S M I T H S O N I A N C O N T R I B U T I O N S T O B O T A N Y N U M B E R 9
Genera of Bamboos
Native to the New World
(Gramineae: Bambusoideae)
F. A. McClure
(EDITED BY THOMAS R. SODERSTROM)
SMITHSONIAN INSTITUTION PRESS
City of Washington
1978
ABSTRACT
McClure, F. A. (edited by Thomas R. Soderstrom). Genera of Bamboos Native
to the New World (Gramineae: Bambusoideae). Smithsonian Contributions to
Botany, number 9, 148 pages, 49 figures, 1973.-Seventeen genera of bamboos
native to the New World are classified, compared, and illustrated, including
four new genera and four new species. The introduction of several taxonomic
characters based on hitherto neglected morphological features, both vegetative
and reproductive, is undertaken with the objective of improving traditional
perspectives. Two reforms are urged as essential to the elevation of the level
of refinement at which future taxonomic treatments of bamboos may be executed.
These are (1) the general adoption of improved collecting methods
correlated with more extensive and sustained field observations, and (2) the
fostering, through interdisciplinary collaboration, of progressive development and
integration of diversified studies of documented materials drawn from a common
source for each individual taxon.
OFFICIAL PURLICATION DATE is handstamped in a limited number of initial copies and is re.
corded in the Institution’s annual report, SmBthsonian Year. SI PRESS NUMBER 4780. SERIE~ COVER
DESIGN: Leaf clearing from the katsura tree Cercidiphyllum japonicum Siebold and Zuccarini.
Library of Congress Cataloging in Publication Data
McClure, Floyd Alonzo, 1897-1970.
Genera of Bamboos native to the new world (Gramineae: Bambusoideae)
(Smithsonian contributions to botany no. 9.)
Bibliography: p.
1. Bamboo-America. 1. Title. 11. Series: Smithsonian Institution. Smithsonian contributions
QKI ~ 2 7 4 7 no. 9 [QK495.G74] 581’.08s [584’.93] 72-8954
to botany no. 9.
For sale by the Superintendent of Documents, US. Government Printing Office, Washington, D.C. 20402
Price $2.60 domestic postpaid, or $2.25 GPO Bookstore
h
On 15 April 1970, Dr. Floyd A. McClure passed away in his bamboo garden
while removing a plant to give to a young friend. McClure had devoted his entire
career to the bamboos, which he described as a “symbol of uprightness, chivalry,
and devotion.” His first book, The Bamboos: A Fresh Perspective, and the present
work represent as rich a heritage as any man could hope to leave to the botanical
world. Of all his numerous publications, these two books represent the synthesis of
a lifetime study by one of the most productive minds in the botanical sciences, a
mind which, to our good fortune, had matured to its full potential.
McClure, not only as a botanist but as a man, inspired and encouraged all
those with whom he came in contact. My first acquaintance with him was here
at the Smithsonian Institution in August 1961, since which time I had cherished
his friendship and advice. In 1967, he persuaded me to extend my interest in the
anatomy of the monocotyledons to include the vascular architecture of the bamboo
culm. Utilizing the optical shuttle system and the data analyzer projector,
this study confirmed the growing consensus that the structure of the bamboo
culm, like that of maize, does not represent a typical monocotyledon stem as had
been previously accepted.
A little over a year before McClure’s death, Dr. Thomas R. Soderstrom,
Smithsonian agrostologist, joined him in his bamboo studies; thus we are assured
that McClure’s life work will be carried on. Daily sessions between the two
continued to the very day of his death. Soderstrom, following McClure’s wishes,
has undertaken the difficult and demanding task of seeing to fruition this
present and final work of a great man and botanist.
McClure’s original intention had been to revise the bamboos for Die
Natiirlichen Pflanzenjamilien; but when he realized that this would be too large
a task, he decided to publish first the genera of the New World as represented
in this book. Although the bulk of the manuscript was in near-readiness at the
time of his death, significant portions of the work remained undone. It thus
fell to Soderstrom to conduct an intensive program of organizing the material.
Every effort was made to complete the book just as McClure would have,
drawing upon his copious notes whenever possible to finish the uncompleted
portions. Fortunately, Mrs. McClure, who had worked closely with her husband
throughout his career and was completely knowledgable of the details of
McClure’s manuscript-preparation methods, was able to provide much needed
help. Mr. Elmer Smith, a botanical illustrator at Harvard University, returned
to the Smithsonian on two occasions for extended periods to complete all the
figures. He had worked closely with McClure over the years, and the two had
developed a personal rapport that transcended the usual scientist-artist relationship.
I want to congratulate Dr. Soderstrom for accepting the commitment to
complete this monumental work, which, in its final form, I am sure would have
pleased Dr. McClure. I also wish to stress that he has had to put aside his own
research for extended periods to carry out this task.
iii
I would also like to thank the National Science Foundation for its continued
support of this work after Dr. McClure’s death and Dr. Eric Holttum, who
painstakingly reviewed the manuscript before submission to the Smithsonian
Institution Press. Mr. H. K. Airy Shaw kindly reviewed the Latin descriptions,
making a number of useful suggestions, and Dr. Alicia Lourteig compared parts
of the text that appear in French with the original herbarium notes in Paris. Dr.
Cleofk E. Calder6n gave assistance and the benefit of her expertise in the final
stages of manuscript preparation.
McClure was always concerned with the precise usage of words, and he
developed over the years a glossary of bamboo terms which first appeared in his
earlier book. Since publication of that glossary he modified some of the terms
and wrote definitions for several new ones. Because of the practicality of having
all terms appear in one reference, the decision was made to include in this
volume the entire glossary, even though some of these terms are not used in
connection with bamboos of the New World.
The archival materials and bamboo library of Dr. McClure are now
incorporated into the Hitchcock-Chase Grass library in the Department of
Botany, Smithsonian Institution. These materials will continue to be curated,
updated, and utilized. Without the extensive holdings of bamboo specimens,
literature, card files, and cross-references that Dr. McClure had accumulated over
a lifetime, a book of this magnitude could never have been completed.
EDWARD S. AYENSU
Chairman
Department of Botany
Smithsonian Znstitution
Washington, D.C.
