed, concavoconvex, finely perforate, transparent; periphery rounded, lobate; NUMBER 31 chambers about twice as long as wide, arcuate, somewhat inflated, arranged in about two whorls with four chambers in peripheral whorl; sutures curved, tangential, slightly depressed on spiral side, radial, straight, moderately depressed on umbilical side; umbilicus large, open; aperture basal umbil- ical opening, remaining open in previous chambers; radial umbilical striae indistinct or lacking. This is a common species of Glabratella in our samples, occurring in six samples from station 3 and 17 samples from station 1. The tiny trans- parent, strongly compressed test with wide umbilical opening clearly sets this species apart from other described species of this genus. Descriptive statistics of the maximum diameter, minimum diameter, and thickness measured in mm for the holotype and three paratypes are: maximum diameter minimum diameter thickness standard mean deviation range 0.108 0.021 0.085-0.135 0.094 0.021 0.070-0.120 0.048 0.012 0.035-0.060 Figured holotype: USNM 211166. Figured paratype: USNM 211161. Paratypes: USNM 211278, 211282. standard mean deviation range station 1 0.44 0.68 0-3 station 3 0.52 2.89 0-20 Glabratella hexacamerata Seiglie and Bermudez PLATE 5: FIGURES 7-9 Glabratella hexacamerata Seiglie and Bermudez, 1965:31, pl. 1: figs. 6, 7. This is the most common species of Glabratella in the present suite of samples. It is easily recog- nized by its possession of six globular chambers per whorl. The low convex spiral side is roughened; the umbilical side is smooth. The umbilicus is large and open. Figured hypotype: USNM 211194. Hypotype: USNM 211264. standard mean deviation range station 1 0.31 0.69 0-3 station 3 0.79 1.15 0-5 89 Glabratella testigoensis Seiglie and Bermudez Glabratella testigoensis Seiglie and Bermudez, 1965:37, pl. 2: figs. 4-7; pl. 4: figs. 4-6. A single specimen of Glabratella occurring at station 3 appears to be conspecific with topotypic specimens of G. testigoensis in the USNM collec- tions. Although the present specimen is less rugose on the spiral side than most specimens on the topotypic slide, it appears to fall within the range of variation represented on that slide. The speci- men was lost during study. Glabratella spp. PLATE 5: FIGURES 13-21 Description of the form represented by hypotype USNM 211189 and treated ecologically as Glabra- tella sp. A follows. Test small, moderately highly trochospiral; wall finely perforate and roughened on the spiral side, imperforate but covered by rows of granules on umbilical side; periphery rounded, lobate; chambers strongly inflated, arranged in two to two and one-half whorls with five chambers in the peripheral whorl, increasing rapidly in size as added, a thin vertical costa on entire umbilical side of each covered by radiating rows of granules, distal end of each chamber, with exception of final chamber, protruded as a spine on the peripheral edge of spiral side of each chamber, then extending toward center of test along spiral side of chamber to chamber of previous whorl, final chamber bear- ing only rudimentary costa along spiral side; sutures deeply incised, radial; umbilicus moderately large, filled with granules; aperture not seen. Only two specimens of this form were observed, but the morphologic characteristics appear to be unique, quite unlike any other species present in this study, any species of Glabratella in the USNM collections, or any species in the literature. Figured hypotype: USNM 211189. Description of the form represented by hypotype USNM 211168 and treated ecologically as Glabra- tella sp. B follows. Test small, low trochospiral; wall finely perforate on spiral side, imperforate on umbilical side; periphery rounded, lobate; cham- bers inflated, somewhat longer than high, arranged in two whorls, six in final whorl, increasing gradu- ally in size as added; sutures arcuate, strongly 90 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY depressed on spiral side, nearly flush on umbilical side; umbilicus large, open; ornamentation on chambers of final whorl fine striae radiating from umbilicus and extending about halfway to periph- ery; aperture interiomarginal umbilical arch, previous apertures remaining open. In general morphology, this species closely resembles minute specimens of Rosalina floridana (Cushman), differing mainly in possession of fine radiating umbilical striae and in lacking the char- acteristic rosalinid interiomarginal arched aperture. Indeed, Douglas and Sliter (1965) have suggested Uiat R. floridana may belong in the genus Glabra- tella on the basis of reported plastogamic repro- duction and the open pustulose aperture. Juvenile specimens of Rosalina floridana with rosalinid apertures and no umbilical striae, however, are numerous throughout the present suite of samples. It is possible that characteristic specimens of R. floridana represent the "microspheric" stage of this species and that the present specimens represent the "megalospheric" reproductive stage, but this cannot now be demonstrated. Lee et al. (1963) observed the life cycle of R. floridana and reported only slight morphologic differences between the two generations; however, .they found no significant differences in relative size of either proloculus or mature test, finally resorting to coiling direction statistics as the best method of discerning the type of generation. Although all three specimens of this group of Glabratella spp. represented by hypotype USNM 211168 exhibit sinistral coiling as reported by Lee et al., for the megalospheric generation, specimens of Rosalina floridana in the present sam- ples are about equally left and right coiled. There- fore, it seems probable that a new species is repre- sented here, although insufficient material is present to erect a new species at this time. Figured hypotype: USNM 211168. Description of the form represented by hypotype USNM 211191 and treated ecologically as Glabra tella sp. C follows. Test minute, low trochospiral, wall very finely perforate; periphery rounded, lobate; chambers strongly inflated, appearing some- what longer than high on spiral side, triangular, about as high as wide on umbilical side, arranged in about two whorls, with six in peripheral whorl; sutures arcuate on spiral side, radial and nearly straight on umbilical side, deeply depressed on both sides; umbilicus small, open; umbilical area lacking granules or striae; aperture minute umbili- cal opening. Although the characteristics of this species appear unlike those of any described species, an insufficient number of specimens is present to make a new name. Figured hypotype: USNM 211191. Glabratella sp. A standard station 1 mean 0.02 deviation 0.14 range 0-1 station 3 0.02 0.14 0-1 Glabratella sp. B standard station 1 mean 0.02 deviation 0.14 range 0-1 station 3 0.04 0.20 0-1 Glabratella sp. C standard station 1 mean 0.00 deviation 0.00 range 0 station 3 0.04 0.20 0-1 Genus Glabratellina Seiglie and Bermudez, 1965 Glabratellina sp. PLATE 5: FIGURES 22-24 Test small, moderately high trochospiral, coarsely perforate and rugose on spiral side, imperforate and covered by granules on umbilical side; periphery rounded, lobate; chambers moderately inflated, slightly longer than high, arranged in two and one- half to three whorls, five to six chambers in periph- eral whorl, chambers increasing gradually in size as added; sutures arcuate, depressed on both sides; granules on umbilical side arranged in numerous radiating rows, covering umbilicus and extending to periphery; aperture a narrow extraumbilical- umbilical slit along base of linguiform umbilical extension of final chamber. Of the previously described species, Glabratellina sp. seems to be most like G. duclozi Seiglie and Bermudez, from the Miocene of Cuba, differing in having a more compressed test, less inflated cham- bers, and lacking the strong costae on the spiral side. Glabratellina sp. is a rare species throughout the present suite of samples, occurring in three NUMBER 31 91 samples from station 1 and five samples from sta- tion 3. Figured hypotype: USNM 211159. Hypotype: USNM 211280. standard mean deviation range station 1 0.08 0.35 0-2 station 3 0.15 0.50 0-3 Glabratellina sagrai (Todd and Bronnimann) PLATE 5: FIGURES 25-27 Rosalina sagrai Todd and Bronnimann, 1957:37, pl. 9: fig. 22. Two small specimens conspecific with the holo- type of "Rosalina" sagrai Todd and Bronnimann have been found in the present material. This spe- cies is best placed in the genus Glabratellina, based on the characteristics exhibited on the umbilical side of the test. This species has a small plano- convex test. The moderately high-spired spiral side is finely perforate, the flattened umbilical side is imperforate. The periphery is subacute and slightly lobate. The chambers are arranged in three whorls, with six to seven chambers in the peripheral whorl. Sutures are limbate, slightly depressed, curved and oblique on the spiral side, depressed and radial, but indistinct on the umbilical side. The umbilical area is covered by granules that extend toward the periphery along the sutures and obscure the sutures. The chambers bear fine radiating striae on the um- bilical side where not obscured by granules. The final chamber bears a linguiform umbilical exten- sion or flap that is characteristic of the genus Glabratellina, and beneath which lies the aperture that extends about two-thirds of the distance from the umbilicus to the periphery. Figured hypotype: USNM 211192. standard mean deviation range station 1 0.06 0.32 0-2 station 3 0.00 0.00 0 Genus Angulodiscorbis Uchio, 1953 Angulodiscorbis corrugata (Millett) PLATE 6: FIGURES 1-3 Discorbina corrugata Millett, 1903:700, pl. 7: fig. 5. Discorbis corrugata (Millett).—Bermudez, 1949:237, pl. 15. figs. 7-9. Numerous specimens that appear to belong to this species occur in the present suite of samples. This small species has a high trochospiral test, the moderately inflated chambers of which are arranged three to three and a half whorls, with five chambers per whorl. The sutures are flush in early whorls, but depressed in the peripheral whorl. The rather small umbilicus is generally filled with granular material. The aperture is a narrow interiomarginal slit, extending from the umbilicus about halfway to the periphery. The umbilical side of the test is cov- ered by radial rows of granules, and the spiral side is dominated by five strong costae that extend along the centers of the chambers from the proloculus to the somewhat angular periphery. This species closely resembles the illustration of Glabratellina duclozi of Seiglie and Bermudez (1965), but lacks the linguiform umbilical extension characteristic of that genus and, by inference, of that species. This species differs from Angulodiscorbis pyra- midalis (Heron-Allen and Earland, 1924) in hav- ing five chambers in each whorl instead of the four characteristic of that species. Seiglie and Bermudez (1965) attributed five chamber-per-whorl specimens to A. pyramidalis, although the original descrip- tion and figures specify four chambers per whorl. Two specimens in the USNM collections labeled Angulodiscorbis "pyramidata" (Heron-Allen and Earland) from Fiji were inspected and found to have four chambers per whorl as originally de- scribed. The specimens identified by Bermudez (1949) as Discorbis corrugata (Millett) from the upper Oligo- cene of the Dominican Republic have been closely inspected and appear conspecific with the present suite of specimens, indicating a range for the spe- cies of Oligocene to Recent. Although Seiglie and Bermudez (1965) placed this species in Glabratella, it is more reasonably placed in Angulodiscorbis, following the generic criteria presented by Loeblich and Tappan in the Treatise (1964). Figured hypotype: USNM 211201. standard mean deviation range station 1 0.52 1.11 0-6 station 3 2.04 3.48 0-18 station 1 station 3 92 Family SIPHONINIDAE Cushman, 1927 Genus Siphonina Reuss, 1850 Siphonina pulchra Cushman Siphonina pulchra Cushman, 1919:42, pl. 14: fig. 7, 1922a:49, pl. 7: figs. 11, 12; 1931:69, pl. 14: figs. 2, 3.—Phleger and Parker, 1951:24, pl. 12: fig. 15.—Parker, 1954:532, pl. 10: figs. 10, 11. The species occurs rarely only at station 1. All specimens but one are juveniles, but appear to be conspecific with Cushman's specimens from the Miocene of Cuba. Hypotype: USNM 211364. standard mean deviation range station 1 0.15 0.36 0-1 station 3 0.00 0.00 0 Genus Siphoninella Cushman, 1927 Siphoninella} sp. A single specimen occurring at station 1 is so re- ferred. It fits the generic description given by Loe- blich and Tappan (1964:C591), but does not have the fimbriate keel so common in the family Siphon- inidae. It is rather small, with a single whorl, coiled early stage, followed by two biserial chambers and a uniserial final chamber ending in a large, flaring, open aperture with a phialine lip. The specimen is compressed and fairly flat surfaced, with a rounded margin. The pores are medium sized, fairly dense, and evenly distributed in the translucent wall. The sutures are narrow and quite depressed. It is not possible to identify this form more closely at this time. Conceivably, it may be a juvenile specimen whose plan of growth would change further, but this seems unlikely. Hypotype: USNM 211340. Family ASTERIGERINIDAE d'Orbigny, 1839 Genus Asterigerina d'Orbigny, 1839 Asterigerina carinata d'Orbigny Asterigerina carinata d'Orbigny, 1839:118, pl. 5: fig. 25; pl. 6: figs. 1, 2.—Cushman, 1921:60, pl. 14: figs. 7, 8; 1922a:54, figs. 4-6; 1931:77, pl. 15: figs. 4, 5.—Phleger and Parker, SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 1951:26, pl. 14: fig. 2— Bandy, 1954:135, pl. 31: fig. 5.— Parker, 1954:532, pl. 10: figs. 16, 17.—Todd and Low, 1971:15, pl. 3: fig. 3. This abundant species occurs at both stations 1 and 3. This is the characteristic species of Asteri- gerina in the Caribbean, originally described by d'Orbigny from the beach sands of Cuba and Ja- maica. Hypotype: USNM 211356. mean 3.90 3.75 range 0-26 0-13 standard deviation 5.83 3.42 Asterigerina sp. A single, broken but living specimen of Asteri- gerina was found in a sample from station 1. Al- though this specimen is clearly different from A. carinata, it is insufficiently well preserved to allow specific identification. Hypotype: USNM 211239. Family EPISTOMARIIDAE Hofker, 1954 Genus Epistomaroides Uchio, 1952 Epistomaroides} sp. A single, small, broken specimen from station 3 has been questionably referred to Epistomaroides on the basis of probable small secondary chambers on the umbilical side of the test and apparent slight retral processes over the sutures, as are characteris- tic of the genus. Because of the small size and fragility of the specimen, more positive identifica- tion cannot be made. Specimen was lost during study. Genus Eponidella Cushman and Hedberg, 1935 Eponidella sp. A single broken specimen of an apparently un- reported species has been found in the present suite of samples at station 1. This specimen has a bicon- vex, low trochospiral, coarsely perforate test. The periphery is rounded and entire. The chambers are somewhat inflated, arranged in one and one-half whorls, each subdivided into small chamberlets vis- NUMBER 31 93 ible on the umbilical side. There are nine chambers in die peripheral whorl. The sutures are limbate, slightly curved, and flush. The umbilicus is filled by a large umbilical plug. The final chamber is broken so that the aperture is not preserved, but the aperture of the penultimate chamber is a single- lipped, oval areal opening. Specimen was lost dur- ing study. Superfamily SPIRILLINACEA Reuss, 1862 Family SPIRILUNIDAE Reuss, 1862 Subfamily SPIRILLININAE Reuss, 1862 Genus Spirillina Ehrenberg, 1843 Spirillina vivipara Ehrenberg PLATE 6: FIGURES 4-6 Spirillina vivipara Ehrenberg, 1843:323, 422, pl. 3(VII): fig. 41.—Cushman, 1922a:37, pl. 5: fig. 7; 1931:3, pl. 1: figs. 2-4; 1941:11.—Phleger and Parker, 1951:25, pl. 13: figs. 3, 4.— Parker, 1954:522, pl. 8: figs. 15, 16. Ouier synonymous specimens are found so re- ferred in die unfigured USNM collections from the Dry Tortugas, Puerto Rico, Tongue of the Ocean, and Bermuda. In the Discovery Bay material speci- mens are rare and with one exception occur only at station 3. Figured hypotypes: USNM 187958. standard mean deviation range station 1 0.02 0.14 0-1 station 3 0.38 0.87 0-4 Genus Mychostomina Berthelin, 1881 Mychostomina revertens (Rhumbler) PLATE 6: FIGURES 7-12 Sprillina vivipara Ehrenberg var. densepunctata Cushman.— Cushman and Parker, 1931:18, pl. 4: fig. 1. fConicospirillina atlantica Cushman, 1947:91, pl. 20: fig. 8. Mychostomina revertens (Rhumbler).—Smith and Isham 1974:66, pl. 1: figs. 1-3, 7-9. A few specimens of this species occur in quite a few Discovery Bay samples. The specimen figured by Cushman and Parker is conspecific; however, three unfigured specimens from their collection re- ferred as above are Spirillina vivipara. The synon- ymy of "Conicospirillina atlantica" is questioned only for the reason that the specimens so referred from off the southeast coast of the United States are generally larger than Mychostomina revertens of the present report and conspecific specimens from the West and South Pacific. They certainly are congeneric. Figured hypotypes: USNM 187956, 187957. Hypotypes: USNM 211309, 211310, 211311, 211312, 211313, 211314, 211315, 211316, 211317. standard mean deviation range station 1 0.77 1.29 0-7 station 3 1.98 2.42 0-9 Subfamily PATELLININAE Rhumbler, 1906 Genus Patellina Williamson, 1858 Patellina advena Cushman fig. 9.- Patellina advena Cushman, 1922c:135, pl. 31: Cushman and Parker, 1931:18, pl. 4: fig. 3. Patellina corrugata Williamson.—Cushman, 1941:10. This species is very rare in our material and oc- curs only at station 3. The similarity between this species and Patellina corrugata Williamson is great. In the two cases where populations of specimens have been seen, however, they seem distinct. All die many unfigured specimens from die Arctic in die USNM collections referred to P. corrugata seem cor- rectly so. Several unfigured specimens from off the southeast coast of the United States are referred to P. corrugata, but appear to fall into bodi species, with little suggestion of gradation. The holotype (the only primary type available) of P. advena is from the Oligocene of Mississippi. It would be very helpful to see populations of speci- mens from a stratigraphic continuum and more geo- graphic areas in the Recent to observe more fully the distinctness of die two species. As it stands, die species appear distinct, but a minor uncertainty exists. Among Recent specimens, Cushman and Parker (1931:18) remarked that P. corrugata is a cold-water form and P. advena a warm-water form. The latter has been identified from the Recent of die South and Southwest Pacific; Juan Fernandez Islands, 94 Chile; and Rio de Janeiro Harbor. Unfigured speci- mens in the USNM collections from Bermuda and Jamaica identified as P. corrugata appear to be P. advena. Cushman and Parker's (1951:23, pl. 17: fig. 4) specimen identified as P. corrugata (from 128 m in the Gulf of Mexico) certainly appears to be P. corrugata. Hypotype: USNM 211319. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY ini, but they closely resemble small specimens from that area. In this regard, it is interesting to note that the only large specimens in the USNM collec- tions are from the Rimini area, and that these are the only specimens that show the complex channel- ing on the umbilical side of the test that is con- sidered "typical" for this species. Hypotype: USNM 211352. standard standard station 1 station 3 mean 0.00 0.10 deviation 0.00 0.59 range 0 0-4 station 1 station 3 mean 17.25 20.31 deviation 15.36 31.40 range 3-99 0-158 Superfamily ROTALIACEA Ehrenberg, 1839 Family ROTALIIDAE Ehrenberg, 1839 Subfamily ROTALIINAE Ehrenberg, 1839 Genus Ammonia Bninnich, 1772 Ammonia beccarii (Linn6) Nautilus beccarii Linne, 1758:710. Rotalia beccarii (Linne).—Cushman, 1922a:52, pl. 8: figs. 7-9; 1928:104, pl. 15: figs. 1-7; 1931:58, pl. 13: figs. 1, 2; 1944:35, pl. 4: fig. 22.—Said, 1949:37, pl. 4: fig. 5—Post, 1951:176.— Parker, 1952:457, pl. 5: fig. 5.—Parker, Phleger, and Peirson, 1953:13, pl. 4: figs. 20-22, 25-30. "Rotalia" beccarii (Linn£) variants.—Phleger, Parker, and Peirson, 1953:42, pl. 9: figs. 14, 15.—Parker, 1954:531, pl. 10: figs. 1,2,5,6. Ammonia beccarii (Linne).—Phleger, 1964:379, pl. 2: figs. 22- 25.—Buzas, 1965:62, pl. 4: fig. 1. Rosalina parkinsonia d'Orbigny, 1839:99, pl. 4: figs. 25-27. Rotalia beccarri (Linne), var. parkinsonia (d'Orbigny).— Phleger and Parker, 1951:23, pl. 12: fig. 6. Rotalia beccarii (Linne) var. tepida Cushman, 1926:79, pl. 1; 1931:61, pl. 13: fig. 3.—Phleger and Parker, 1951:23, pl. 12: fig. 7.—Post, 1951:176.—Parker, 1952:457, pl. 5: fig. 8. Rotalia cf. R. beccarii (Linn£) var. parkinsonia (d'Orbigny).— Cushman, Todd, and Post 1954:360, pl. 89: fig. 22. Rotalia beccarii (Linn£) var. sobrina Shupack, 1934:6, fig. 4.— Post, 1951:176— Parker, 1952:457, pl. 5: fig. 8. Living specimens of this widespread shallow- water species occur throughout these samples, often in large numbers. Only one of the present samples was found to be devoid of living specimens of Ammonia beccarii. The morphology of the test ap- pears to vary with the environment, so that in cer- tain samples so-called "varieties" dominate. None of the present specimens approach the large size ex- hibited by typical material from the sands of Rim- Ammonia jacksoni, new species PLATE 6: FIGURES 13-18 Test small, strongly convex on spiral side, slightly convex on umbilical side; wall finely perforate on spiral side, imperforate near umbilicus on umbilical side; periphery narrowly rounded, slightly lobate to entire; chambers inflated, arranged in two to three whorls, six to ten chambers in peripheral whorl, increasing gradually in size as added; all chambers visible on spiral side, only those of pe- ripheral whorl visible on umbilical side; sutures limbate, gently curved, slightly depressed on umbili- cal side, becoming open near umbilicus, nearly tangential, slightly depressed to flush on spiral side; umbilicus moderately large, filled by numerous closely spaced pillars that extend to proloculus; umbilical ends of chambers thickened, covered by papillae or granules, as are sutures near umbilicus, giving entire umbilical area granulate appearance; aperture interiomarginal slit (closed in some speci- mens) on umbilical side of final chamber. This small species is fairly common in many sam- ples from station 1, but only two specimens have been found in samples from station 3. It is morpho- logically quite distinct from Ammonia beccarii (Linn£), possessing more numerous chambers per whorl and more strongly tangential sutures on the spiral side, as well as a densely papillate umbilical area. Glabratellina sagrai (Todd and Bronnimann) is somewhat similar in appearance, but differs in having a subacute periphery, a flattened umbilical side, fewer chambers per whorl, and completely closed sutures on the umbilical side. This species is named in honor of J. B. C. Jack- son, who collected the samples and aided this study in many ways. NUMBER 31 95 Descriptive statistics of the maximum diameter, minimum diameter, and thickness measured in mm for the holotype and three paratypes are: standard mean deviation range maximum diameter 0.164 0.045 0.120-0.205 minimum diameter 0.144 0.037 0.105-0.180 thickness 0.104 0.036 0.070-0.140 Figured holotype: USNM 211164. Figured paratype: USNM 211162. Paratypes: USNM 211271, 211272. standard mean deviation range station 1 0.98 1.50 0-6 station 3 0.04 0.20 0-1 Family ELPHIDIIDAE Galloway, 1933 Subfamily ELPHIDIINAE Galloway, 1933 Genus Elphidium Montfort, 1808 Elphidium advenum (Cushman) Polystomella advena Cushman, 1922a:56-57, pl. 9: figs. 11, 12. Elphidium advenum (Cushman).—Todd and Bronnimann, 1957:39, pl. 6: figs. 5-7.—Parker, Phleger, and Peirson, 1953:7, pl. 3: fig. 11.—Bandy, 1956:193, pl. 30: fig. 18. Some of our specimens do not have a carina, but clearly are within the range of variation of the species. Specimens closely resembling the holotype and paratype are more inflated and have more lim- bate sutures and better development of septal bridges than the carinaless specimens. The latter resemble Elphidium rugulosum (Cushman and Wickenden), but are easily distinguished by their raised umbilicus. All specimens of Elphidium ad- venum examined (including the holotype and para- type) have an optically granular wall structure. For this reason, and because other characters, such as pores on the apertural face, are not consistent, we choose a conservative view of the genus Elphidium in opposition to the multigenera proposed by Loe- blich and Tappan (1964) and many other authors. Hypotype: USNM 211284. standard mean deviation range station 1 0.42 0.87 0-3 station 3 1.10 1.43 0-6 Elphidium excavatum (Terquem) Polystomella excavata Terquem, 1875:25, pl. 2: fig. 2. Elphidium incertun var. clavatum Cushman, 1930:18-19, pl. 7: figs. 8, 9. Elphidium clavatum Cushman.—Loeblich and Tappan, 1953:98, pl. 19: figs. 8-10.—Buzas, 1965:58, pl. 2: figs. 6, 7, pl. 3: figs. 1, 2; 1966:591, pl. 71: figs. 1-8.—Todd and Bronnimann, 1957:39, pl. 6: fig. 10. This widespread common northern species occurs rarely. As Lutze (1965) and Feyling-Hanssen (1972) have convincingly pointed out, the species called Elphidium clavatum Cushman by most American workers is a junior synonym of E. excavatum (Ter- quem). The individuals from Jamaica are smaller than E. excavatum from colder waters and do not have the brownish color that is so characteristic of this species. Nevertheless, the umbilical bosses, variable septal bridges, and sutural pattern are characteristic of E. excavatum. Todd and Bronni- mann found this species in the Gulf of Paria, but it has rarely been reported from die tropics. Hypotype: USNM 211287. standard mean deviation range station 1 0.21 0.65 0-3 station 3 0.02 0.14 0-1 Elphidium gunteri Cole Elphidium gunteri Cole, 1931:34, pl. 4: figs. 9, 10.—Parker, 1954:508, pl. 6: fig. 16.—Bandy, 1956:194, pl. 30: fig. 19 — Lankford, 1959:2098, pl. 2: fig. 7. This species occurs rarely at station 1. Specimens have fewer chambers (eight to ten) than the speci- mens deposited in the USNM collections, but in other respects are identical. Hypotype: USNM 211285. standard mean deviation range station 1 0.08 0.28 0-1 station 3 0.00 0.00 0 Elphidium kugleri (Cushman and Bronnimann) Cribroelphidium kugleri Cushman and Bronnimann, 1948:18- 19, pl. 4: fig. 4. The specimens in this study as well as two of Cushman and Bronnimann's paratypes are optically radial in wall structure. The specimens are very 96 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY station 1 station 3 211290, distinctly perforate and compare well with Cush- man and Bronnimann's specimens, although some of ours have as many as nine chambers instead of the six described by Cushman and Bronnimann. Some of the specimens appear brownish in color. All lack umblical bosses. Hypotype: USNM 211286. standard mean deviation range station 1 0.06 0.24 0-1 station 3 0.54 1.17 0-6 Eliphidium norvangi, new species PLATE 7: FIGURES 1-4 Test small, planispiral, involute; periphery rounded, slightly lobate; chambers seven to eleven in final whorl, initial and final chambers bearing many spike-shaped papillae; sutures distinct, slightly limbate; sutural pores a single row extend- ing to umbilical area; septal bridges short, distinct, on all sutures; wall calcareous, finely perforate, optically radial, transparent to translucent; umbili- cus slightly depressed, containing many spike-shaped papillae; aperture and lower half of apertural face bearing many spike-shaped papillae. This species is quite distinctive because of the numerous relatively large spike-shaped papillae. All specimens in our material have these papillae. No named species of Elphidium appears to resemble it. Moreover, examination of all primary types of 94 species in the USNM collection bears out that the species described here is unique. Small size prob- ably contributed to its obscurity. We have named this species in honor of the late Dr. Aksel Norvang of the Zoological Museum, Copenhagen, who made outstanding contributions to our knowledge of foraminifera. Descriptive statistics of the maximum diameter, minimum diameter, and thickness measured in mm for the holotype and six paratypes are: standard mean deviation range maximum diameter 0.126 0.023 0.090-0.160 minimum diameter 0.104 0.011 0.090-0.120 thickness 0.073 0.020 0.050-0.100 Figured holotype: USNM 211216. Figured paratype: USNM 211217. Paratypes: USNM 211288, 211289, 211291, 211292. standard mean deviation range station 1 1.46 2.66 0-14 station 3 0.13 0.33 0-1 Elphidium rugulosum (Cushman and Wickenden) PLATE 7: FIGURES 5, 6 Elphidium articulatum d'Orbigny var. rugulosum Cushman and Wickenden, 1929:7, pl. 3: fig. 8. Elphidium rugulosum (Cushman and Wickenden).—Bandy, 1956:194, pl. 30: fig. 21. Specimens closely resemble Cushman and Wick- enden's holotype and paratypes and Bandy's hypo- types. On most specimens septal bridges are poorly developed and characteristically, rugose material fills in the sutures and umbilical area. Individuals from our study and Cushman and Wickenden's paratypes were found to have a granular wall struc- ture. The type-locality of this species is Cumberland Bay, Juan Fernandez Islands, Chile. Bandy reported it from the Gulf of Mexico, but it is not reported by other authors. This species probably does have a wide distribution in the Caribbean because some of the specimens in the USNM collection that are identified as Elphidium matagordanum (Kornfeld) and E. mexicanum Kornfeld probably belong to this species. Figured hypotype: USNM 211215. range 0-9 mean 0.85 1.77 standard deviation 0.99 2.19 Elphidium spp. Specimens represented by hypotype USNM 239849 are very small juveniles and do not have the characters by which species can be distinguished. They may belong to either Elphidium advenum (Cushman) or E. rugulosum (Cushman and Wick- enden). Specimens represented by hypotype USNM 239850 are larger and "coarse" in appearance with numerous irregular septal bridges. They may repre- sent aberrant forms of E. excavatum (Terquem) or E. gunteri Cole. Hypotypes: USNM 239849, 239850. NUMBER 31 standard standard 97 mean deviation range mean deviation range station 1 0.25 0.44 0-1 station 1 0.06 0.32 0-2 station 3 0.08 0.35 0-2 station 3 0.00 0.00 0 Superfamily GLOBIGERINACEA Carpenter, Parker, and Jones, 1862 Family HANTKENINDAE Cushman, 1927 Subfamily HASTIGERININAE Bolli, Loeblich, and Tappan, 1957 Genus Globigerinella Cushman, 1927 Globigerinella aequilateralis (Brady) Globigerinella aequilateralis (H. B. Brady).