21 Sepember 1971
iv
Foreword
My first acquaintance with Dr. McClure came through his articles on plants
cultivated at Canton, which I read in Singapore, where I was also making
myself acquainted with plants grown by Chinese people. We first met in Java
in 1929, about which time I began to try to take some interest in bamboos,
provoked by the fact that nearly all the bamboo plants in the Botanic Gardens
at Singapore had lost their labels, and the consideration that bamboos were
plants of importance which should not be neglected. This led me to examine
native Malayan bamboos when I had the opportunity of botanical travel. In this
rather casual study I was helped and stimulated by McClure’s successive papers
on his Chinese bamboos, from which I learned much. Our paths met again
after World War 11, in Singapore, London, and Washington, where I had the
privilege of his hospitality on two occasions. I have enjoyed reading the text
of this book, which brings many happy memories into my conscious thought,
and am honored to be asked to write this Foreword, in which I wish to try to
assess the significance of this, his last contribution to recorded knowledge of
the plants to which he devoted so much of his life.
Most early descriptions of bamboos were based on flowering specimens by
herbarium botanists who had never seen the living plants. Inevitably such
descriptions omitted vegetative characters without which field botanists and
cultivators could not name their plants, many of which flower only at long
intervals. Inevitably, also, the herbarium botanists, familiar with the plants of
north temperate regions, compared bamboo spikelets and flowers with those of
the grasses with which they were familiar, and tried to fit bamboos into the
grass pattern, not realizing that the reverse process leads to a better understanding
of the whole.
Effective vegetative description, and its correlation with floral description,
of the great bamboos of Southeast Asia was begun by Kurz, and was continued
by Gamble in his comprehensive monograph of Indian bamboos (1896). Gamble,
however, did not extend his thinking on the comparative morphology of the
inflorescence beyond that of Munro and Bentham, and little further critical
thought had been given to the subject when in 1924 McClure began his studies
of the bamboos of southern China (related to those of India but in many
cases distinct). Between 1924 and 1940 he established a plantation of six hundred
bamboo plants, collected during numerous travels, so that he could watch their
growth and flowering, In so doing he studied vegetative branching at culm
nodes (which Japanese botanists had found important in describing their own
bamboos) and also made detailed observations on the development of inflorescences.
For the first time he correlated vegetative branching with that of the
inflorescence, and established clearly a basic distinction between the two types
of inflorescence, referred to in the present book as iterauctant (indeterminate)
and semelauctant (determinate). Iterauctant branching matches exactly the
vegetative branching at culm nodes and its precise description involves a more
careful observation of the sheathing organs at the bases of spikelet-like structures
V
than had previously been undertaken. This is one of the keys to the understanding
of bamboo classification. Japanese botanists also studied the branching
of bamboo rhizomes, a subject of great practical importance to growers; McClure
did the same for the bamboos of China, concerning which no such records had
been made.
Thus when McClure went to tropical America in 1942 he had acquired
an understanding of bamboos as living plants wider and more detailed than
that of any of his predecessors, and enriched by his own original thought. For
the first time he applied to American bamboos the same kind of comprehensive
observation and thought, and this has resulted in a synthesis, presented in this
book, of unique significance. He examined many native bamboos over a wide
range in the American tropics and subtropics, collecting specimens with all the
kind of detailed observation he had found to be necessary; he also examined
type-specimens and other material of all previously described species, attempting
to correlate specimens which often failed to show all the characters he believed
to be significant. He made a new bamboo garden, in which once more he could
study growing plants. The work is not completed; the imperfections of earliercollected
specimens leave gaps in our knowledge. Although clear distinctions
between species in most genera remain to be established, the present work is a
firm foundation on which further work can be based, and an indispensible guide
to further thought and action. The excellent illustrations, designed to show
clearly a wealth of significant detail, are a very important complement to the text.
My own knowledge of bamboos as living plants is confined to the species
of Malaya and New Guinea, and some from India seen in cultivation. The
tropical American species dealt with in the present book are in the main very
different from those of the Malayan region, and prompt some thoughts, which
I hope may be of interest, though I have not an adequate knowledge of Japanese
bamboos. These are also important to the following statement concerning
the evolutionary significance of bamboos as members of the family Gramineae.
It seems to me possible that the development of woody culms, which is a
distinctive feature of bamboos, may have originated more than once in the
family; that is, the bamboos as we know them may not be monophyletic in this
respect. It seems likely that the Gramineae had developed a reduced and
specialized type of inflorescence before woody forms appeared (the very specialized
woody nature of bamboos is unique and cannot be a relic of any
proto-monocotyledonous stock). The large bamboos of the genera Bambusa,
Dendrocalamus, and Gigantochloa (which I believe are closely allied, though
separated by Munro on fruit-characters for which he had little evidence, but
which were repeatedly copied by later authors), as well as Schizostachyum and
its allies, all have six stamens and a well-developed pericarp. They also all
have a caespitose habit with pachymorph rhizome-elements, and inflorescences
of the iterauctant type that matches exactly vegetative branching, with a
prophyllum at the base of every branch, Thus these bamboos, predominant in
Southeast Asia, in addition to the maximum development of woody habit, also
show a combination of primitive characters. Of this great group of “primitive”
bamboos, only Bambusa migrated to the Americas from Asia, presumably at a
time when the land-bridge to North America from Asia experienced a mild
climate.
The other well-known genus shared by America and Asia is Arundinaria.
One may also postulate for it a similar migration (even as restricted by McClure)
vi
since it is much more diversified in Asia than America. This genus has been
variously interpreted, both in America and Asia, by different authors. Japanese
botanists have recognized that some of their bamboos, originally included in
Arundinaria, differed in many ways, and have attempted to recognize new genera.
But the type-species of Arundinaria is North American, and so the species of
Asia must be judged by that type. In the present book McClure deals exhaustively
with A,. gigantea, and indicates what he regards as the essential characters
of the genus, from which he excludes many American species formerly included
in it. Unfortunately he does not indicate which species of the Old World he
regards as congeneric with the American type. I believe that students of Japanese
bamboos would find the treatment of Arundinaria in the present work valuable
in a reassessment of their own species of the segregate genera.
Arundinaria has a semelauctant inflorescence in which the spikelets are
grass-like and the branches of the inflorescence lack prophylla at their bases.
In the Bambusa group of genera there is a prophyllum as first foliar organ on
every branch, leafy or flowering, right up to the prophylla called paleae which
enclose the true floral elements. In most grasses there are no prophylla (and no
bracts) at any branch of the inflorescence, but the paleae persist. There is, thus,
a gap between the prophylla of vegetative branches and the paleae, so that the
homology of vegetative prophylla and paleae was long unrecognized. The
semelauctant bamboos sometimes show intermediate stages, with prophylla
present at the base of inflorescence branches (Figures 3 ~ , 19~). The presence
of such intermediate stages may be significant as indicating possible evolutionary
lines of transition from the Bambusa to the grass types of inflorescence. It seems
to me likely that the transition has occurred on several lines. The bamboos of
Japan might provide interesting evidence, if considered from this standpoint.
In any case, precise observation of bracts and prophylla in relation to flowering
branches of semelauctant bamboos may indicate significant diagnostic characters
for recognition of species.
Arundinaria has leptomorph rhizomes from which new culms arise as
axillary structures; that is, the growth of the rhizomes is monopodial. I have
produced evidence (Holttum, 1955) that sympodial branching of the stem predominates
in all families of monocotyledons, and monopodial branching is
certainly the exception in Gramineae. The latter occurs in some genera of
Paniceae, but I believe it is rare in other divisions of the family, In some other
families (e. g., Palms) it has developed separately in distinct groups of genera,
and it may have done the same among woody Gramineae. Possibly therefore the
occurrence of leptomorph rhizomes in Awndinaria is less important as evidence
of relationship to other genera than some other characters.
In this book McClure describes a new species of bamboo which he includes
in the genus Yushania, previously known only from Taiwan. This looks like a
third migrant (or could both have evolved from a common northern ancestor
no longer extant?). Other genera of American bamboos, now more clearly distinguished
than formerly, may be found to give indications of relationships
with Asia.
One group of species, however, appears to be quite exclusive to the New
World, and surely originated there, namely those here included in Chusquea and
its immediate allies Neurolepis and Swallenochloa. Munro (1868: 13, 52) remarked
on the resemblance between Chusquea and some panicoid grasses, and
also to the fact that in essentials of spikelet structure they are not far apart.
vii
It is greatly to be regretted that McClure was not able to complete a survey
of the bamboo genera of the Old World, of which he had such a wide and
intimate knowledge. He accumulated, however, an immense amount of the
reference material necessary for such a survey, which, together with his manuscript
notes, is available as a basis for further study by his students. I hope such
study will result in published works, so that more of the fruits of McClure's
labors may become available to botanists, foresters, and others in Southeast
Asia, and may be put to practical as well as scientific use. I believe that the
peculiar and very remarkable properties of the bamboos of tropical Asia could
find new uses in modern technology, but they have not yet received the attention
they deserve. McClure's bamboo plantation at Canton could yet be of
great service in providing more information about these plants.
A new study of bamboo anatomy, especially nodal anatomy, is certainly
a key to understanding the branch patterns which McClure and others have
shown to be distinctive characters. Plant anatomists have been too content
to describe only internodal structure (which is, of course, important). Nodal
anatomy is extremely complex, and doubtless difficult to understand, but on it
depends the dynamics of growth of bamboos and of some other monocotyledons.
Because there is not secondary growth of concentric woody tissue in bamboos,
the primary structure of the node has to be far more precise than in dicotyledons,
and intercalary growth at the base of each culm internode needs a far more
complex organization. When this is better understood, respect for bamboos as
highly specialized organisms may well be greatly enhanced. The question has
also to be asked: How is the control of all this complexity organized?
R. E. HOLTTUM
Royal Botanic Gardens
Kew, England
November 1971
viii
Acknowledgments
The laboratory studies and field work that serve as the basis for the following
treatise cover a period dating back to 1943. In that year I initiated, under the
auspices of the Smithsonian Institution, a project for the U. S. National Research
Council: a survey of bamboo species of a special category occurring in Mexico,
Central American countries, Colombia, Venezuela, and Brazil (McClure: 1944).
Opportunities to continue the collection and field study of bamboos of the
Western Hemisphere were afforded during my incumbency (1944-1954) as
Field Service Consultant on Bamboo with the Office of Foreign Agricultural
Relations of the U. S. Department of Agriculture. More recently, as Consultant
on Tropical Forestry to the Maria Moors Cabot Foundation of Harvard University
(1956-1959) I was given the opportunity to visit European herbaria to
study types and critical specimens of Western Hemisphere bamboo genera,
and to advance the organization of pertinent data from the literature and from
personal experience. Since 1955, the Department of Botany of the Smithsonian
Institution, as administrator of grants-in-aid from the U. S. National Science
Foundation, has fostered the progress of my bamboo studies directed toward
the revision of the descriptions of the bamboo genera of the world, for the
second edition of Engler and Prantl’s Die Naturlichen Pflanrenfamilien.
Facilities for the study of the Gramineae (including the Bambusoideae) now
available at the United States National Herbarium were greatly enriched by the
fruits of the many mutually related activities pursued by Dr. Agnes Chase during
the course of her long and distinguished career as agrostologist. Among these
fruits are innumerable notes and photographs illuminating types and critical
specimens of gramineous plants that she investigated in the principal herbaria of
Europe and America. In many cases these notes and photographs are accompanied
by duplicate specimens or fragements from type-collections-specimens
presented as tokens of appreciation for expert curatorial services (identifications,
annotations, etc.) freely, but often at great personal sacrifice, bestowed upon
taxonomically critical materials in extant (but in many cases neglected) classical
collections. The acquisition of specimens representative of the grass (and
bamboo) flora of the world was greatly augmented by a system of exchanges
supported by long sets of duplicates by Dr. Chase on numerous wide-ranging
field trips carried out in the United States and other New World countries.
Fruitful discussions with Thomas R. Soderstrom, Associate Curator, Division
of Grasses, and David B. Lellinger, Associate Curator, Division of Ferns, Depaftment
of Botany, Smithsonian Institution, are gratefully acknowledged.
The skills of several artists went into the preparation of the line drawingssome
based on my own original sketches, others made specifically for the present
work: Elmer W. Smith, Mrs. Gesina Berendina Threlkeld, Mrs. Martha Niepold,
and Florence Mekeel (Mrs. H. J. Lambeth).