—Phleger and Parker, 1951:35, pl. 19: fig. 14. Globigerinella aequilateralis (Brady).—Todd and Bronni- mann, 1957:40, pl. 12: fig. 7.—Cifelli and Smith, 1970:35, pl. 4: figs. 2-4. One specimen of this planktonic species was found at station 1. One must assume that some planktonic specimens washed ashore while still liv- ing and that we captured a few before they died. There is nothing morphologically peculiar about most to suggest any other history (such as the mor- phologic differences observed between "water- column" and "bottom" populations of the same planktonic species). Whether or not this genus is a junior synonym of Hastigerina Thomson, with this species the type, is beyond the scope of this study. Hypotype: USNM 211302. Family GLOBOROTALIIDAE Cushman, 1927 Subfamily GLOBOROTAUINAE Cushman, 1927 Genus Globorotalia Cushman, 1927 Globorotalia inflata (d'Orbigny) Globigerina inflata d'Orbigny.—Phleger and Parker, 1951:34, pl. 19: figs. 10, 11.—Cifelli and Smith, 1970:28, pl. 2: figs. 4, 5. Three living specimens of this common plank- tonic species were found at station 1. Hypotype: USNM 211306. Globorotalia menardii (d'Orbigny) Pulvinulina menardii (d'Orbigny).—Cushman, 1922a:50, pl. 8: figs. 3, 4. Globorotalia menardii (d'Orbigny).—Phleger and Parker, 1951:36, pl. 20: figs. 1, 2.—Bandy, 1956:194.—Cifelli and Smith, 1970:40. Four living and characteristic specimens of this planktonic species were found at station 1 in the Discovery Bay material. Hypotype: USNM 211307. standard mean deviation range station 1 0.08 0.28 0-1 station 3 0.00 0.00 0 Family GLOBIGERINIDAE Carpenter, Parker, and Jones, 1862 Subfamily GLOBIGERININAE Carpenter, Parker, and Jones, 1862 Genus Globigerina d'Orbigny, 1826 Globigerina quinqueloba Natland Globigerina cf. G. quinqueloba Natland.—Todd and Bron- nimann, 1957:40, pl. 12: figs. 2, 3. A single living specimen of this common plank- tonic species was found at station 1. It falls within the limits of the species, sensu stricto, though it could be placed in Globigerina quinqueloba Nat- land subspecies egelida Cifelli and Smith. A rela- tively much larger population would be needed to determine this more certainly; therefore, it seems best simply to leave it at the species level. Hypotype: USNM 211301. Globigerina rubescens Hofker, 1956 Globigerina rubescens Hofker, 1956:234, pl. 32: fig. 26, pl. 35: figs. 18-21. A single individual belonging to this species was observed at station 1. Hypotype: USNM 211308. 98 Genus Globigerinoides Cushman, 1927 Globigerinoides conglobatus (Brady) Globigerinoides conglobata (H. B. Brady).—Phleger and Parker, 1951:35, pl. 19: fig. 15.—Bandy, 1956:194. Globigerinoides conglobatus (Brady).—Cifelli and Smith, 1970:37, pl. 5: figs. 2-5. A single somewhat immature individual was found living at station 1. Hypotype: USNM 211304. Globigerinoides ruber (d'Orbigny) Globigerina rubra d'Orbigny.—Cushman, 1921:55, pl. 12, fig. 6; 1922a:36,pl. 14: figs. 1,2. Globigerinoides rubra (d'Orbigny).—Phleger and Parker, 1951:35, pl. 19: fig. 16.—Bandy, 1954:136, pl. 31: fig. 6; 1956:194.—Todd and Bronnimann, 1957:40, pl. 12: fig. 5. Globigernoides ruber (d'Orbigny).—Cifelli and Smith, 1970:38, pl. 5: fig. 6. Rare living individuals of this very common planktonic species are found at station 1 in the pres- ent material. Presumably they were collected shortly after being washed into very shallow water. They appear to be water-column species; some, however, have white rather than red tests. Hypotype: USNM 211305. standard mean deviation range station 1 0.10 0.37 0-2 station 3 0.00 0.00 0 Subfamily CATAPSYDRACINAE Bolli, Loeblich, and Tappan, 1957 Genus Globigerinita Bronnimann, 1951 Globigerinita glutinata (Egger) Globigerinita glutinata (Egger).—Cifelli and Smith, 1970:35, pl. 4: fig. 5. Two specimens were found living at station 1. The species is widespread and locally common in water-column plankton in the Caribbean. Hypotype: USNM 211303. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Superfamily ORBITOIDACEA Schwager, 1876 Family EPONIDIDAE Hofker, 1951 Genus Eponides de Montfort, 1808 Eponides} sp. PLATE 7: FIGURES 7-9 A single small specimen from station 3 is ques- tionably assigned to Eponides. It has a biconvex trochospiral test that is coarsely perforate on the spiral side, imperforate near the umblicus and on the apertural face. The chambers are arranged in about two whorls, with six chambers in the periph- eral whorl. The periphery is rounded, lobate. The sutures are limbate, depressed, sigmoid on the um- bilical side, tangential on the spiral side. The aper- ture is a wide interiomarginal extraumbilical- umbilical arch. This species differs from Loeblich and Tappan's (1964:C678) definition of Eponides in having a coarsely perforate spiral side and in having a rounded periphery. Because only a single specimen is present, no study of the internal structure has been attempted. Figured hypotype: USNM 211169. Genus Neoeponides Reiss, 1960 Neoeponides antillarum (d'Orbigny) Rotalina (Rotalina) antillarum d'Orbigny, 1839:75, pl. 5: figs. 4-6. Truncatulina antillarum (d'Orbigny).—Cushman, 1921:57, pl. 13: figs. 7, 8. Eponides antillarum (d'Orbigny).—Cushman, 1931:42, pl. 9: fig. 2.—Phleger and Parker, 1951:20, pl. 10: figs. 9, 10.— Bandy, 1954:136, pl. 30: fig. 8.—Parker, 1954:528, pl. 9: figs. 14, 15. Pulvinulina incerta Cushman, 1922a:51, pl. 9: figs. 1-3. This species is fairly common at station 3, but only one specimen has been found at station 1. These specimens seem to be smaller than average specimens from the USNM collections. Hypotype: USNM 211349. standard standard mean deviation range mean deviation range station 1 0.06 0.24 0-1 station 1 0.02 0.14 0-1 station 3 0.00 0.00 0 station 3 0.33 0.60 0-2 NUMBER 31 Family AMPHISTEGINIDAE Cushman, 1927 Genus Amphistegina d'Orbigny, 1826 Amphistegina gibbosa d'Orbigny PLATE 7: FIGURES 10-12 Amphistegina gibbosa d'Orbigny, 1839:120, pl. 8: figs. 1-3. Amphistegina lessonii d'Orbigny.—Cushman, 1931:79, pl. 16: figs. 1-3.—Phleger and Parker, 1951:26, pl. 13: figs. 13, 14, pl. 14: fig. 1.—Bandy, 1956:192. This is the shallow-water species of Amphistegina in the Caribbean area. Numerous specimens have been found in samples from station 3, but only occasional specimens appear in material from sta- tion 1. The taxonomic problem posed by this species is a particularly difficult one. A. gibbosa was named by d'Orbigny in 1839, with the comment that this spe- cies approaches only A. "quoyii" in its character- istics, differing principally in the shape of the chambers and in lacking medial partitions on the "rosette" side of the test. Among the species of Amphistegina listed in d'Orbigny's Tableau methodique de la classe des Cephalopodes (1826), A. quoyii was the first, indi- cated as living at Rawack, in Australia. No descrip- tion of the species was given, however, and no illustration was listed in the text. Nevertheless, in the captions to the illustrations, plate 17, figures 1-4 are listed as "A. quoyii." In the same publica- tion, d'Orbigny listed as the third species of Am- phistegina, A. lessonii from Mauritius in the Indian Ocean. For A. lessonii, he gave no description, but listed figures 1-4 of plate 17 as illustrations of his species, plus his plaster model no. 98. According to die plate captions, however, A. lessonii was not figured. Comparison of the illustrations on plate 17 with the model of A. lessonii suggests that two separate species of Amphistegina are represented; the illustrations are of a lenticular compressed spe- cies, the model is of a strongly biconvex, much less compressed species. The most logical conclusion to be drawn is that the reference to the illustrations in the text is incorrect, that the illustrations are of A. quoyii, and that this is the name that should be ap- plied to the Pacific species, which is similar, but not identical to the Caribbean species as noted by d'Orbigny (1839). The name A. lessonii, which has generally been applied to the current species, 99 should be used for the inflated forms from the Indian Ocean characterized by d'Orbigny's model. Figured hypotype: USNM 211202. standard mean deviation range station 1 0.17 0.48 0-2 station 3 11.92 13.44 0-64 Family CIBICIDIDAE Cushman, 1927 Subfamily CIBICIDINAE Cushman, 1927 Genus Cibicides de Montfort, 1808 Cibicides mayori (Cushman) Truncatulina mayori Cushman, 1924:39, pl. 12: figs. 3, 4. Cibicides mayori (Cushman).—Cushman, Todd, and Post, 1954:371, pl. 91: figs. 29, 30.—Todd, 1957, pl. 92: fig. 10; 1965:53, pl. 22: fig. 7. A few specimens that appear conspecific widi the holotype and paratypes of this species occur at sta- tion 3 in the present samples. These specimens are planoconvex, with inflated chambers and depressed sutures. The periphery is lobate, with a thin serrate keel. The interiomarginal slit aperture extends along the umbilical side from the slightly depressed umbilicus to the periphery. A supplementary su- tural aperture is present on the spiral side at the posterior corner of the final chamber, remaining open in previous chambers. Hypotype: USNM 211265. standard mean deviation range station 1 0.00 0.00 0-0 station 3 0.10 0.31 0-1 Cibicides pseudoungerianus (Cushman) PLATE 7: FIGURES 13-15 Truncatulina pseudoungeriana Cushman, 1922b:97, pl. 20: fig. 9. Cibicides pseudoungerianus (Cushman).—Bandy, 1956:193, pl. 31: fig. 8. The present suite of specimens has been com- pared with the holotype of "Truncatulina pseudo- ungeriana" Cushman, from the Oligocene of Mississippi, and appears conspecific. This small species has a plano-convex to slightly biconvex test, 100 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY with an entire to somewhat lobate periphery that bears a narrow, imperforate keel. There are eight to ten chambers in the peripheral whorl, with carved, limbate sutures that are depressed on the umbilical side, flush to slightly raised on the spiral side. Early chambers on the spiral side are obscured by a thickening of the wall over the previous whorls. The aperture extends from the umbilicus across the periphery and along the bases of the final few chambers on the spiral side. Most of the reported occurrences of Cibicides pseudoungerianus from Recent sediments are actu- ally of other species of Cibicides and other genera. Bandy's specimens, however, and some unfigured specimens in the USNM collections from shallow Atlantic water off the southeast United States ap- pear to belong to this species. Figured hypotype: USNM 211200. mean 1.98 1.92 range 0-11 0-17 station 1 station 3 standard deviation 2.61 2.84 Cibicides sp. Two small, probably immature specimens of Cibicides from station 3 are insufficiently diagnostic to allow specific identification. Hypotype: USNM 211266. standard mean deviation range station 1 0.00 0.00 0 station 3 0.04 0.20 0-1 Family PLANORBULINIDAE Schwager, 1877 Genus Planorbulina d'Orbigny, 1826 Planorbulina mediterranensis d'Orbigny Planorbulina mediterranensis d'Orbigny, 1826:280, pl. 14: figs. 4-6, 6 bis.—Cushman, 1922a:45, pl. 6: figs. 1, 2; 1931:129, pl. 24: figs. 5-8.—Bandy, 1954:137, pl. 31: fig. 3.—Parker, 1954:545, pl. 13: fig. 9. Two individuals from station 3 appear to be- long to this widely distributed species. These speci- mens may be distinguished from the more abundant Planorbulinella acervalis (Brady) by having cham- bers that are much wider than high, with broadly rounded peripheries, and that are not arranged in the alternating annular series characteristic of Planorbulinella. Hypotype: USNM 211353. standard mean deviation range station 1 0.00 0.00 0 station 3 0.04 0.20 0-1 Family ACERVULINIDAE Schultze, 1854 Genus Acervulina Schultze, 1854 Acervulina sp. PLATE 7: FIGURES 16-18 This species has an attached test, with a low trochospiral juvenile stage, later adding irregularly shaped inflated chambers in an apparently random pattern. The periphery is rounded to subacute. The wall is moderately perforate. The sutures are de- pressed. There is no apparent primary aperture, which distinguishes this form from broken or juve- nile specimens of Planorbulinella acervalis (Brady). The present specimens appear conspecific with specimens in the USNM collection labeled "Acer- vulina inhaerens Schultze." Included with the latter are intermixed specimens of Planorbulinella acer- valis (Brady), Planorbulina mediterranensis d'Or- bigny, and other rather nondescript encrusting forms. This species lacks, however, the coarse, funnel-shaped pores that appear to be the diagnostic feature of Acervulina inhaerens as figured and de- scribed by Schultze. Although numerous specimens of this species are to be found in the present samples, the various species appear to be so poorly defined that erection of yet another species of Acervulina at this time seems ill advised. Figured hypotype: USNM 211203. standard mean deviation range station 1 0.17 0.48 0-2 station 3 0.85 1.66 0-9 Genus Planorbulinella Cushman, 1927 Planorbulinella acervalis (Brady) PLATE 7: FIGURES 19-21 Planorbulina acervalis Brady, 1884:657, pl. 92: fig. 4.— NUMBER 31 101 Cushman, 1921:55, pl. 12: fig. 8; 1922a:45, pl. 6: fig. 3; 1931:130, pl. 25: fig. 1; 1941:13, pl. 2: figs. 7, 8. Planorbulina mediterranensis d'Orbigny.—Cushman, 1915:28, pl. 12: fig. 1; 1941:13, pl. 2: fig. 9. This variable attached species has a juvenile trochospiral stage that is difficult to distinguish from small specimens of Cibicides pseudoungarianus (Cushman), differing in having a subacute periph- ery, a lesser number of chambers per whorl, and in having slightly to strongly depressed sutures on the umbilical side. The umbilical side is strongly con- vex; the spiral side is planar to concave. Mature specimens have new chambers added in annular alternating series; the chambers are about as wide as they are high. Mature chambers do not reach to the umbilicus, as do those of die juvenile stage. Chambers are flattened to concave on the spiral side, and convex, somewhat inflated on the umbilical side. Marginal apertures with narrow lips are present at the base of the apertural face of the final chamber and at the suture with the preceding chamber, the posterior aperture remain- ing open in earlier chambers. The periphery is acute in the mature stage, generally with a thin, irregular, imperforate keel. Some large trochoid specimens from the present samples have been included in this species because they exhibit apertures similar in shape and place- ment to those of mature specimens of this species, although they do not develop the annular alter- nating chambers but retain the juvenile trochoid type of development. Figured hypotype: USNM 211188. station 1 station 3 Hypotype: USNM 211350, 211351. standard mean deviation range 1.35 1.56 0-8 5.08 3.99 0-18 Family CYMBALOPORIDAE Cushman, 1928 Genus Cymbaloporetta Cushman, 1928 Cymbaloporetta atlantica (Cushman) PLATE 7: FIGURES 22-24 Tretomphalus atlanticus Cushman, 1934:86, pl. 11: fig. 3, pl. 12: fig. 7. Juveniles of this species closely resemble small specimens of Rosalina globularis d'Orbigny, lack- ing only the well-developed umbilical chamber flaps of that species. Chambers are added in a low trochospiral coil until the one and one-half whorl stage, when the characteristic alternating annular chambers are added, generally four to a whorl. Some large individuals eventually add a single large float chamber to form a "typical" tretom- phaloid chamber arrangement. Because the type-species of Tretomphalus (T. bulloides) is conspecific with the type-species of Rosalina (R. globularis), the various species as- signed to Tretomphalus must be reassigned to odier genera (for more information, see Douglas and Sliter, 1965). Because this species has a charac- teristically cymbaloporettid stage in its ontogeny, Douglas and Sliter suggest placement in Cymbalo- poretta. This assignment seems reasonable and is followed in the present study. Figured hypotype: USNM 211199. Hypotype: USNM 211268. standard mean deviation range station 1 0.83 1.64 0-7 station 3 7.06 8.39 0-54 Cymbaloporetta squammosa (d'Orbigny) PLATE 8: FIGURES 1-3 Rosalina squammosa d'Orbigny, 1839:91, pl. 3: figs. 12-14. Cymbalopora squammosa (d'Orbigny).