Essential to the fulfillment of the objectives of this study has been the
opportunity to examine the types and critical specimens of the genera under
consideration. In this connection, it is a pleasure to acknowledge the courtesies
extended by the responsible officials of the following botanical institutions:
The Herbarium of the Field Museum of Natural History, Chicago; the Herbarium
of the New York Botanical Garden; the Herbarium of the Arnold Arboretum,
Cambridge; The Herbarium of the Museum National d’Histoire
Naturelle, Paris; The Herbarium of the Royal Botanic Gardens, Kew; The
Herbarium of the British Museum (Natural History), London; the Herbarium
of the Botanische Museum, Berlin-Dahlem; The Herbarium of the University
of Pisa; The Makino Herbarium, and the Herbarium of the Botanical Institute,
Tokyo; and the United States National Herbarium (Smithsonian Institution),
Washington, D. C.
F. A. MCCLURE
Smithsonian Institution
Washington, D. C.
I March 1970
I I
To Conrad Chapman, who from the beginning has manifested a sustained
and sustaining interest in the objectives and the fruition of my bamboos studies.
X
Contents
Page ... Preface, by Edward S. Ayensu . . . . . . . . . . . . . . . . . . . . .
Foreword, by R. E. Holttum . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments, by F. A. McClure . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Key to the Recognized Genera of Bamboos Native to the New World . . . .
Apoclada McClure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Arthrostylidium Ruprecht . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Athroostachys Bentham . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
6
Arundinaria Michaux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Atractantha McClure, new genus . . . . . . . . . . . . . " . . " " " " 42
Aulonemia Goudot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Bambusa Schreber, Subgenus Guadua (Kunth) Hackel . . . .
Colanthelia McClure and E. W. Smith, new genus . . . . . . . . . . . . . . . . . . . . 77
Glaziophyton Franchet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Chusquea Kunth . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Elytrostachys McClure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Merostachys Sprengel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Myriocladus Swallen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Neurolepis Meisner . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Rhipidocladum McClure, new ge . . . . . . . . . . . . . . . . . . . . . . .
Swallenochloa McClure, new genus . . . . . . . . . . . . . . . . . . . . . . . .
Yushania K. H. Keng . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Literature Cited 140 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xi

Genera of Bamboos
Native to the New World
(Gramineae: Bambusoideae)
F. A. McClure
(EDITED BY THOMAS R. SODERSTROM)
True knowledge can only be acquired piecemeal,
by the patient interrogation of nature.
Sir Edmund Whittaker
Introduction
It is now more than three quarters of a century
since the preliminary publication of Hackel’s
treatment of the bamboo genera of the world, prepared
as part of the first edition of Engler and
Prantl’s Die Naturlichen Pflanzenfumilien (1887).
The fact that several supplements were published
shortly after is eloquent of Hackel’s feeling that
his work on this group was incomplete and of an
essentially tentative nature. It is sobering to reflect
that, although the number of bamboo species
known to science has more than doubled in the
meantime, the nature of the material collected has
changed but little. The problem of achieving an
adequate perspective on the taxonomy of the group,
however, has become immensely more complicated,
rather than simplified, by much that has been done
in this interval. Specimens collected continue to be
fragmentary, field notes are still brief or lacking,
F. A. McClure, Honorary Research Associate, Department of
Botany, Smithsonian Institution, Washington, D. C. 20560
(Deceased 1970). Thomas R. Soderstrom, Department of
Botany, Smithsonian Institution, Washington, D. C. 20560.
and the published descriptions of entities have
continued to omit reference to features of fundamental
taxonomic importance, especially those relating
to the nature of the branching habit of all
axes of the plant. Published supplementary studies
in the fields of anatomy and cytology have not, in
general, improved the situation as they might have,
had the workers in these fields collaborated more
closely with the taxonomist.
The present treatment is neither monographic
nor definitive in its intent or scope. It is offered as
representing a preliminary step forward in the
taxonomic conquest of the bamboos of the New
World. The dominant objective of the studies on
which this treatise is based has been to facilitate
the recognition of the generic affiliation of each
bamboo native to the Western Hemisphere that
requires identification. To this end-and within
the limits imposed by available materials-an effort
has been made to bring into sharper focus the
image (dependable features) and the apparent
present boundaries of each genus. It has been possible
to improve the concept and the circumscription
of some genera by including generally
neglected morphological features. The elimination
2 SMITHSONIAN CONTRIBUTIONS TO BOTANY
of traditionally included features that do not hold
good has improved the concept and the circumscription
of some genera, and has suggested a
revised taxonomic disposition of others. The diverse
patterns oE branching habit in the vegetative apparatus
and in the inflorescence have been given
particular attention, both here and in a broad
survey published earlier (McClure, 1966b). The
exploitation of the taxonomic potential of these
patterns is limited to those entities in the available
specimens of which this feature is represented.
The incompleteness and disparities (notably the
incongruent representations of morphological and
ontogenetic aspects of the plant) that characterize
the available specimens, field notes, and descriptions
of most recorded bamboo species exert serious
restrictions upon the development of uniformly
satisfactory circumscriptions of the genera. I have
been able to improve and olarify the image of
representative species of some of the genera through
field studies. Available plants of the type-species
and/or one or more representative species of fourteen
genera of bamboos of the New World and
the Old World, respectively, have been maintained
under cultivation for a number of years for progressive
collection and observation of their successive
stages of development, on a comparative basis.
Owing to the incomplete and fragmentary representation
of some genera in the existing herbarium
material and the fact that living material of some
of the genera has not been accessible to me, the
drafting of generic descriptions that are uniform
in their coverage of every detail is still an impossibility,
In comparison with previous treatments,
however, the coverage of taxonomically significant
features has been greatly extended as far as gross
morphology is concerned. Attention has thus been
directed to structures and to dynamic aspects of
the bamboo plant that have hitherto been largely
or completely neglected. It is hoped that those
who collect bamboo specimens for identification or
for description and naming and those who address
themselves to the description and classification of
the bamboos will in the future find useful leads
here, and that the taxonomic conquest of the bamboos
will have received a perceptible impetus. The
new demands for evidence from other disciplines
means that in order to be up-to-date the collector
must abandon the idea that the conventional leafy
flowering specimens are sufficient for purposes of
description and classification. He must have in
mind not only the morphologist but the anatomist,
cytologist, chemist, geneticist, biochemist, ecologist,
etc., so that material for the study of each entity in
as many disciplines as possible may be available
under the same herbarium voucher.
In some cases, the nomenclatural type of a genus
(especially where its choice was fortuitous) represents
a species that is morphologically peripheral
to the main body of the currently recognized com
ponents of the genus. In any case, the depth of
presently attainable taxonomic perspectives varies
from genus to genus. For this reason, the emphasis
on phylogeny is uneven, tentative, and of necessity
extremely limited in this treatment of the
genera. During the course of the phylogenetic die
versification of bamboo taxa, some of the characteristic
expressions of different individual features
(conventionally referred to as "characters") have
recombined (in disregard of quasi-generic boundaries).
This may be due to the persistence of
genetic compatibility within populations of related
plants, or between groups of related species, some
of which have already acquired combinations of
morphological features of such strong divergence as
to suggest, respectively, either specific or generic
segregation by the taxonomist.
The proposed consolidation of currently recognized
genera that apparently are not set off from
each other by clearly disjunct patterns of morphological
feature combinations is prompted by
the desire to facilitate identifications to genus. Uncertainties
often arise as a result of feature combinations
that constitute intergradations between
two or more such genera, especially where these
intergradations have either been ignored or given
separate generic rank. An effort has been made to
eliminate these uncertainties by allowing the intergradations
to fall within a single genus with an
expanded circumscription. They then yoke toge
ther the recognizable extremes under one generic
name. Where feasible, the diversities are then given
informal status as either subgenera or sections.
However, it is possible that more comprehensive
morphological studies, supplemented by correlated
studies in other disciplines and deepened by the
use of the electron scanning microscope of both
vegetative and reproductive structures, may force
the revision of the generic lines proposed here.
NUMBER 9
It appears that some groups of closely related
bamboos flout the intellectual concepts of morphological
homogeneity and morphological disjunctions,
at either the “species level” or the “genus
level”-or both, Of course, the terms “genus level”
and “species level” lack a sound semantic basis,
since the component elements of both genera and
species are, in fact, determined primarily on the
basis of circumscription (fences) that reflect morphological
similarities and disjunctions rather than
levels of evolutionary advancement. As every perceptive
taxonomist recognizes, the concepts “relatively
advanced” and “relatively primitive” are
properly focused upon the individual features or
structures of an organism, rather than upon the
organism as a whole or a group of organisms. Carlquist
(1961:7) states the point very simply in these
words: “One must remember that only characteristics,
not plants, or species, are primitive or advanced.”
This is because features commonly evolve
independently-as though possessing their own individual
phylogeny-rather than simultaneously
with other characters, as though linked with them,
genotypically. Proposed phylogenetic arrangements
of taxa, or groups of taxa, can be given meaning
only by a projection of the relative incidence of
supposedly “advanced” and supposedly “primitive”
features in their component elements.
The temporal and morphological divergencies
that, in most bamboos, isolate the vegetative state
from the sexually reproductive state, combine to
give rise to a physical isolation between the specimens,
descriptions, and names that, in many cases,
independently represent these two states of a given
species. As a result of the practical difficulties involved
in bringing adequate representation of the
essential morphological features of the vegetative
state of a bamboo into authentic association with
the morphological features of its reproductive state,
a dichotomy similar to that which has produced
the mycological category “Fungi Imperfecti” has
developed in the taxonomy of the bamboos. In
comparison with the fungi, however, most bamboos
confront both the collector and the taxonomist with
formidable obstacles to the resolution of these
difficulties. These obstacles are imposed by a very
protracted vegetative state, and morphological gradations
that appear in acropetal sequence within
each component axis of the plant. These and other
gradations appear also in temporal sequence among
the component parts of the relatively massive body
of the plant as it develops toward sexual maturity.
Another stumbling block has its origin in the
radical changes in appearance and composition
that often take place in the inflorescences during
their development, especially in those genera where
they are of the iterauctant (indeterminate) type
(McClure, 1966b:93 et seq.). The generic affinities
suggested by the morphological features of the vegetative
apparatus of a given species may be, and
often are, quite different from those suggested by
the morphological features of the reproductive apparatus
of the same species. Therefore, the full
integration of the vegetative features with the reproductive
features of each species is the only rational
approach to the achievement of a realistic
circumscription of the genera. This calls for drastic
improvements in collecting methods and in taxonomic
studies as well. The morphology and the
ontogeny of the whole plant must eventually be
assembled and brought under consideration. Until
this has been accomplished the taxonomist cannot
function effectively (or respectably) in collaboration
with specialists in integrating the results of
studies on paramorphological aspects of the bamboos
for the progressive clarification of their taxonomy.
A realistic approach to the solution of the
“genus problem” requires the building up, maintenance,
and sustained competent comprehensive
interdisciplinary study of living collections embracing
both the vegetative and the reproductive states
of critical species of each genus.
In the first sentence of Chapter I of his illuminating
review entitled Comparative Plant Anatomy
(1961), Carlquist reminds us that “comparative
anatomical studies of angiosperms have achieved
a remarkable record within the past century, and
one may safely say that few systematic studies would
fail to benefit from incorporation of anatomical
data.” And again, on page 120, “certainly no generic
monograph can be said to be complete without
studies on leaf anatomy.” This “guide to taxonomic
and evolutionary application of anatomical
data in angiosperms” serves the useful purpose of
emphasizing the impressive number and diversity
of the anatomical features that have been brought
into focus by studies of angiosperms other than
the bamboos. The sparsity of references to bam4
SMITHSONIAN CONTRIBUTIONS TO BOTANY
boos incorporated by Carlquist in his book emphasizes
the retarded state of our current knowledge
of the anatomy of this group of plants. It is regrettable,
however, that so few of the published
anatomical studies of the bamboos are sufficiently
comprehensive or well documented to afford a reliable
basis for conclusions as to their taxonomic
significance for the differentiation of genera. It remains
to be seen whether really comprehensive
studies on their anatomy will be significant for the
taxonomy of the bamboos as they have been shown
to be in some other groups of the Gramineae.