—Cushman, 1922a:41, pl. 6: figs. 4-6. Cymbaloporetta squammosa (d'Orbigny).—Cushman, 1931:83, pl. 16: fig. 4. Cymbalopora poeyi (d'Orbigny).—Cushman, 1931:83, pl. 16: fig. 4. This species has a juvenile trochospiral stage, later developing alternating annular chambers, gen- erally six in each whorl. The juvenile stage some- what resembles that of Rosalina globularis, differ- ing in lacking the chamber flaps and the depressed sutures on the umbilical side that are characteristic of that species, and in possessing a large umbilical plug and a much thicker test wall. This species has a much thicker and more opaque test than that of Cymbaloporetta atlantica, and has six annular chambers in the mature stage compared to four chambers of C. atlantica. Also, C. squammosa lacks the final tretomphaloid float chamber found in some large specimens of C. atlantica. 102 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY range 0-13 0-15 Figured hypotype: USNM 211198. Hypotype: USNM 211270. standard mean deviation range station 1 6.40 8.13 0-47 station 3 3.71 4.08 0-17 Cymbaloporetta tobagoensis (Bronnimann) PLATE 8: FIGURES 4-6 Cymbalopora tobagoensis Bronnimann, 1949:183, figs, a, b. Cymbaloporetta bradyi (Cushman).—Todd and Bronnimann, 1957:37, pl. 11: fig. 9. In this species, the annular mature chambers characteristic of Cymbaloporetta are added in a less tightly organized fashion than those of C. atlantica and C. squammosa, resulting in a low, flattened ir- regularly conical test. The wall is moderately to coarsely perforate on the spiral side, intermediate between the thickened, coarsely perforate wall of C. squammosa and the thin, finely perforate wall of C. atlantica. The juvenile stage differs from those of the other two species in possessing depressed su- tures on both spiral and umbilical sides, and in having pores on the umbilical side. The juvenile stage differs from Rosalina globularis in lacking chamber flaps, and in possessing a large umbilical plug. Figured hypotype: USNM 211197. Hypotype: USNM 211269. mean 2.83 5.71 standard deviation station 1 station 3 2.88 4.55 9: figs. 9, 10.—Parker, Phleger, and Peirson, 1953:5, pl. 4: figs. 14, 15—Parker, 1954:513, pl. 7: fig. 9.—Lankford, 1959, pl. 2: fig.. 17. Virgulina punctata d'Orbigny.—Cushman and Parker, 1931:15.—Cushman, 1941:9.—Todd and Bronnimann, 1957:33, pl. 8: fig. 8. Virgulina schreibersiana Czjzek.—Bandy, 1954:139, pl. 31, fig. 10. The holotype and a paratype have been exam- ined. Three Discovery Bay specimens were ex- amined for wall structure and found to be clearly optically granular. The species is fairly well repre- sented at station 3. Some of the abundant unfigured specimens referred to "Virgulina punctata" from the southeast coast of the United States and the Dry Tortugas (especially the former) in the USNM collections are Fursenkoina pontoni, though the majority are F. punctata. Of other USNM unfigured specimens referred to F. punctata, the following judgments are made: of the seven from Bermuda, both F. punctata and F. pontoni are represented; the one from Montego Bay, Jamaica is F. punctata; the one from Puerto Rico is F. pontoni', die two from off Tobago probably are F. pontoni; and the two from off Guadaloupe are F. pontoni. Though Fursenkoina pontoni was named from Miocene sediments, the above-cited Recent forms certainly appear conspecific. Perhaps the Miocene age of type F. pontoni has led to some of the con- fusion of Recent specimens with F. punctata, which was named by d'Orbigny from the Recent of Cuba; F. punctata is much more compressed and tends more toward biseriality than F. pontoni, and tends to be larger. Figured hypotype: USNM 211183. station 1 station 3 Superfamily CASSIDULINACEA d'Orbigny, 1839 Family CAUCASINIDAE Bykova, 1959 Subfamily FURSENKOININAE Loeblich and Tappan, 1961 Genus Fursenkoina Loeblich and Tappan, 1961 Fursenkoina pontoni (Cushma) PLATE 8: FIGURES 7, 8 Virgulina pontoni Cushman, 1932:80, pl. 12: figs. 10, 11; 1937:19, pl. 2: fig. 27.—Phleger and Parker, 1951:19, pl. standard mean deviation range 0.00 0.00 0 4.04 7.99 0-51 Fursenkoina punctata (d'Orbigny) Virgulina punctata d'Orbigny, 1839:139, pl. 1: figs. 35, 36.— Cushman, 1921:52, pl. 11: fig. 15; 1922a:31, pl. 3: fig. 9; 1937:23, pl. 3: fig. 27—Parker, 1954:513, pl. 7: fig. 11. Virgulina schreibersiana Czjzek.—Bandy, 1956:198. Only a single living specimen of this species was found at station 3 in the Discovery Bay material. Hypotype: USNM 211298. NUMBER 31 Genus Sigmavirgulina Loeblich and Tappan, 1957 Sigmavirgulina tortuosa (Brady) PLATE 8: FIGURES 9-12 Bolivina tortuosa H. B. Brady.—Cushman and Parker, 1931: 16, pl. 3: fig. 22.—Cushman, 1937:133, pl. 17: figs. 11-19; 1941:10.—Todd and Bronnimann, 1957:34, pl. 8: fig. 24. Todd and Bronnimann's figured specimen is weathered and superficially resembles the thin, twisted form of Bolivina subexcavata discussed herein; it is, however, Sigmavirgulina tortuosa. Of two specifically referred unfigured specimens col- lected by Bronnimann in the same study, one is S. tortuosa and the other is a thin, twisted Bolivina subexcavata. Of three unfigured specimens in the USNM collections from off Onslow Bay, North Carolina and referred to Sigmavirgulina tortuosa, two are S. tortuosa and one is a fairly characteristic Bolivina subexcavata. It has been difficult to determine the variability and exact morphology of this species in the Dis- covery Bay material as few specimens occur in any one sample and there is a disproportionate number of juveniles. This later situation is the case with many of die rarer species here and is believed to indicate that the species live more abundantly nearby in deeper water or otherwise different en- vironments and tend mainly to be washed in acci- dentally. The fact that the number of individuals here is greater, though still rare, at the deeper station (3) supports this. Comparison with large populations of Sigmavir- gulina tortuosa from other localities has assisted greatly in understanding the plan of growth and variability of these present specimens and of the species and genus themselves. The major taxonomic problem arises out of understanding the coiling in die early part of the test. The fact that Loeblich and Tappan (1964:C733) noted a very early axial twisting, leading them to conclude that the first two series of chambers were coiled in a sigmoiline fashion and that the examined specimens had a granular rather than radial wall led them to sep- arate this, the type-species from Bolivina, to which it had been assigned. It is also compared with Fur- senkoina, with which it is thought to be closely related. Specimens in the present study examined for wall structure showed Fursenkoina to be opti- cally granular and Bolivina optically radial. Like- 103 wise, numerous specimens of Sigmavirgulina tor- tuosa have been examined; they are not "beauti- fully" radial, but do show numerous calcite crosses. These examined specimens then perhaps should be referred to the genus Compresserigerina, radial, but they differ morphologically in having large pores and not tending to become unserial and have a neck, and they certainly appear to belong to the species tortuosa. The entire problem of radial ver- sus granular wall is in doubt. Leaving that aside, there appears to be sufficient morphologic distinc- tion to the species tortuosa to place it in a distinct genus, and if that is to be, Sigmavirgulina tortuosa is correctly the type-species. The higher taxonomic placement, dependent on assumed wall structure, as described by Loeblich and Tappan (1964) leaves many problems. As with the others encountered herein, we mainly follow the placement of Loeblich and Tappan fairly closely for convenience, in spite of this problem. Recognizing Sigmavirgulina tortuosa as distinct from Bolivina, leaving aside the problem of wall structure, brings up other problems. There is a con- siderable degree of variation in the amount of axial twisting at the base and the descriptive terms of this twisting must be understood. Specimens range from what would usually be called slightly twisted (perhaps 15°) basally, but clearly biserial through- out, to forms with such a basal twist (approaching 720°) that a problem arises as to how to "term" this twist. It could be called a sigmoiline coil or spire or it could be defined as a series of biserial chambers with a very rapidly rotating axis of coil- ing. Unfortunately, those specimens tending to- ward this last situation have a bulbous base with a thick wall (probably a result of maintaining the continuity of chambers being offset so rapidly) that makes seeing the exact positions and relations of the chambers difficult. The precise arrangement of early chambers showed by Loeblich and Tappan (1964, fig. 601: la) has not been seen in the present study. Their etched basal view (1964, fig. 601: 2) showing "twisted biserial early chamber arrange- ment" has been seen and better illustrates the prob- lem of description and understanding. From the exterior (as with la), only less twisted biserial ar- rangements have been clearly seen herein, the wall thickness being a deterrent with specimens twisted more than approximately 180°. The variation in twisting and concomitant wall 104 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY and test thickness, along with some other variations within the species, make juveniles difficult to recog- nize as belonging to the same species and initial dif- ficulty in recognizing the species as a single tax- onomic unit. All mature specimens show considerable flaring. They become more regularly biserial, though con- tinuing to twist, after the initial twisting. The surface is smooth and translucent except often in the early twisted part. There, ornamentation may consist of up to a few short costae in addition to the keel, which superficially may look like a costa at this point. There may also or instead be some very short spines that appear to be related to the pores and perhaps project from or around them— this occurring on large-pored specimens. In such cases, the pores may be seen as linear projections back into the thick wall. These lines are the pores and are located in the early part of the test, but not always in association with spines. Most specimens are only somewhat compressed in the adult chambers, but this varies somewhat and, varying, may be markedly thickest along the median and go to sharp and/or keeled margins. That is the more usual case; they may otherwise only become less thick toward a rounded margin. The majority of specimens are keeled, either throughout or become keeled in the last few to pair of chambers. Some specimens show a keel as an extension of the limbate sutures. The sutures often are not distinct or depressed in the early, twisted portion, but usually become so. They may be thick- ened or narrow, or go from thickened to narrow. The depression of the sutures often hardly indents the margin when it is sharp or keeled, even though depression may (or may not) be reasonably marked in the body of the test. The pores on most specimens are relatively few and large, but pore pattern is variable. Pore size may be small or medium throughout or decrease markedly in the last one or two chambers, some- times disappearing at the top of the rather bulbous final chambers of rather juvenile specimens. Pores may be relatively evenly distributed or may be con- centrated along the bases of chambers or become so in adult chambers, or may be or become concen- trated (as above) in rows parallel to the bases of chambers and, occasionally, also along the margins. A strange feature of some large pores is that in the early, twisted, and immediately following part of the test, the pores appear to be located as pits in centers of small, sharp pustules; these pustules may be slightly extended to be the spines mentioned above. Also, as mentioned, such pores, either in association with spines or not, are often linear- appearing from being seen extending through the thick but translucent wall. The aperture in adults usually extends as a slit from the base some distance upward toward the apex of the final chamber, on keeled (most) speci- mens surrounded by the keel, which forms a lip, but with a lip if not keeled. Occasionally, the aperture begins above the base of the final chamber and is more terminal. A tooth plate can be seen. In juveniles there is more apertural variability; the less twisted forms have the adult-type aperture, though a bit more oval, while the more bulbous and twisted forms have a more open, ovate slit located basally on the side of the depression formed by the join of two chambers. This depressed loca- tion is very seldom found in adults because of the rising and smooth overlap of chambers. With most specimens, there is little difficulty dis- tinguishing them from other species. A few small specimens, with little twist and pores along the bases of the chambers, resemble a form of Bolivina doniezi Cushman and Wickenden, but that species does not twist. In some cases, the compressed and twisted form here included in Bolivina subexcavata Cushman and Wickenden and clearly transitional with characteristic specimens of that species has either been taken for or included with Sigmavirgu- lina tortuosa. The resemblance is, however, super- ficial. Sigmavirgulina tortuosa twists earlier and usually more profoundly and has a smooth wall, usually with large pores, while the Bolivina has a grainy appearing wall, small pores, and a lesser but longer continuing twist. Figured hypotype: USNM 211176, 211177. Hypotype: USNM 211334, 211335, 211336, 211337, 211338. standard mean deviation range station 1 0.50 0.82 0-4 station 3 2.50 1.66 0-6 NUMBER 31 Genus Coryphostoma Loeblich and Tappan, 1962 Coryphostoma? limbata (Brady) subsp. costulata (Cushman) PLATE 8: FIGURES 13, 14 Bolivina limbata H. B. Brady var. costulata Cushman, 1922a: 26, pl. 3: fig. 8; 1937:187, pl. 21: fig. 30. Loxostomum mayori (Cushman).—Cushman and Parker, 1931:16, pl. 3: fig. 24.—Cushman, 1941:11 [part].—Todd and Bronnimann, 1957:35, pl. 8: fig. 38.