In circumscription of genera, the taxonomic usefulness
of features still to be revealed by paramorphological
studies of the bamboos remains to be
effectively explored. For the time being, at least,
the illumination that is potential in such distinguished
exploratory studies as those carried out
by Shibata (1900), Brandis (1907), Ohki (1932),
Takenouchi (1931a,b), Freier (1941), Metcalfe
(1960), and others (cited under the pertinent genera)
is eclipsed by the inadequate array of species
and features covered, and by the shadow of persistent
uncertainties with respect to where the truly
generic boundaries lie and, in some cases, with respect
to the identification or the documentation of
those species that were studied. The extent to
which the present work may appreciably improve
the view is limited by the lack of evidence as to
whether, and at what points, paramorphological
features will strengthen or modify the image of any
genus as it is here portrayed.
Valuable perspective on the incidence and the
varations of a number of anatomical features of 47
species listed under 25 genera of bamboos are presented
by Metcalfe (1960). Insofar as the bamboos
are concerned, major attention is given in this
work to leaf anatomy. The generic perspectives
afforded by these recorded data are summarized
by Metcalfe (1960584) as follows: “The anatomical
data obtained for the leaves of the various bamboos
examined by the present writer have been
brought together in Table I. Perusal of this table
confirms the opinion already expressed that there
are no clear-cut distinctions between any of the
genera that have been examined so far as leaf structure
is concerned.”
Ohki (1934) published a summary of his studies
of spodograms of the leaf epidermis in representative
species of Bambusa, Dendrocalamus, Phyllo.
stachys, Pleioblastus, Sasa, Semiarundinaria, and
Sinobambusa. Of Ohki’s results, Metcalfe (1960:
584) has the following to say:
The facts recorded by Ohki seem to the present writer to
support the view that there are no very clear-cut divisions
between the genera on the basis of characters revealed in
spodograms. On the other hand, variations in the occurrence
and distribution of macro-hairs, micro-hairs, and pridtlehairs,
variations in the width of the bands of bulliform cells,
differences in the number and distribution of papillae on the
long-cells have specific diagnostic value. This applies also to
quantitave characters such as the dimensions of stomata and
prickle-hairs. For the indentification of some species it is important
to know whether the walls of the basal cells of the
micro-hairs are smooth or scabrous, and whether or not the
walls of the prickle-hairs bear protuberances.
The hazards of attributing to all members of a
taxon (genus or species) a feature whose presence
has been demonstrated in a single specimen or a
single member has not always been heeded by the
taxonomist. Metcalfe (1960:584) observes: “Thus
we find uniseriate micro-hairs only in Guaduellu
oblonga amongst the species examined by the author.
Then again, Phyllostachys is the only genus
from certain species of which fusoid-cells appear
to be absent.”
Some of the taxonomic characters most promising
for the elucidation of the bamboos at the genus
level are to be found in features of structures that
are rarely represented at all in extant herbarium
specimens-and never adequately so. This lack
could be remedied either by extensive travel and
repeated visits to wild stands of representative taxa,
or by the establishment and maintenance of vast
living collections for the same purpose. In either
case, the lifetime of one person will not be enough.
The traditional and still prevalent disproportionate
emphasis on the importance of the diversity
comprehended by features of the reproductive
structures as the all-but-exclusive source of fundamental
taxonomic characters has resulted in the
acceptance of mere fragments of the plant as specimens
adequately documenting most known taxa.
It has also given us a vast but relatively sterile
literary heritage. These unchallenged bad examples,
aided by the priority principle embodied in
the ICBN have tended to perpetuate-even encourage-
undue haste in both the collection of specimens
for preservation, and the preparation of their
formal description for publication. The body of
NUMBER 9 5
uncorrelated literature has become undigestible.
The elevation of the level of refinement at which
future taxonomic treatments of bamboos may be
executed is conditioned upon the improvement of
certain procedures. In order to make this progress
possible, collectors must adopt the objective of
documenting for the taxonomist a more detailed
image of each taxon collected. This would involve
more deliberation in the selection of the components
of the conventional array of pressed specimens.
It would provide for more extensive and
more sustained field observations, and the recording
of the results in the form of notes and sketches
and/or photographs to accompany the pressed
specimens. It would involve the simultaneous collection
of other appropriately selected, preserved,
and correlated study materials by way of fostering
the progressive development and integration of diversified
studies, under interdisciplinary collaboration,
of documented materials from a common
source for each individual taxon (McClure, 1966b:
6, 147). The study set of any given collector number
should be made to represent the plant as
completely as possible before duplicate specimens
are segregated for distribution.
Plant taxonomy is currently in a period of accelerated
evolution. Perspectives on its problems
and its methods are changing. Under the stimuli
exerted by a developing dissatisfaction with the
results achievable through traditional methods, and
by the developing interest of specialists and their
published studies in several fields ancillary to morphology,
perceptive taxonomists are actively seeking
new ways to bridge the gaps between the way
things are in nature and the ways in which they are
pictured in conventional treatises. It is anticipated
that, through the gradual accumulation and integration
of the results of interdisciplinary collaboration
in morphological and paramorphological
studies, the present work will be superseded by a
greatly improved and more comprehensive portrayal
of the bamboos-one that more adequately
illuminates the details of their individualities in
multidisciplinary depth and clarifies their generic
and phylogenetic relationships, both within and
beyond the subfamily Bambusoideae.
ABOUT THE KEY.-A key is by many botanists admired
on the basis of the conciseness of its leads.
Partly for this reason, the array of contrasting features
used in botanical keys is, by convenience,
kept to a minimum. However, the user of such
keys may be misled where exceptions to the impressions
created by unduly abbreviated leads are
left unmentioned. The negligent perpetuation of
loose interpretations of structural complexities can
thwart the desire to make an effective key. Inadvertently
incorporated ambiguities may betray both
the maker of a key and the user of it.
The relatively lavish use of contrasting characters
that appears at some points in this key is motivated
by practical considerations. Effective definition
of the assumed disjunctions between some
pairs of closely related bamboo genera demands it.
Again, specimens typical of extant herbarium material
of most bamboos afford such a sparsely diversified
and, in many ways, incongruent representation
of the taxonomically critical structures that it
is only by means of either a very broad scoop or a
net with small meshes that many of them may,
with any degree of confidence, be directed into a
labeled pigeonhole. Because of limitations of available
documented coverage it has not been possible
to make genus-wide assessment of the potential of
some features as sources of contrasting characters;
however, the taxonomically useful array of contrasting
characters has been augmented, both by drawing
upon several morphological features hitherto
neglected and by clarifying others that tradition
interprets loosely. Some characters elsewhere given
unqualified generic importance have had to be
eliminated at certain junctures.