—Bandy, 1956:295. [part, not p. 31: fig. 11]. The holotype of "Bolivina limbata H. B. Brady var. costulata Cushman" and the holotype and five paratypes of "Bolivina mayori Cushman" have been examined. (Both were described from off Florida by Cushman, 1929.) So have numerous other speci- mens of both taxa from essentially the "type" areas and odier areas. Examination strongly suggests that two taxa are represented, but that original descrip- tions have led to some confusion in differentiating these similar forms. Overall, "Loxostomum mayori" has a more delicate test, tends to have rounder mar- gins, and has from none to numerous striae or fine costae of varying length and position, as opposed to die few strong basal costae of "Loxostomum limba- tum costulatum." Bandy's figured specimen appears to belong to L. mayori, while, of the other two un- figured specimens deposited with it, one could be either species, while one clearly is "L. limbatum costulatum." Unfigured specimens included in Cush- man and Parker's material from Rio de Janeiro Harbor and the unfigured specimens treated by Cushman in 1941 (from Old Providence Island) ap- pear to belong to both taxa. "Loxostomum cf. mayori (Cushman)" of Phleger and Parker from the Gulf of Mexico (1951:17, pl. 8: fig. 5) appear to belong to L. mayori. Relatively rare, mainly juvenile specimens were found living in this study. More common, empty (dead) tests found probably indicate that the spe- cies lives nearby, adjusted to another environment. The present specimens are characterized by a few, strong costae over the large, usually bulbous, proloculus and sometimes first few following cham- bers. A few striations or finer costae are present on the later chambers of some larger specimens. Also characteristic are the sutures, which are very lim- bate toward the median line of the test, which, with the chamber overlap, forms a zigzag pattern 105 of clear subtriangles in the biserial part of the test. Early, the sutures are arcuate, though they may become straight and oblique later. Early, they are either flush or slightly depressed. In larger specimens they may become narrow and slightly depressed (if they were not depressed early) in the later part. Some larger tests are twisted. The mar- gins are usually rounded, but sometimes carinate in part, or occasionally all, of the test. The pores are of medium size, not very dense, and of fairly even distribution, except that in a few cases there are concentrations along suture edges. The pores are very distinct because the walls are so clear. The aperture is large—long with a fairly wide slit and a thick lip. There is an internal process that reaches the aperture to form a tooth plate in the final chamber of some specimens, but in others seems not to extend all the way up to the aperture nor all the way down from the apex to the base. The generic placement of this species is ques- tionable in some morphologic features, but, more basically, we question the limits and wall struc- tures of the genus Coryphostoma as given by Loeblich and Tappan (1964:C733). The first re- marks given with their original generic description (1962:111) allow for somewhat broader limits than does the description itself and the direct compari- sons with other genera. The significant remark states that because of the revision of Loxostomum, many species previously so referred were left name- less and the name Coryphostoma is proposed for them. This, then, leaves the present taxon only differing from characteristic Coryphostoma (except in wall structure) in not being rounded in cross section, but rather flattened, but usually with rounded margins (as opposed to Loxostomum's quadrate, double-keeled outline and sometimes den- ticulate aperture with no internal toothplate). Both Coryphostoma and Loxostomum are stated to have granular walls. Several specimens of the present taxon have been examined and found to have op- tically radial walls. The wall is not "beautifully" radial, but many calcite crosses can be found. We also found this with Sigmavirgulina tortuosa, and we feel that the original descriptions of the walls as granular, with no exceptions, are in error. We cannot say that this is so for all species of Sigma- virgulina and Coryphostoma, but it is for the two species here discussed (and S. tortuosa is the type- species of Sigmavirgulina). The present Loxosto- 106 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY mum-like specimens certainly do not belong to Loxostomoides or to Rectobolivina ("radial") on the basis of considerable morphologic differences. We are unable to resolve the taxonomic problem further than to raise these questions. We have, therefore, tentatively placed the form "Loxostomum limbatum costulatum" in Coryphostoma. It appears best placed there if one disregards the "radial- granular" wall question, which is subject, at best, to considerable interpretation. Figured hypotype: USNM 211213. standard mean deviation range station 1 0.31 0.62 0-2 station 3 1.00 1.30 0-5 Coryphostoma? sp. Two specimens, one from each station, are so referred. They closely resemble the type-species of the genus, Coryphostoma plaita (Carsey), except they are smaller and the apertures appear larger. The aperture also appears to be in a terminal pit, with a higher "back" side, reminiscent of some of the Pleurostomellidae. It may, however, simply be a large slit. No apertural teeth nor connecting tubes between chambers can be seen. The wall is optically radial. It is because of the problems of toothplate, aperture, and wall structure that the genus is questioned. The only pleurostommellid- like form it resembles at all closely is "Pleurosto- mella vicksburgensis Howe," which is similarly bi- serial, but that form differs in the aperture, which is not similarly sunken, in having larger pores, and in being larger. The material is too scanty for further determination. Hypotype: USNM 211347. Family CASSIDULINIDAE d'Orbigny, 1839 Genus Cassidulina d'Orbigny, 1826 Cassidulina cf. C. subglobosa Brady Cassidulina subglobosa Brady.—Phleger and Parker, 1951:27, pl. 14: fig. 13 [not figs. 11, 12].—PParker, 1954:536, pl. 11: fig. 6 [not figs. 4, 5, 7-9].—Lankford, 1959, pl. 3: fig. 15. Cassidulina sp. A. Todd and Bronnimann, 1957:37, pl. 11: fig. 10. Only 12 specimens have been obtained, from station 3, in the present material. Therefore, syn- onymy is particularly difficult in questionable cases. In the case of Phleger and Parker and that of Parker, the specimens questionably synonymized are very similar to our specimens, but not identi- cal, and belong to populations that they state in- clude large, characteristic specimens also (as those figures cited as not conspecific with the present specimens). This may also be the case with Lank- ford's material, but no comments are made and the figured specimen appears conspecific. Phleger and Parker's citation is as Cassidulina subglobosa in the text but as Cassidulina cf. subglobosa on the plate description and on the slide containing the speci- men. Parker's (1954) specimen is described as a "small compressed variant." Our specimens are too few to identify more closely, but they are not char- acteristic C. subglobosa, being much too compressed and small. They are also similar to Cassidulina barbara Buzas, but more globose, with less de- pressed sutures, a slightly different aperture, and are somewhat smaller. A new species may be repre- sented or perhaps all are juvenile or variant C. subglobosa. Among the many unfigured specimens in the USNM collections from off the southeast coast of the United States are many very similar specimens included with the larger and more characteristic C. subglobosa. The taxonomic prob- lem is beyond the scope of this report. The present specimens are very small, smooth, transparent to translucent, and optically granular in wall structure, with small pores. They show about six or seven chambers in the peripheral whorl (four complete viewed from the apertural side). They have slightly limbate, flush to very slightly depressed sutures. Specimens are subglobose in shape, with a rounded margin. The apertural face stands up parallel to the axial plane of the test, clearly showing the aperture. It is small, subtrian- gular, with a slight V-shaped projection of the wall into the aperture near the base on one side; though the wall is slightly thickened around the aperture, no internal processes have been seen. Hypotype: USNM 211341. standard mean deviation range station 1 0.00 0.00 0 station 3 0.23 0.47 0-2 in the umbilical area. standard Hypotype: USNM 211293. station 1 mean 0.08 deviation 0.28 range 0-1 standard station 3 0.02 0.14 0-1 mean deviation range station 1 1.33 1.37 0-6 station 3 1.98 2.42 0-9 Anomalina} 1 •en.i-fnrm.ix ("Ht* rrm-All«*n and V Mirlr>irtA\ NUMBER 31 Family NONIONIDAE Schultze, 1854 Subfamily NONIONINAE Schultze, 1854 Genus Nonionella Cushman, 1926 Nonionella auricula Heron-Allen and Earland Nonionella auricula Heron-Allen and Earland, 1930:192, pl. 5: figs. 68-70.—Todd and Bronnimann, 1957:32, pl. 5: fig. 32. Specimens from Jamaica most closely resemble individuals in the Cushman collection collected by P. Bronnimann from the Gulf of Paria, Trinidad. Some individuals resemble the paratypes of Nonio- nella atlantica Cushman, but the latter are generally more inflated and characteristically contain pustules 107 This rather small species occurs in several sam- ples, mostly from station 1. It has a low trochoid test, with about one and one-half whorls visible on the flattened spiral side of the largest individuals. The chambers are somewhat inflated on the um- bilical side. The periphery is broadly rounded. There are eight to nine chambers in the ultimate whorl. Sutures are limbate, flush to slightly de- pressed. The umbilicus is small but open, with the umbilical corner of each chamber thickened and imperforate. The aperture is an interiomarginal extraumbilical-umbilical slit extending to the pe- riphery. Figured hypotype: USNM 211171. Family ALABAMINIDAE Hofker, 1951 Genus Trichohyalus Loeblich and Tappan, 1953 Trichohyalus aguayoi (Bermudez) Discorbis aguayoi Bermudez, 1935:204, pl. 15: figs. 10-14.— Todd and Bronnimann, 1957:37, pl. 9: fig. 24. Discorinopsis aguayoi (Bermudez).—Parker, Phleger, and Peir- son, 1953:7, pl. 4: figs. 23, 24.—Phleger and Ewing, 1962: 178, pl. 5: figs. 5,6. This species occurs in samples from station 1 only. Hypotype: USNM 211365. standard mean deviation range station 1 0.17 0.48 0-2 station 3 0.00 0.00 0 Family ANOMALINIDAE Cushman, 1927 Subfamily ANOMALININAE, 1927 Genus Anomalina d'Orbigny, 1826 Anomalina glabrata Cushman PLATE 8: FIGURES 15-17 Anomalina glabrata Cushman, 1924:39, pl. 12: figs. 5-7.- Cushman, Todd, and Post, 1954:370, pl. 91: fig. 23. PLATE 8: FIGURES 18-20 Discorbina reniformis Heron-Allen and Earland, 1915:696, pl. 52: figs. 7-14. Anomalina} maculosa Todd, 1957:296, pl. 92: fig. 12. The generic assignment of this species is uncer- tain because it exhibits characteristics that preclude its inclusion in any described genus. It has a small, slightly trochospirally coiled test that is moderately to coarsely perforate except at the sutures and on the apertural face. The chambers are inflated and the sutures are flush, resulting in a globular test. The test wall is extremely thick, obscuring the de- tails of chamber arrangement. The periphery is rounded, entire. There are usually six to nine cham- bers in the peripheral whorl, early chambers in- creasing rapidly in size as added, later chambers in- creasing in size more slowly. The umbilicus is closed. The apertural face of the last chamber ap- pears to be covered by an imperforate callus, the bottom edge of which is free, resulting in a low slitlike interiomarginal aperture varying in length, ranging from the umbilicus to the periphery in some specimens to a very short peripheral slit in others. Because only a small number of specimens are present in these samples, no attempt at wall- 108 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY structure study has been attempted, and no place- ment in the Treatise (Loeblich and Tappan, 1964) classification is possible. For convenience, the ques- tionable assignment to Anomalina by Todd (1957) has been followed, because it does exhibit many characteristics of that poorly defined genus. The assignment to Discorbina by the original authors is less acceptable, because Discorbina has been placed in synonymy with Discorbis in the Treatise (Loeblich and Tappan, 1964), and the present speci- mens clearly do not exhibit the characteristics of that genus. Figured hypotype: USNM 211170. Hypotype: USNM 211262, 211366, 211367. standard mean deviation range station 1 0.08 0.28 0-1 station 3 0.02 0.14 0-1 Superfamily ROBERTINACEA Reuss, 1850 Family CERATOBULIMINIDAE Cushman, 1927 Subfamily CERATOBULIMININAE Cushman, 1927 Genus Lamarckina Berthelin, 1881 Lamarckina sp. Test small, planoconvex, auriculate, about two times higher than wide; wall finely perforate, trans- lucent; periphery acute with a thick, narrow keel, lobate; chambers inflated, arranged in a single whorl, eight chambers in largest specimen, cham- bers increasing rapidly in size as added; sutures limbate, slightly depressed on flat spiral side, deeply incised on convex umbilical side; umbilicus large, depressed, partially covered by large plate extend- ing from lower part of final chamber to prolocular area and first few chambers; aperture an arch be- neath the chamber flap on proximal side of the flap. Although the distinctive features of these speci- men ts appear unlike those of any previously re- ported species, an insufficient number is present to permit formal naming of a new species. Hypotype: USNM 211371. standard mean deviation range station 1 0.04 0.20 0-1 station 3 0.08 0.28 0-1 Family ROBERTINIDAE Reuss, 1850 Genus Cerobertina Finlay, 1939 Cerobertina sp. Two extremely small specimens of Cerobertina have been found at station 3. These specimens are compressed, concavo-convex, about twice as high as they are wide. The wall is translucent, finely per- forate. The periphery is narrowly rounded, entire in early development, becoming somewhat lobate. About seven to eight flattened chambers are visible from either side. The chambers increase rapidly in size as added. The sutures are radial, slightly curved, flush. The chamberlets on the umbilical side are very small and indistinct. The aperture is a low silt along the umbilical side of the depressed apertural face of the final chamber. These speci- mens are unlike other described species of this genus, but too small a sampling is present to allow erection in another species. Hypotype: USNM 211370. standard station 1 mean 0.00 deviation 0.00 range 0 station 3 0.04 0.20 0-1 Genus Ungulatella Cushman, 1931 Ungulatella} sp. Three very distinct specimens have been found at station 1. They appear to belong to Ungulatella (and may represent a new species). They differ from the generic description given by Loeblich and Tap- pan (1964:C782), however, in that they have a smooth, not rough, wall and they have more over- lap of chambers than Loeblich and Tappan de- scribe as seen on the side from which the aperture is visible. They are most similar to Ungulatella gracilis Seiglie from shallow water off Venezuela (which also is smooth) but show more chamber overlap and have a more rounded, rather than sharp, rim around the depression in the final cham- ber that contains the largest part of the aperture. NUMBER 31 109 In the present specimens also, the aperture is more Hypotype: USNM 211228. standard mean deviation range station 1 0.06 0.32 0-2 station 3 0.00 0.00 0 comma shaped than loop shaped and superficially looks like a circle. The material is insufficient to be more certain of the taxonomic designation. Appendix Number of Individuals Observed in 20 ml Replicate Samples TABLE A.