An innovation that seems to serve a useful purpose
here incorporates in either lead of a given
pair, as occasion arises, mention of one or more
characters which, while they do not provide additional
contrasts, are common to the taxa designated
by that particular lead for subsequent differentiation
from each other.
Conventional botanical keys commonly use indention
as the means of matching members of each
pair of contrasting leads, and for setting off adjacent
pairs from each other, visually. Besides
being uneconomical in many ways, this usage occasions
avoidable inconveniences and disadvantages.
It confronts both the preparer of the manuscript
and the typesetter with hurdles that are not
always successfully cleared. A key of the design
offered here is easier to prepare, easier to set in
6 SMITHSONIAN CONTRIBUTIONS TO BOTANY
type, easier to proofread, and less costly to publish,
than one embodying indentation. After a little
practice in its use, one needs make no greater effort,
either to follow the path that leads to the
right name, or to retrace one’s steps in case one
loses his way.
Key to the Recognized Genera of Bamboos Native to the New World
la. Culms in the vegetative state normally not developing branches .......................................... 2
1b.
2a. Flowering culms leafless, the above-ground internodes exceedingly thin walled and provided
with pith septa at close intervals, the lowermost one greatly elongated: branch
buds solitary at culm nodes: inflorescence never terminal to a culm, short-peduncled,
the rachis deliquescent: spikelets terminating in a depauperate sterile anthecium ........
........................................................................................................................ Glaziophyton
Flowering culms leafy, the above-ground internodes not exceedingly thin walled, and
not provided with pith septa at close intervals, the lowermost one shorter than those
that follow it: branch buds lacking at culm nodes; inflorescence always terminal to a
Culms in the vegetative state normally developing branches ............................................ 3
2b.
3a.
3b.
4a.
4b.
5a.
5b.
6a.
6b.
7a.
7b.
8a.
8b.
9a.
9b.
1Oa.
culm, long peduncled, the rachis excurrent: spikelets terminating in a perfect floret
............................................................................................................................... ~ e u r o ~ p ~ s
Primary (first-order) branches (elements of the branch complement) at each midculm
node more than one; stamens 3 .......................... , . , , . , , , , , , . , , . . , , . . , , , , , . . , , , , , ,
Primary (first-order) branches (elements of the bra complement) at e
node single; stamens 3 ......................................................................
Transitional glumes regularly 4, the first two (“empty glumes”) in size and
distinct from the second two (“sterile lemmas”) ......................................................... ..5
Transitional glumes 0-1-2, empty glumes and sterile lemmas not distinguishable ....... 6
Culm internodes lacking a lumen (filled with pith as in Zea mays): initial (primary)
branch buds and primary (first-order) branches at each midculm node typically many
(rarely only 3 ) , of constellate insertion, the middle bud or branch several times as
large as those that accompany it: culms arising from well-developed rhizomes of either
pachymorph or leptomorph form in a given species, or from rhizomes of both forms
in the same plant (as in Chusquea fendleri) ............................................. Chusquea
(primary) branch buds and primary
(first-order) branches at each midculm node usually 3, 1- (rarely) 5, of level insertion,
the middle one dominant but not several times as large as those that flank it; culms
arising from weakly developed pachymorph rhizomes and, in some species, also by
tillering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Swallenochloa, new genus
Inflorescences iterauctant: terminal segment of each rachis conspicuously elongate: spikelets
terminating in a rudiment borne on a bristle-like prolongation of the rachilla
Inflorescences semelauctant; terminal segment of each rachis not conspicuously elongate:
Culms scandent‘; inflorescences capitate . . . . . . . . . . . . . . . . ................................ Athroostachys
Culms self supporting, the internodes typically more or less conspicuously sulcate,
inflorescences not capitate .......................................................................................... 8
Inflorescence a raceme: branches at midculm node to 5 ...................................... Apoclada
Inflorescence a panicle: branches at midculm node to 3 ................................... Yushania
Culm internodes provided with a lumen: initial
behind the palea of the solitary perfect floret ...................... Atmctantha, new genus
spikelets terminating in a depauperate sterile anthecium ........................................... 7
Primary element of the branch complement at midculm nodes flat, unsegmented, appressed
and wholly adnate to the surface of the culm: second-order branches of
apsidate insertion and displayed in fan-shaped array; inflorescences semelauctant .. , ,I0
Primary element of the branch complement at midculm nodes terete above its base,
segmented and diverging from the culm; and dominant over the fasciculate second-
Midculm sheaths (as far as known) with the blade very much narrower at its base than
the apex of the sheath proper, abruptly rounded to a constricted base, and more or
less strongly reflexed; the surface of the lower internodes in young culms commonly
shows color mottling; inflorescences typically spicate racemes, the rachis in some species
(as in Merostachys pluriflora) more or less strongly contracted, sometimes producing
a congestion of the spikelets; spikelets typically of secund orientation, either (a)
order branches: inflorescences either semelauctant or iterauctant ............................ 11
NUMBER 9 7
containing a single perfect flower (as in Merostachys speciosa) and terminating in a
rudiment borne on a bristle-like prolongation of the rachilla, the latter more or less
completely concealed in the narrow, canaliculate sulcus of the palea, or (b) containing
two or more perfect flowers (as in Merostachys pturipora), the prolongation
of the rachilla then usually more robust than bristle-like, the sulcus of the uppermost
fertile palea broader and not canaliculate, and the terminal structure a depauperate
(not rudimentary) sterile anthecium; transitional glumes at the base of the spikelet 2,
strongly differing in size and shape, both empty, the first an “empty glume,” the second
a “sterile (empty) lemma;” empty glume, sterile lemma and fertile anthecia all of
brittle, extremely fragile consistency and of a typical grayish color at maturity;
mature fruit oblong or ovoid, not compressed, the pericarp leathery or crustaceous,
thickened toward the base and apex, the sulcus and the basal position of the embryo
not manifest externally ........................................