—Station 1 NOV 1969 DEC JAN 1970 MAR SPECIES Acervalis sp Ammobaculites exilis Ammodiscus minimus Ammodiscus sp ; Ammonia beccarii Ammonia Jacksoni Amphistegina gibbosa Angulodiscorbis corrugata Anomalina glabrata Anomalina (?) reniformis Archaias angulatus Asterigerina carinata Asterigerina sp Baggina aff. B. philippinensis ... Bolivina doniezi Bolivina lowmani Bolivina paula Bolivina rhomboidalis Bolivina striatula . Bolivina subexcavata Bolivina spp Bolivina (?) spp , Bolivina cf. B. compacta , Bolivina cf. B. subexcavata , Bronnimannia caribaea Buliminella elegantissima Buliminella milletti Buliminella parallela . Cancris sagra Cassidulina cf. C. subglobosa .... Cerobertina sp , Chrysalidinella aff. C. miocenica Cibicides mayori Cibicides pseudoungerianus , Cibicides sp , Clavulina tricarinata Coryphostoma (?) limbata costulata Coryphostoma (?) sp Cyclogyra planorbis Cymbaloporetta atlantica Cymbaloporetta squammosa Cymbaloporetta tobagoensis Discorbinella minuta Dlscorbin ella sp Discorbis granulosa , Discorbis mira Discorbis murrayi Eggerella cf. E. advena Eggerella of. E. humbolti Elphidium advenum Elphidium excavatum Elphidium gunteri Elphidium kugleri Elphidium norvangi Elphidium rugulosum Elphidium spp Eoeponidella pacifica Epistomaroides (?) sp Epistominella (?) sp Eponidella sp Eponides (?) sp Esosyrinx (?) sp Fissurina cf. F. agassizi Fissurina cf. F. milletti Fissurina goreaui Fursenkoina pontoni Fursenkoina punctata 12 3 4 2 2 1 27 52 30 25 16 6 3 2 1 1 2 4 4 16 1 1 35 51 53 89 19 20 18 46 12 3 4 30 10 15 16 11 2 3 111 27 36 31 74 12 9 7 29 14 2 1 3 3 2 1 4 5 4 4 110 12 3 4 11 5 8 9 1 3 2 9 2 2 1 1 21 24 45 30 9 18 32 13 1 1 1 11 2 3 2 3 2 1 12 3 4 17 11 9 3 2 12 1 1 22 35 52 31 6 12 67 33 13 1 12 2 1 2 20 2 1 1 3 1 12 3 112 1 12 3 4 12 10 6 11 3 12 2 2 2 2 3 3 3 2 1 23 31 10 27 13 21 6 20 1 111 1 3 1 2 17 2 41 30 5 5 5 1 13 2 4 1 2 12 2 4 2 2 1 1 NUMBER 31 111 APR MAY JUL AUG SEP OCT 12 3 4 12 3 4 12 3 4 12 3 4 12 3 4 12 3 4 12 3 4 13 2 32 25 1 1 1 1 5 13 5 17 16 2 21 29 20 99 1 14 7 11 34 1 1 1 13 15 5 11 1 8 12 11 15 4 1 14 7 17 31 1 2 25 12 25 4 1 1 1 1 1 2 24 12 32 19 12 3 23 31 7 16 1 2 3 11 27 48 73 51 19 36 66 26 1 13 3 3 3 16 33 27 43 13 9 22 38 1 1 2 1 7 11 1 20 33 26 22 43 22 7 6 1 1 1 4 1 1 1 3 5 5 2 2 9 1 4 2 16 32 31 23 10 21 26 13 1 3 2 13 7 21 3 13 12 3 36 19 40 44 27 7 31 39 26 23 2 12 49 35 40 11 29 35 23 7 1 2 6 2 2 1 1 1 3 6 11 13 11 12 4 5 15 6 2 7 2 1 8 13 1 1 1 6 1 2 1 2 1 9 19 4 4 1 9 22 1 4 5 8 2 1 5 3 4 47 14 6 1 5 4 13 3 13 2 1 1 13 1 1 3 8 2 6 1 1 1 1 1 3 2 2 1 1 2 2 1 1116 1 1 112 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE A.—Continued NOV 1969 DEC JAN 1970 MAR SPECIES 12 3 4 12 3 4 12 3 4 9 24 Gaudryina exilis Gaudryina sp Gaudryina (?) sp Gaudryina cf. G. exilis .... Glabratella altispira Glabratella braziliensis ... Glabratella compressa Glabratella hexacamerata ... Glabratella testigoensis ... Glabratella sp. A Glabratella sp. B Glabratella sp. C Glabratella cf. G. carinata Glabratellina sagrai Glabratellina sp Globigerina quinqueloba .... Globigerina rubescens Globigerinella aequilateralis Globigerinita glutia&ta .... Globigerinoides conglobatus Globigerinoides ruber Globorotalia inflata Globorotalia menardii Glomospira glomerata Glomospira gordialis Glomospira sp Haplophragmoides sp Haplophra gmoides (?) sp. Helenina anderseni Karreriella (?) sp Lagenammina (?) sp Lamarckina sp Laryngosigma sp Lenticulina sp Lituotuba sp Mychostomina revertens Neoconorbina terquemi Neoeponides antillarum Nonionella auricula Patellina advena Patellinella sp Peneroplis bradyi Peneroplis pertusus Peneroplis proteus Planorbulina acervalis Planorbulina mediterranensis Rectobolivina raphana (?) ... Reophax nana Reophax sp Reussella atlantica Rosalina bulbosa Rosalina candeiana Rosalina concinna Rosalina floridana Rosalina globularis Rosalina subaraucana Rosalina sp Rosalina (?) sp Rotaliammina aff. R. mayori Sagrina pulehe11a Sagrina pulchella primativa Sigmavirgulina tortuosa Siphogenerina c ostata Siphonina pule ra Siphoninella (?) Spirillina vivipara Stetsonia minuta Textularia agglutinans , Textularia (?) sp , Tiphotrocha cf. T. comprimata Trichohyalus aguayoi , Trifarina occidentalis , Trochammina cf. T. advena ... Trochammina cf. T. quadriloba Trochammina sp Ungulatella (?) sp 3 3 1 1 1 2 4 8 3 1 1 3 20 10 17 12 2 4 3 14 12 1 1 1 111 3 4 3 3 17 8 3 2 14 2 2 1 3 3 1 1 2 1 1 1 19 10 23 10 1 2 1 12 21 4 13 2 1 1 112 1 2 14 4 9 31 13 1 1 3 4 8 1 1 3 14 13 5 22 NUMBER 31 113 APR MAY JUN JUL AUG SEP OCT 12 3 4 12 3 4 1 1 1 2 1 1 1 1 112 2 1 1 1 2 2 2 3 2 1 1 1 3 2 3 3 1 2 1 1 1 1 1 2 1 1 1 8 1 3 1 5 12 15 15 10 112 3 2 11 1 1 4 2 22 8 29 10 2 8 9 9 17 5 5 3 1 1 1 1 2 2 13 9 4 1 2 7 8 52 4 6 3 4 5 1 1 18 13 7 13 11 8 6 4 14 2 1 1 13 9 20 12. 37 24 30 1 10 11 14 27 1 11 16 31 2 12 14 13 19 33 2 1 1 1 18 24 25 4 114 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE B.—Station 3 NOV 1969 DEC JAN 1970 FEB MAR SPECIES 12 3 4 7 17 1 2 Acervalis sp Ammobaculites exilis Ammodiscus minimus Ammodiscus sp , Ammonia beccarii , Ammonia jacksoni Amphistegina gibbosa , Angulodiscorbis corrugata Anomalina glabrata , Anomalina (?) reniformis Archaias angulatus Asterigerina carinata , Asterigerina sp , Baggina aff. B. philippinensis Bolivina doniezi Bolivina lowmani Bolivina paula , Bolivina rhomboidalis , Bolivina striatula Bolivina subexcavata Bolivina spp Bolivina (?) spp , Bolivina cf. B. compacta Bolivina cf. B. subexcavata , Bronnimannia caribaea Buliminella elegantissima Buliminella milletti Buliminella parallela Cancris sagra Cassidulina cf. C. subglobosa .... Cerobertina sp Chrysalidinella aff. C. miocenica Cibicides mayori Cibicides pseudoungerianus Cibicides sp Clavulina tricarinata Coryphostoma (?) limbata costulata Coryphostoma (?) sp Cyclogyra planorbis Cymbaloporetta atlantica Cymbaloporetta squammosa Cymbaloporetta tobagoensis Discorbinella minuta Discorbinella sp Discorbis granulosa Discorbis mira Discorbis murrayi Eggerella cf. E. advena Eggerella cf. E. humboldti Elphidium advenum Elphidium excavatum Elphidium gunteri , Elphidium kugleri Elphidium norvangi Elphidium rugulosum , Elphidium spp Eoeponidella pacifica Epistomaroides (?) sp Epistominella (?) sp Eponidella sp Eponides (?) sp Esosyrinx (?) sp Fissurina cf. F. agassizi Fissurina cf. F. milletti Fissurina goreaui Fursenkoina pontoni Fursenkoina punctata 1 2 6 2 3 3 9 11 38 1 3 6 15 12 2 3 2 3 1 3 1 1 36 14 20 23 26 14 16 21 5 5 12 14 2 24 24 3 12 10 6 9 6 10 13 3 1 3 1 1 1 1 10 21 19 20 27 20 7 9 2 2 2 2 12 4 2 3 3 3 1 15 9 5 1 2 19 25 8 4 2 3 2 111 1 1 15 19 23 24 10 6 1 1 1 1 2 1 4 3 13 6 1 2 1 9 7 1 1 1 1 9 2 4 31 1 18 1 3 12 4 13 3 5 5 6 2 111 1 28 55 29 16 7 15 34 24 3 2 1 1 2 3 2 2 7 5 1 2 13 5 5 17 2 3 2 2 1 3 29 10 19 11 3 12 8 3 35 11 12 3 15 3 3 13 1 2 8 18 31 32 12 7 20 18 11 12 10 12 2 2 4 3 1 4 2 6 2 NUMBER 31 115 APR 12 3 4 12 3 4 12 3 4 2 3 2 14 21 28 7 1 2 8 139 25 4 14 5 3 16 14 158 5 1 28 45 46 16 3 5 46 7 8 12 34 11 17 17 8 9 17 20 9 4 4 3 5 112 1 12 20 30 13 4 12 1 0 7 2 21 11 10 12 64 2 1 4 11 1 5 13 2 6 6 5 1 1 5 5 2 6 1 13 1 3 1 20 28 44 6 16 12 16 8 2 10 16 13 4 3 13 2 4 39 38 48 34 21 32 33 39 2 2 12 4 2 13 26 61 8 21 19 57 5 26 4 2 11 1 1 25 38 32 28 18 24 18 14 12 1 2 3 2 2 1 1 73 42 21 45 54 12 7 10 1112 29 22 10 41 5 13 7 21 3 14 3 1 47 8 22 8 4 3 8 2 2 1 1 1 2 2 3 5 2 3 11 3 2 1 1 1 3 1 10 11 7 19 7 2 5 12 1 1 2 2 11 18 31 5 3 12 5 3 3 2 4 2 1 12 54 15 9 3 10 13 8 3 15 10 3 3 1 1 14 60 27 15 3 4 1 12 4 2 2 11 17 12 7 4 8 5 14 13 8 12 11 4 3 1 10 26 41 5 2 4 3 3 8 2 3 3 5 1 5 2 12 4 1 2 13 20 11 13 1 1 2 2 4 3 3 12 2 2 1 5 4 6 1 1 10 9 11 15 3 2 3 10 3 5 2 14 2 2 4 9 9 10 7 2 1 5 18 11 2 5 2 5 4 3 2 6 6 1 10 7 2 1 2 1 2 1 1 3 4 1 1 1 8 3 17 1 1 1 1 1 3 1 6 11 1 11 11 51 18 1 5 12 116 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE B.—Continued NOV 1969 DEC JAN 1970 FEB MAR SPECIES 12 3 4 12 3 4 12 3 4 12 3 4 Gaudryina exilis Gaudryina sp Gaudryina (?) sp , Gaudryina cf. G. exilis Glabratella altispira Glabratella braziliensis ..., Glabratella compressa Glabratella hexacamerata .... Glabratella testigoensis ..., Glabratella sp. A Glabratella sp. B Glabratella sp. C Glabratella cf. G. carinata Glabratellina sagrai Glabratellina sp Globigerina quinqueloba Globigerina rubescens , Globigerinella aequilateralis Globigerinita glutinata Globigerinoides conglobatus Globigerinoides ruber , Globorotalia inflata , Globorotalia menardii Glomospira glomerata Glomospira gordialis Glomospira sp Haplophragmoides sp Haplophra gmoides (?) sp. Helenina anderseni , Karreriella (?) sp , Lagenammina (?) sp , Lamarckina sp , Laryngosigma sp , Lenticulina sp , Lituotuba sp , Mychostomina revertens Neoconorbina terquemi , Neoeponides antillarum Nonionella auricula Patellina advena Patellinella sp , Peneroplis bradyi , Peneroplis pertusus Peneroplis proteus , Planorbulina acervalis , Planorbulina mediterranensis Rectobolivina raphana (?) Reophax nana Reophax sp Reussella atlantica , Rosalina bulbosa , Rosalina candeiana , Rosalina concinna , Rosalina floridana Rosalina globularis Rosalina subaraucana , Rosalina sp Rosalina (?) sp Rotaliammina aff. R. mayori , Sagrina pulchella Sagrina pulchella primativa Sigmavirgulina tortuosa Siphogenerina costata Siphonina pulcra Siphoninella (?) Spirillina vivipara Stetsonia minuta Textularia agglutinans Textularia (?) sp Tiphotrocha cf. T. comprimata Trichohyalus aguayoi , Trifarina occidentalis Trochammina cf. T. advena Trochammina cf. T. quadriloba Trochammina sp Ungulatella (?) sp 3 6 1 1 3 2 2 4 2 11 2 4 5 3 4 8 5 5 3 17 7 4 7 3 3 15 4 1 1 29 11 39 27 1 2 1 2 1 1 8 3 4 1 2 1 2 5 6 15 3 23 5 9 7 1 1 28 28 11 1 1 1 1 1 1 1 1 1 4 2 4 5 4 5 18 7 7 2 6 2 12 3 1 2 12 1 1 2 2 1 1 1 3 1 1 1 2 1 6 18 8 5 2 10 4 32 16 12 1 38 10 2 1 3 1 1 3 4 1 1 1 1 21 20 40 23 1 4 7 4 113 19 17 22 6 10 8 16 22 1 3 1 16 22 24 34 NUMBER 31 117 APR MAY JUN JUL AUG SEP OCT 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 1 3 1 1 1 1 3 2 2 1 1 1 2 1 1 5 3 2 1 1 1 3 2 3 1 1 1 1 1 1 1 1 1 1 1 1 3 9 1 4 1 1 1 1 1 1 1 1 1 2 1 2 2 2 3 4 2 2 4 1 1 1 5 2 3 8 1 1 2 1 9 1 1 1 5 1 1 1 1 1 1 4 4 1 1 3 7 5 1 3 2 2 2 5 2 1 10 6 5 3 4 3 4 10 6 4 1 2 2 2 1 4 4 1 2 1 2 1 3 3 2 3 2 2 2 1 2 4 1 5 5 1 4 1 5 3 7 2 3 8 3 1 2 4 2 5 4 2 10 1 2 3 4 9 15 3 1 4 7 2 3 1 1 10 8 1 7 1 8 1 3 2 7 4 6 5 2 1 3 5 2 1 2 1 4 1 1 1 1 4 1 1 1 1 1 1 2 7 1 7 7 1 1 3 2 3 6 6 0 1 1 1 6 2 2 5 11 5 8 8 6 2 3 4 6 10 10 4 6 4 1 7 3 17 3 3 3 31 7 7 4 45 34 51 13 46 44 26 83 19 19 31 9 21 9 16 10 7 10 6 20 5 2 2 7 12 4 10 9 5 36 31 18 20 13 8 20 1 5 15 30 12 19 3 5 1 12 19 7 9 9 2 5 5 8 2 1 1 3 2 4 3 1 2 1 6 2 1 1 1 1 1 2 2 1 3 2 1 1 3 1 3 6 3 2 1 1 1 1 3 1 1 1 1 5 1 5 2 3 4 2 3 5 2 4 2 3 3 4 5 2 6 2 1 2 2 2 3 2 1 1 1 1 2 1 1 1 3 1 1 1 4 1 1 1 1 1 1 1 1 1 3 1 1 1 !2 13 35 21 14 1 49 72 3 1 39 29 47 5 15 1 23 11 24 13 54 33 21 1 20 20 14 12 6 1 13 1 7 20 1 1 12 . Literature Cited Akers, W. H., and J. H. Dorman 1964. Pleistocene Foraminifera of the Gulf Coast. Tulane Studies in Geology, 3(1): 1-93, plates 1-15. Arrhenius, O. 1921. Species and Area. Journal of Ecology, 9:95-99. Bandy, O. L. 1954. Distribution of Some Shallow-Water Foraminifera in the Gulf of Mexico. United States Geological Survey Professional Paper, 254-F:123-141, plates 27-31. 1956. Ecology of Foraminifera in Northeastern Gulf of Mexico. United States Geological Survey Professional Paper, 274-G: 179-204, plates 29-31. Bermudez, P. J. 1935. Foraminiferos de la Costa Norte de Cuba. Memorias de la Sociedad Cubana de Historia Natural, 9(3):129- 224, plates 1-17. 1949. Tertiary Smaller Foraminifera of the Dominican Republic. Cushman Laboratory for Foraminferal Re- search, Special Publication, 25:1-322, plates 1-26. Bliss, C. L. 1958. Periodic Regression in Biology and Climatology. Connecticut Agricultural Experimental Station Bul- letin, 615:3-55. Bock, W. D. 1971. A Handbook of the Benthonic Foraminifera of Florida Bay and Adjacent Waters. Miami Geological Society Memoir, 1:1-72. Boltovskoy, E. 1959. Foraminiferos recientes del sur de Brasil y sus relaciones con los de Argentina e India del Oeste. Argentina Servicio Hidrografia Naval Publico, H(1005): 1-124, plates 1-20. Boltovskoy, £., and H. Lena 1969. Seasonal Occurrences, Standing Crop and Produc- tion in Benthic Foraminifera of Puerto Deseado. Contributions from the Cushman Foundation for Foraminiferal Research, 20:87-95. Bradshaw, J. S. 1955. Preliminary Laboratory Experiments on Ecology of Foraminiferal Populations. Micropaleontology, 1:351- 358. Brady, H. B. 1884. Report on the Foraminifera Collected by H. M. S. Challenger during the Years 1873-76. Part 22 in volume 9 in Zoology in Report . . . Challenger, xxl + 814 pages, 115 plates. London. Bronnimann, P. 1949. Pflanzenbewohnende Tropische Foraminiferen Nebst Beschreibung von Cymbalopora tobagoensis n. sp. Verhandlungen Basel Naturforschenden Gesellschaft, 60:179-185. Buzas, 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 (Foraminifer) in Long Island Sound through Canonical Analysis and Invariant Char- acters. Journal of Paleontology, 40:585-594, plates 71-72. 1968. On the Spatial Distribution of Foraminifera. Con- tributions from the Cushman Foundation for Fora- miniferal Research, 19:1-11. 1969. Foraminiferal Species Densities and Environmental Variables in an Estuary. Limnology and Ocean- ography. 14:411-422. Buzas, M. A., and T. G. Gibson 1969. Species Diversity: Benthonic Foraminifera in West- ern North Atlantic. Science, 163:72-75. Cifelli, R., and R. K. Smith 1970. Distribution of Planktonic Foraminifera in the Vicinity of the North Atlantic Current. Smithsonian Contributions to Paleobiology, 4:1-52, plates 1-6. Cochran, W. G. 1963. Sampling Techniques. 413 pages. New York: John Wiley & Sons, Inc. Cole, W. S. 1931. The Pliocene and Pleistocene Foraminifera of Flor- ida. Florida Geological Survey Bulletin, 6:1-79, plates 1-7. Cushman, J. A. 1915. A Monograph of the Foraminifera of the North Pacific Ocean, Part 5: Rotalidae. United States Na- tional Museum Bulletin, 71:1-87, plates 1-31. 1919. Fossil Foraminifera from the West Indies. Carnegie Institution of Washington Publication, 291:21-71, plates 1-15. 1921. Foraminifera from the North Coast of Jamaica. Proceedings of the United States National Museum, 59(2360):47-82, plates 11-19. 1922a. Shallow-Water Foraminifera of the Tortugas Re- gion. Carnegie Institution of Washington, 17(311):3- 85, plates 1-14. 1922b. The Byram Calcareous Marl of Mississippi and Its Foraminifera. United States Geological Survey Pro- fessional Paper, 129E:79-122. 1922c. The Foraminifera of the Mint Spring Calcareous Marl Member of the Marianna Limestone. United States Geological Survey Professional Paper, 129F: 123-152, plates 29-35. 1922d. The Foraminifera of the Atlantic Ocean, Part 3: Textulariidae. United States National Museum Bul- letin, 104:1-149, plates 1-26. 1924. Samoan Foraminifera. The Carnegie Institution of 118 NUMBER 31 119 Washington Publication, 342:1-75, plates 1-25. 1925. Recent Foraminifera from British Columbia. Con- tributions from the Cushman Laboratory for Fora- miniferal Research, l(2):38-47, plates 6, 7. 1926. Recent Foraminifera from Puerto Rico. The Car- negie Institution of Washington Publications, 344: 73-84, plate 1. 1928. On Rotalia beccarii (Linne). Contributions from the Cushman Laboratory for Foraminiferal Research, 4(4): 103-107, plate 15. 1930. The Foraminifera of the Atlantic Ocean, Part 7: Nonionidae, Camerinidae, Peneroplidae, and Alveo- linellidae. United States National Museum Bulletin, 104:1-79, plates 1-18. 1931. The Foraminifera of the Atlantic Ocean, Part 8: Rotaliidae, Amphisteginidae, Calcarinidae, Cymbal- oporettidae, Globorotaliidac. Anomalinidae, Planor- bulinidae, Rupertiidae, and Homotremidae. United States National Museum Bulletin, 104:1-179, plates 1-26. 1932. Notes on the Genus Virgulina. Contributions from the Cushman Laboratory for Foraminiferal Research, 8:7-20, plates 2-3. 1933. Some New Recent Foraminifera from the Tropical Pacific. Contributions from the Cushman Laboratory for Foraminiferal Research, 9:77-95, plates 8-10. 1934. Notes on the Genus Tretomphalus, with Descrip- tions of Some New Species of a New Genus, Pyro- pilus. Contributions from the Cushman Laboratory for Foraminiferal Research, 10(4):79-101, plates 11- 13. 1937. A Monograph of the Subfamily Virgulininae of the Foraminiferal Family Buliminidae. Cushman Lab- oratory for Foraminiferal Research, Special Publi- cation, 9:1-228, plates 1-24. 1941. Recent Foraminifera from Old Providence Island Collected on the Presidential Cruise of 1938. Smith- sonian Miscellaneous Collections, 99(9):1—14. plates 1,2. 1943. Tretomphalus myersi, a New Species from the Pacific. Contributions from the Cushman Laboratory for Foraminiferal Research, 19(2):26-27, plate 6. 1944. Foraminfera from the Shallow Water of the New England Coast. Cushman Laboratory for Foraminif- eral Research, Special Publication, 12:1-37, plates 1-4. 1947. New Species and Varieties of Foraminifera from off the Southeastern Coast of the United States. Con- tributions from the Cushman Laboratory for Foram- iniferal Research, 23(4):86-92, plates 19-20. Cushman, J. A., and P. Bronnimann 1948a. Some New Genera and Species of Foraminifera from Brackish Water of Trinidad. Contributions from the Cushman Laboratory for Foraminiferal Research, 24(1):15-21, plates 3-4. 1948b. The Additional New Species of Arenaceous Foram- inifera from Shallow Water of Trinidad. Contribu- tions from the Cushman Laboratory for Foraminif- eral Research, 24(2):37-42, plates 7-8. Cushman, J. A., and E. D. Cahill 1933. Miocene Foraminifera of the Coastal Plain of the Eastern United States. United States Geological Sur- vey Professional Paper, 175-A:l-50, plates 1-13. Cushman, J. A., and B. Kellett 1929. Recent Foraminifera from the West Coast of South America. Proceedings of the United States National Museum, 75(2796):1-16, plates 1-5. Cushman, J. A., and W. McGlamery 1938. Oligocene Foraminifera from Choctaw Bluff, Ala- bama. United States Geological Survey Professional Paper, 189-D:101-119, plates 24-28. Cushman, J. A., and F. L. Parker 1931. Recent Foraminifera from the Atlantic Coast of South America. Proceedings of the United States National Museum, 80(3): 1-24, plates 1-4. Cushman, J. A., and G. M. Ponton 1932. The Foraminifera of the Upper, Middle, and Part of the Lower Miocene of Florida. Florida State Geo- logical Survey Bulletin, 9:1-147, plates 1-17. Cushman, J. A., and R. Todd 1942. The Genus Cancris and Its Species. Contributions from the Cushman Laboratory for Foraminiferal Research, 18(4):72-94, plates 17-24. Cushman, J. A., R. Todd, and R. J. Post 1954. Recent Foraminifera of the Marshall Islands, in Bikini and Nearby Atolls, Part 2: Oceanography (Biologic). United States Geological Survey Profes- sional Paper, 260-H:319-384, plates 82-93. Cushman, J. A., and R. T. D. Wickenden 1929. Recent Foraminifera from off Juan Fernandez Is- lands. Proceedings of the United States National Museum, 75(2780): 1-16, plates 1-6. Dayton, P. K., and R. R. Hessler 1972. Role of Biological Disturbance in Maintaining Diversity in the Deep Sea. Deep-Sea Research, 19:199-208. Dennison, J. M., and W. H. Hay 1967. Estimating the Needed Sampling Area for Sub- aquatic Ecologic Studies. Journal of Paleontology, 41:706-708. Ehrenburg, C. G. 1843. Verbreitung und Einfluss des mikroskopischen Lebens in Slid und Nord-Amerika. . . . Abhand- lungen der koenigliche Akademie der Wissenschaften Berlin, 1841[1843]:291-446, plates 1-4. Fenchel, T. 1970. Studies of the Decomposition of Organic Detritus Derived from the Turtle Grass Thalassia testudinum. Limnology and Oceanography, 15:14-20. Fichtel, L. V., and J. P. C. V. Moll 1798 [1803]. Testacea microscopica aliaque minuta ex gen- eribus Argonauta et Nautilus ad naturam delineata et descripta: Milroskopische und Andere Kleine Schalthiere aus den Geschlectern Argonaute und Schiffer, nach der Natur gezeichnet und Beschrieben, Wien, Camesina. 124 pages, plates 1-24. [1798 edition unavailable. 1803 edition listed in Catalogue . . . British Museum as "a duplicate . . . with a new title page and slight variations."] Finn, J. D. 120 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 1974. A General Model for Multivariate Analysis. 423 pages. New York: Holt, Rinehart and Winston, Inc. Fisher, R. A., A. S. Corbet, and C. B. Williams 1943. The Relation Between the Number of Species and the Number of Individuals in Random Samples of an Animal Population. Journal of Animal Ecology, 12:42-58. Forskal, P. 1775. Descriptiones Animalium, Avium, Amphibiorum. . . . 164 pages. Copenhagen. Gibson, T. G., and M. A. Buzas 1973. Species Diversity: Patterns in Modern and Miocene Foraminifera of the Eastern Margin of North Amer- ica. Geological Society of America Bulletin, 84:217- 238. Gleason, H. A. 1922. On the Relations between Species and Area. Ecology, 3:156-162. Grassle, J. F., and H. L. Sanders 1973. Life Histories and the Role of Disturbance. Deep- Sea Research, 20:643-659. Greig-Smith, M. A. 1964. Quantitative Plant Ecology. 256 pages. London: But- terworths. Heron-Allen, E„ and A. Earland 1915. The Foraminifera of the Kerimba Archipelago (Por- tuguese East Africa), Part II. Transactions of the Zoological Society of London, 20(17):543-794, plates 40-53. 1930. The Foraminifera of the Plymouth District, II. Journal of the Royal Microscopical Society, 50:161- 199, plates 4-5. Hill, M. O. 1973. Diversity and Evenness: A Unifying Notation and Its Consequences. Ecology, 54:427-432. Hofker, J. 1956. Foraminifera Dentata, Foraminifera of Santa Cruz and Thatch-Island Virginia-Archipelago West Indies. University Zoological Museum, Kobenhaxm, 15:1- 237, plates 1-35. Hoglund, H. 1947. Foraminifera in the Gullmar Fjord and the Skage- rak. Zoologiska Bidrag Fran Uppsala, 26:1-328, plates 1-32. Hurlbert, S. H. 1971. The Nonconcept of Species Diversity: A Critique and Alternative Parameters. Ecology, 52:577-586. Jackson, J. B. C. 1972. The Ecology of the Molluscs of Thalassia Com- munities, Jamaica West Indies, II: Molluscan Popu- lation Variability along an Environmental Stress Gradient. Marine Biology, 14:304-337. Jones, T. R., and W. K. Parker 1860. On the Rhizopodal Fauna of the Mediterranean, Compared with That of the Italian and Some Other Tertiary Deposits. Quarterly Journal of the Geo- logical Society of London, 16:292-307. Lankford, R. R 1959. Distribution and Ecology of Foraminifera from East Mississippi Delta Margin. Bulletin of the American Association of Petroleum Geologists, 43(9):2068-2099, plates 1-3. Lankford, R. R., and F. B Phleger 1973. Foraminifera from the Nearshore Turbulent Zone, Western North America. Journal of Foraminiferal Research, 3:101-132. Lee. J. J. 1974. Towards Understanding the Niche of the Foraminif- era. Pages 207-260 in volume 1 in R. H. Hedley and C. G. Adams, editors, Foraminifera. New York: Academic Press Inc. Lee, J. J., M. McEnery, S. Pierre, H. P. Freudenthul, and W. A. Miller 1966. Tracer Experiments in Feeding Littoral Foraminif- era. Journal of Protozoology, 4:659-670. Lee, J. J., and W. A. Muller 1973. Trophic Dynamics and Niches of Salt Marsh Foram- inifera. American Zoology, 13:215-223. Linne\ K. 1758. Regnum Animale. Volume 1 of Systema Naturae. Edition 10, 824 pages. Holmiae [Stockholm]: L. Salvii. Lipps, J. H., and J. W. Valentine 1970. The Role of Foraminifera in the Trophic Structure of Marine Communities. Lethaia, 3:279-286. Loeblich, A. L. Jr., and H. Tappan 1953. Studies of Arctic Foraminifera. Smithsonian Miscel- laneous Collections, 121:1-150, plates 1-24. 1962. Six New Generic Names of the Mycetozoida (Trichi- idae) and Foraminiferida (Fischerinidae, Bulimini- dae, Caucasinidae, and Pleurostomellidae), and a Re- description of Loxostomum (Loxostomidae, new family). Biological Society of Washington Proceed- ings, 75:107-113. 1964. Treatise on Invertebrate Paleontology, part c: Protista 2, Sarcodina Chiefly "Thecameobians" and Foraminiferida. Volumes 1-2, 900 pages. Lawrence: University of Kansas Press for Geological Society of America. MacArthur, R. H. 1972. Geographical Ecology: Patterns in the Distribution of Species. 270 pages. New York: Harper and Row. MacArthur, R. H., and J. W. MacArthur 1961. On Bird Species Diversity. Ecology, 42(3):594-598. Matera, N. J., and J. J. Lee 1972. Environmental Factors Affecting the Standing Crop of Foraminifera in Sublittoral and Psammolittoral Communities of a Long Island Salt Marsh. Marine Biology, 14:89-103. Millett, F. W. 1899. Report of the Recent Foraminifera of the Malay Archipelago Collected by Mr. A. Durrand. Journal of the Royal Microscopical Society, 6:557-564, plate 7. 1903. Report on the Recent Foraminifera of the Malay Archipelago Collected by Mr. A. Durrand. Journal of the Royal Microscopical Society, 15:685-704, plate 7. Muller, W. A. 1975. Competition for Food and Other Niche-Related NUMBER 31 121 Studies of Three Species of Salt-Marsh Foraminifera. Marine Biology, 31:339-351. Murray, J. W. 1973. Distribution and Ecology of Living Benthic Foram- iniferids. 274 pages. New York: Crane, Russak and Company, Inc. Orbigny, A. d' 1826. Tableau Mdthodique de la Classe des C^phalopodes. Annals des Sciences Naturelles, 7:245-314, plates 10-17. 1839. Foraminiferes. Part 2 of Natural History in de la Sagra, Historie physique, politique et naturelle de Vile de Cuba. 224 pages [piates published separately]. Paris: Bertrand. Parker, F. L. 1952. Foraminiferal Distribution in the Long Island Sound-Buzzard Bay Area. Bulletin of the Museum of Comparative Zoology, 106(10):428-473, plates 1-5. 1954. Distribution of the Foraminifera in the Northeastern Gulf of Mexico. Bulletin of the Museum of Com- parative Zoology, lll(10):453-588, plates 1-13. 1958. Eastern Mediterranean Foraminifera. Reports of the Swedish Deep-Sea Expedition, 8(4):217-283, plates 1-6. Parker, F. L., and W. Athearn 1959. Ecology of Marsh Foraminifera in Poponesset Bay, Massachusetts. Journal of Paleontology, 33(2):333- 343. Parker, F. L., F. B Phleger, and J. F. Peirson 1953. Ecology of Foraminifera from San Antonio Bay and Environs, Southwest Texas. Cushman Foundation for Foraminiferal Research, Special Publication, 2:1- 75, plates 1-4. Phleger, F. B 1964. Patterns of Living Benthonic Foraminifera, Gulf of California. American Association of Petroleum Geol- ogists Bulletin, 3:377-394, plates 1-3. Phleger, F. B, and G. C. Ewing 1962. Sedimentology and Oceanography of Coastal Lagoons in Baja California, Mexico. Geological Society of America Bulletin, 73:145-181, plates 1-6. Phleger, F. B, and F. L. Parker 1951. Ecology of Foraminifera, Northwest Gulf of Mexico, Part II: Foraminifera Species. Geological Society of America Memoir 46:1-64, plates 1-20. Phleger, F. B, F. L. Parker, and J. F. Pierson 1953. North Atlantic Foraminifera. Reports of the Swedish Deep-Sea Expedition 7(1): 1-122, plates 1-12. Pielou, E. C. 1966. The Measurement of Diversity in Different Types of Biological Collections. Journal of Theroretial Biol- ogy, 13:131-144. Post, R. J. 1951. Foraminifera of the South Texas Coast. Publications of the Institute of Marine Science, 2(1): 165-176, plate 1. Preston, F. W. 1948. The Commonness, and Rarity, of Species. Ecology, 26:254-283. 1962. The Canonical Distribution of Commonness and Rarity. Ecology, 43(2-3): 185-215, 410-432. Rzehak, A. 1888. Die Foraminiferen der Nummuliten-schichten des Waschberges und Michelsberges bei Stockerau in Nieder-Oesterreich. Austria Geologische Reichsans- talt Verhandlungen, pages 226-229. Said, R. 1949. Foraminifera of the Northern Red Sea. Cushman Laboratory for Foraminiferal Research, Special Pub- lication, 26:1-44, plates 1-4. Sanders, H. L. 1969. Benthic Marine Diversity and the Stability-time Hypothesis. In Diversity and Stability in Ecological Systems. Brookhaven Symposia in Biology, 22:71-81. Saunders, J. B. 1957. Trochamminidae and Certain Lituolidae (Foramini- fera) from the Recent Brackish-Water Sediments of Trinidad, British West Indies. Smithsonian Miscel- laneous Collections, 134:1-16, plates 1-4. Schultze, M. S. 1854. Ueber den Organismus der Polythalamien (Fora- miniferen) nebst Bemerkungen uber die Rhizopoden im allgemeinen. 68 pages, 7 plates. Leipzig: Engel- mann. Seal, H. L. 1964. Multivariate Statistical Analysis for Biologists 207 pages. London: Methuen and Company, Ltd. Seiglie, G. A., and P. J. Bermudez 1965. Monografia de la Familia de Foraminiferos Gla- bratellidae. Geoscience, 12:15-65, plates 1-14. Sheldon, L. 1969. Equitability Indices: Dependence on the Species Count. Ecology, 50:467. Shupack, B. 1934. Some Foraminifera from Western Long Island and New York Harbor. American Museum Novitates, 737:1-12, plate 1. Sliter, W. V. 1970. Inner-Neritic Bolivinitidae from the Eastern Pacific Margin. Micropaleontology, 16:155-174, plates 1-8. Slobodkin, L. B., and H. L. Sanders 1969. On the Contribution of Environmental Predictabil- ity to Species Diversity. In Diversity and Stability in Ecological Systems. Brookhaven Symposia in Biology, 22:82-95. Smith, R. K., and L. B. Isham 1974. Reinstatement of Mychostomina Berthelin, 1881, and Emendation of Spirillina Ehrenberg, 1843, Spirillin- inae, Spirillinidae, and Spirillinacea, all Reuss, 1862. Journal of Foraminiferal Research, 4:61-68. Taylor, L. R. 1961. Aggregation, Variance and the Mean. Nature, 189: 732-735. Terquem, M. O. 1875. Essai sur le classement des animaux qui vinent sur la plage et dans les environs de Dunkerque. 153 pages. Paris. Todd, R. 1957. Geology of Saipan, Mariana Islands, Part 3: Pale- 122 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY ontology (Chapter H, Smaller Foraminifera). United States Geological Survey Professional Paper, 280-H: 265-320, plates 64-93, tables 1-4. 1965. The Foraminifera of the Tropical Pacific Collections of the "Albatross," 1899-1900—Rotaliform Families and Planktonic Families. United States National Museum Bulletin, 161(4): 1-139, plates 1-28. Todd, R., and P. Bronnimann 1957. Recent Foraminifera and Thecamoebina from the Eastern Gulf of Paria. Cushman Foundation for Foraminiferal Research, Special Publication, 3:3-43, plates 1-12. Todd, R., and D. Low 1971. Foraminifera from the Bahama Bank West of An- dros Island. United States Geological Survey Profes- sional Paper, 683-C:l-22. Uchio, T. 1960. Ecology of Living Benthonic Foraminifera from the San Diego, California, Area. Cushman Foundation for Foraminiferal Research, Special Publication, 5:1- 72, plates 1-10. Ujiie, H. 1962. Introduction to Statistical Zonation. The Journal of the Geological Society for Japan, 68:431-450. Warren, A. D. 1957. Foraminifera of the Buras-Scofield Bayou Region, Southeast Louisiana. Contributions from the Cush- man Foundation for Foraminiferal Research 8(1):29- 40, plates 3-4. Williams, C. B. 1964. Patterns in the Balance of Nature. 234 pages. New York: Academic Press. PLATES 124 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 1 Lituotuba sp., hypotype, USNM 211212, X 140: 1, Side view; 2, side view. Haplophragmoides} sp., hypotype, USNM 211211, X 160: 3, Side view; 4, edge view. Ammobaculites exilis (Cushman and Bronnimann), hypotype, USNM 211205, X 100: 5, Side view; 6, edge view. Gaudryina exilis (Cushman and Bronnimann), hypotype, USNM 211207, X 140: 7, Apertural view; 8, side view. Gaudryina cf. G. exilis (Cushman and Bronnimann), hypotype, USNM 211208, X 140: 9, Apertural view; 10, side view. Eggerella cf. E. humboldti (Todd and Bronnimann), hypotype, USNM 211206, X 300: 11, Side view; 12, edge view. Clavulina tricarinata (d'Orbigny), hypotype, USNM 211204, X 160: 13, Apertural view; 14, side view. Fissurina goreaui, new species, paratype, USNM 211209, X 200: 15, Apertural view; 16, side view. Fissurina goreaui, new species, holotype, USNM 211210, X 200: 17, Apertural view; 18, side view. Buliminella elegantissima (d'Orbigny), hypotype, USNM 211182; X 160: 19, Apertural view; 20, side view. Buliminella milletti (Cushman), hypotype, USNM 211181; X 160: 21, Apertural view; 22, side view. Buliminella parallela (Cushman and Parker), hypotype, USNM 211180, X 160: 23, Apertural view; 24, side view. Bolivina doniezi (Cushman and Wickenden), hypotype, USNM 211158, X 200: 25, Apertural view, 26, side view. Bolivina doniezi (Cushman and Wickenden), hypotype, USNM 211157, X 200: 27, Apertural view; 28, side view. NUMBER 31 125 126 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 2 Bolivina paula (Cushman and Cahill), hypotype, USNM 211156, X 160: 1, Apertural view; 2, side view. Bolivina rhomboidalis (Millett), hypotype, USNM 211155, X 140: 3, Apertural view; 4, side view. Bolivina striatula, (Cushman), hypotype, USNM 211154, X 160: 5, Apertural view; 6, side view. Bolivina striatula (Cushman), hypotype, USNM 211153, X 160: 7, Apertural view; 8, side view. Bolivina striatula (Cushman), hypotype, USNM 211152, X 160: 9, Apertural view; 10, side view. Bolivina subexcavata (Cushman and Wickenden), hypotype, USNM 211146, X 200: 11, Apertural view; 12, side view. Bolivina subexcavata (Cushman and Wickenden), hypotype, USNM 211148, X 200: 13, Apertural view; 14, side view. Bolivina subexcavata (Cushman and Wickenden), hypotype, USNM 211147, X 200: 15, Apertural view; 16, side view. Bolivina subexcavata (Cushman and Wickenden), hypotype, USNM 211150, X 200: 17, Apertural view; 18, side view. Bolivina subexcavata (Cushman and Wickenden), hypotype, USNM 211149, X 200: 19, Apertural view; 20, side view. Bolivina subexcavata (Cushman and Wickenden), hypotype, USNM 211151, X 200: 21, Apertural view; 22, side view. Bolivina cf. B. subexcavata (Cushman and Wickenden), hypotype, USNM 211144, X 200: 23, Apertural view; 24, side view. Bolivina cf. B. Subexcavata (Cushman and Wickenden), hypotype, USNM 211145, X 200: 25, Apertural view; 26, side view. NUMBER 31 127 128 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 3 Rectobolivina raphana (Parker and Jones), hypotype, USNM 211214, X 50: 1, Apertural view; 2, side view. Sagrina pulchella d'Orbigny, hypotype, USNM 211179, X 160: 3, Apertural view; 4, side view. Sagrina pulchella d'Orbigny, hypotype, USNM 211178, X 160: 5, Apertural view; 6, side view. Trifarina occidentalis (Cushman), hypotype, USNM 211174, X 160: 7, Apertural view; 8, side view. Trifarina occidentalis (Cushman), hypotype, USNM 211175, X 160: 9, Apertural view; 10, side view. Discorbis granulosa (Heron-Allen and Earland), hypotype, USNM 211196, X 80: 11, Side view; 12, edge view; 13, side view. Discorbis mira (Cushman), hypotype, USNM 211195, X 80: 14, Side view; 15, edge view; 16, side view. Discorbis murrayi (Heron-Allen and Earland), hypotype, USNM 211290, X 160: 17, Side view; 18, edge view; 19, side view. Discorbinella minuta, new species, holotype, USNM 211163, X 200: 20, Side view; 21, edge view; 22, side view. Discorbinella minuta, new species, paratype, USNM 211167, X 200: 23, Side view; 24, edge view; 25, side view. NUMBER 31 129 130 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 4 Rosalina candeiana (d'Orbigny), hypotype, USNM 211186, X 90: 1, Side view; 2 edge view; 3, side view. Rosalina concinna (Brady), hypotype, USNM 211187, X 100: 4, Side view; 5, edge view; 6, side view. Rosalina floridana (d'Orbigny), hypotype, USNM 211185, X 100: 7, Side view; 8, edge view; 9, side view. Rosalina globularis (d'Orbigny), hypotype, USNM 211184, X 100: 10, Side view; 11, edge view; 12, side view. Rosalina subaraucana (Cushman), hypotype, USNM 211173, X 140: 13, Side view; 14, edge view; 15, side view. Baggina aff. B. philippinensis (Cushman), hypotype, USNM 211172, X 140: 16, Side view; 17, edge view; 18, side view. Glabratella altispira, new species, paratype, USNM 211160, X 300: 19, Side view; 20, edge view; 21, side view. Glabratella altispira, new species, holotype, USNM 211165, X 300: 22, Side view; 23, edge view; 24, side view. NUMBER 31 131 132 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 5 Glabratella cf. G. carinata (Seiglie and Bermudez), hypotype, USNM 211193, X 300: 1, Side view; 2, edge view; 3, side view. Glabratella compressa, new species, paratype, USNM 211166, X 230: 4, Side view; 5, edge view; 6, side view. Glabratella hexacamerata (Seiglie and Bermudez), hypotype, USNM 211194, X 300: 7, Side view; 8, edge view; 9, side view. Glabratella compressa, new species, holotype, USNM 211161, X 230: 10, Side view; 11, edge view; 12, side view. Glabratella spp., hypotype, USNM 211189, X 300: 13, Side view; 14, edge view; 15, side view. Glabratella spp., hypotype, USNM 211168, X 300: 16, Side view; 17, edge view; 18, side view. Glabratella spp., hypotype, USNM 211191, X 300: 19, Side view; 20, edge view; 21, side view. Glabratellina sp., hypotype, USNM 211159, X 200: 22, Side view; 23, edge view; 24, side view. Glabratellina sagrai (Todd and Bronnimann), hypotype, USNM 211192, X 200: 25, Side view; 26, edge view; 27, side view. NUMBER 31 133 2 ^ 1 mm * 4 5 6 ■*'' 7 8 9 10 ii ]2 #* 17 13 14 16 * # 18 ^ ™ jpf 19 20 2 i ^^ 2 23 2 25 j Jf J^ 4f#l IK# JgjSv; | 26 > 27 24 134 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 6 Angulodiscorbis corrugata (Millett), hypotype, USNM 211201, X 200: 1, Side view; 2, edge view; 3, side view. Spirilina vivipara (Ehrenberg), hypotype, USNM 187958, X 160: 4, Side view, 5, edge view; 6, side view. Mychostomina revertens (Rhumbler), hypotype, USNM 187956, X 300: 7, Side view; 8, edge view; 9, side view. Mychostomina revertens (Rhumbler), hypotype, 187957, X 300: 10, Side view; 11, edge view; 12, side view. Ammonia jacksoni, new species, paratype, USNM 211162, X 200: 13, Side view; 14, edge view; 15, side view. Ammonia jacksoni, new species, holotype, USNM 211164, X 200: 16, Side view; 17, edge view; 18, side view. NUMBER 31 135 136 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 7 Elphidium norvangi, new species, holotype, USNM 211216, X 200: 1, Apertural view; 2, side view. Elphidium norvangi, new species, paratype, USNM 211217, X200: 3, Apertural view; 4, side view. Elphidium rugulosum (Cushman and Wickenden), hypotype, USNM 211215, X 200: 5, Side view; 6, edge view. Eponides sp., hypotype, USNM 211169, X 160: 7, Side view; 8, edge view; 9, side view. Amphistegina gibbosa (d'Orbigny), hypotype, USNM 211202, X 80: 10, Side view; 11, edge view; 12, side view. Cibicides pseudoungerianus (Cushman), hypotype, USNM 211200, X 100: 13, Side view; 14, edge view; 15, side view. Acervulina sp., hypotype, USNM 211203, X 100: 16, Side view; 17, edge view; 18, side view. Planorbulinella acervalis (Brady), hypotype, USNM 211188, X 80: 19, Side view; 20, edge view; 21, side view. Cymbaloporetta atlantica (Cushman), hypotype, USNM 211199, X 200: 22, Side view; 23, edge view; 24, side view. NUMBER 31 137 138 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY PLATE 8 Cymbaloporetta squammosa (d'Orbigny), hypotype, USNM 211198, X 80: 1, Side view; 2, edge view; 3, side view. Cymbaloporetta tobagoensis (Bronnimann), hypotype, USNM 211197, X 140: 4, Side View; 5, edge view; 6, side view. Fursenkoina pontoni (Cushman), hypotype, USNM 211183, X 80: 7, Apertural view; 8, side view. Sigmavirgulina tortuosa (Brady), hypotype, USNM 211177, X 160: 9, Apertural view; 10, side view. Sigmavirgulina tortuosa (Brady), hypotype, USNM 211176, X 160: 11, Apertural view; 12, side view. Coryphostoma? limbata (Brady) subspecies costulata (Cushman) hypotype, USNM 211213, X 100: 13, Apertural view; 14, side view. Anomalina glabrata (Cushman) hypotype, USNM 211171, X 140: 15, Side view; 16, edge view; 17, side view. Anomalina} reniformis (Heron-Allen and Earland), hypotype, USNM 211170, X 140: 18, Side view; 19, edge view; 20, side view. NUMBER 31 139 REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION Manuscripts intended for series publication receive substantive review within their originating Smithsonian museums or offices and are submitted to the Smithsonian Institution Press with approval of the appropriate museum authority on Form SI-36. Requests for special treatment—use of color, foldouts, casebound covers, etc.—require, on the same form, the added approval of designated committees or museum directors. Review of manuscripts and art by the Press for requirements of series format and style, completeness and clarity of copy, and arrangement of all material, as outlined below, will govern, within the judgment of the Press, acceptance or rejection of the manuscripts and art. Copy must be typewritten, double-spaced, on one side of standard white bond paper, with lVi" margins, submitted as ribbon copy (not carbon or xerox), in loose sheets (not stapled or bound), and accompanied by original art. Minimum acceptable length is 30 pages. Front matter (preceding the text) should include: title page with only title and author and no other information, abstract page with author/title/series/etc, following the establish- ed format, table of contents with indents reflecting the heads and structure of the paper. First page of text should carry the title and author at the top of the page and an unnum- bered footnote at the bottom consisting of author's name and professional mailing address. Center heads of whatever level should be typed with initial caps of major words, with extra space above and below the head, but with no other preparation (such as all caps or underline). Run-in paragraph heads should use period/dashes or colons as necessary. Tabulations within text (lists of data, often in parallel columns) can be typed on the text page where they occur, but they should not contain rules or formal, numbered table heads. Formal tables (numbered, with table heads, boxheads, stubs, rules) should be sub- mitted as camera copy, but the author must contact the series section of the Press for edito- rial attention and preparation assistance before final typing of this matter. Taxonomic keys in natural history papers should use the alined-couplet form in the zoology and paleobiology series and the multi-level indent form in the botany series. If cross-referencing is required between key and text, do not include page references within the key, but number the keyed-out taxa with their corresponding heads in the text. Synonymy in the zoology and paleobiology series must use the short form (taxon, author, year:page), with a full reference at the end of the paper under "Literature Cited." For the botany series, the long form (taxon, author, abbreviated journal or book title, volume, page, year, with no reference in the "Literature Cited") is optional. Footnotes, when few in number, whether annotative or bibliographic, should be typed at the bottom of the text page on which the reference occurs. Extensive notes must appear at the end of the text in a notes section. If bibliographic footnotes are required, use the short form (author/brief title/page) with the full reference in the bibliography. Text-reference system (author/year/page within the text, with the full reference in a "Literature Cited" at the end of the text) must be used in place of bibliographic footnotes in all scientific series and is strongly recommended in the history and technology series: "(Jones, 1910:122)" or ". . . Jones (1910:122)." Bibliography, depending upon use, is termed "References," "Selected References," or "Literature Cited." Spell out book, journal, and article titles, using initial caps in all major words. For capitalization of titles in foreign languages, follow the national practice of each language. Underline (for italics) book and journal titles. Use the colon-parentheses system for volume/number/page citations: "10(2):5-9." For alinement and arrangement of elements, follow the format of the series for which the manuscript is intended. Legends for illustrations must not be attached to the art nor included within the text but must be submitted at the end of the manuscript—with as many legends typed, double- spaced, to a page as convenient. Illustrations must not be included within the manuscript but must be submitted sepa- rately as original art (not copies). All illustrations (photographs, line drawings, maps, etc.) can be intermixed throughout the printed text. They should be termed Figures and should be numbered consecutively. If several "figures" are treated as components of a single larger figure, they should be designated by lowercase italic letters (underlined in copy) on the illus- tration, in the legend, and in text references: "Figure 9b." If illustrations are intended to be printed separately on coated stock following the text, they should be termed Plates and any components should be lettered as in figures: "Plate 9b." Keys to any symbols within an illustration should appear on the art and not in the legend. A few points of style: (1) Do not use periods after such abbreviations as "mm, ft, yds, USNM, NNE, AM, BC." (2) Use hyphens in spelled-out fractions: "two-thirds." (3) Spell out numbers "one" through "nine" in expository text, but use numerals in all other cases if possible. (4) Use the metric system of measurement, where possible, instead of the English system. (5) Use the decimal system, where possible, in place of fractions. (6) Use day/month/year sequence for dates: "9 April 1976." (7) For months in tabular list- ings or data sections, use three-letter abbreviations with no periods: "Jan, Mar, Jun," etc. Arrange and paginate sequentially EVERY sheet of manuscript—including ALL front matter and ALL legends, etc., at the back of the text—in the following order: (1) title page, (2) abstract, (3) table of contents, (4) foreword and/or preface, (5) text, (6) appendixes, (7) notes, (8) glossary, (9) bibliography, (10) index, (11) legends. M :■■ ■