Midculm sheaths (as far as known) with the blade
apex of the sheath proper, broadly triangular and not at all reflexed; the surface of
the internodes in young culms typically plain green. Inflorescences typically with
either spicate or open racemose branching, the rachis either deliquescent or excurrent
(when excurrent, either straight or more or less strongly geniculate; inflorescences
with excurrent rachises in some species secund (strongly so in Rhipidocladum maxonii;
weakly so in R. raceiniflorum and its allies) ; spikelets each containing few to several
perfect flowers and terminating regularly in a depauperate sterile anthecium; transitional
glumes at the base of the spikelet usually 3, rarely 2 or 4, all progressively
approaching the first fertile lemma in size and shape, the first two empty, the third
rarely and the fourth always (when present) subtending a depauperate flower; glumes
and anthecia neither very brittle nor of extremely fragile consistency, at maturity
usually representing variously tinted versions of stramineous color, in some species
punctate with minute green dots; mature fruit oblong, usually more or less noticeably
compressed, the pericarp pergamineous (leathery or crustaceous in R. verticillatum) ,
the sulcus and the basal position of the embryo usually more or less clearly manifest
externally . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rhipidocbdum, new genus
Rhizomes leptomorph; developed midculm branch complements of restricted insertion,
the primary element dominant-in some species (as in Arundinaria gigantea s p . tecta,
for example) occasionally either suppressed temporarily or remaining solitary; inflorescences
semelauctant; stigmas 3 ........................................................... A ~ n d i n a & ~
Rhizomes pachymorph; stigmas in most cases 2 (often 3 in Bambusa) .......................... 12
Inflorescences iterauctant, forming pseudospikelets ................................................... 13
Inflorescences semelauctant, not forming pseudospikelets ................................................ 14
Culm branches (at least at lower culm nodes) thorny (except in the unarmed northern
forms of Bam’busa arnpZexifoNa and 8. aculeata) ; spikelets proper not pedicellate (the
distal segment of the rachis not elongated) . . . . . . . . . . . . . . . . . . . Bambusa subgenus Guadua
Culm branches all unarmed; spikelets proper rendered pediceuate by the elongated
distal segment of the rachis ....................................................................... Eb’trostahYs
Primary branch at midculm nodes appressed, not swollen basally, and not bearing buds
at its congested proximal nodes; culms and branches in most known species characterized
by one or more long internodes followed by several nodes closely crowded
together by the suppression (abortion) of the intervening internodes; the Cuh
sheaths and branch sheaths as well as the leaf sheaths each thickened and indurate;
leaf blades sessile or subsessile, typically thick, of leathery texture, and broadly
rounded at the base (rather grass-like and only perceptibly narrowed at the base in
Myriocladus maguirei) ; inflorescences typically of elongate profile, with an excurrent
rachis, of open racemose or paniculate branching (a condensed linear panicle in M.
cardonae), the primary branches of the rachis in some species forming secund racemes;
spikelets each containing usually 2 perfect flowers, occasionally 3 and, in at least one
species, sometimes up to 5 ....................................................................... Myn’ocwus
Developed branch complements at midculm nodes with the primary element patent,
more or less prominently swollen basally, and potentially proliferating from buds at
the congested proximal nodes; the culms, and sometimes their branches, in a few
species (as in Arthrostylidium schomburgkii and Aulonemia queko) with a long
internode (usually the first above-ground one of culms) followed by 2 or more nodes
lob.
1 la.
llb.
12a.
12b.
13a.
13b.
14a.
14b.
8 SMITHSONIAN CONTRIBUTIONS TO BOTANY
closely crowded together by the suppression (abortion) of the intervening internodes;
the culm sheaths and leaf sheaths in a few species (as in Aulonernia eflusa and A. depexa)
thickened, indurate, and provided wih leathery or parchment-like blades; leaf
blades with well-developed petioles in A. effusa and A. deflexa; inflorescences either
spicate, subspicate, open racemose, purely paniculate, or with both racemose and
paniculate branching in the same plant; spikelets each containing few to many
perfect flowers .....................................................................................
15a. Base of the primary axis typically extending downward beyond the
of the prophyllum or lowermost circumcingent sheath to form a low, downwardtapered
promontory; inflorescence a raceme with zigzag rachis; plants of tropical or
subtropical climates ................................................................................ Arthrostylidim
l5b. Base of the primary axis not extending downward into a promontory; inflorescence a
panicle (sometimes intermixed with racemose branches in CohntheZio) with straight
rachises; plants of temperate or cold climates ................................................................ 16
16a. Sheath at midculm nodes provided with scissing from) a conspicuous persistent
................................................... ............................... Colanthelia, new genus
Im nodes typically lackin spicuous basal girdle
Apoclada McClure
FIGUW 1-5
Apoclada McClure, in Reitz, ed., 1967:57.
Plants unicespitose; unarmed, Rhizomes pachymorph.
Culm habit unknown, the internodes k
sulcate. Branch complement at each midculm node
arising from (what appears to be) more than one
primary bud, the primary (first-order) component
consisting of more than one axis (2-5) crowded
together and inserted in a horizontal line, each
primary axis dominant over branches of higher orders
potentially arising by proliferation from buds
(when these are present) at its proximal nodes.
Leaf blades with transverse veinlets typically not at
all or only weakly manifest externally.
Inflorescences semelauctant, each typically consisting
of but a single spikelet (weakly racemose in
Apoclada arenicola; some spikelets reduced to a
single floret in Apoclada diversa). Prophylla usually
absent (present in Apoclada diversa when a
flowering branch is reduced to a solitary spikelet).
Transitional glumes of conventional nature typically
lacking at the base of each spikelet-when
present (e.g., Figure 3 ~ ) typically consisting of leaf
sheaths bearing reduced blades, but highly diverse
(as between species) in form, number, and spatial
relation to the first fertile lemma. Spikelets usually
made pedicellate by being terminal to a peduncle,
sometimes (as occasionally in Apoclada diversa)
sessile; containing few (rarely but one) to several
to many perfect florets, and terminating in a depauperate
sterile anthecium. Lemma (when subtending
a functional flower) embracing its palea
only basally at maturity. Palea broadly sulcate and
2-keeled dorsally, the margins not at all or only
slightly and partly overlapping. Lodicules typically
3, the anterior 2 asymmetrical and paired, the posterior
one smaller and symmetrical. Stamens normally
3 (exceptionally as sometimes observed in
Apoclada diversa, varying from 3 to 6 in spikelets
from the same specimen), the filaments filiform,
free. Stigmas 2. Mature fruit (known only in
Apoclada simplex) an oblong or subfusiform, sulcate,
mucronate caryopsis, the pericarp glabrous,
appreciably thickened and crustaceous, the sulcus
broad, the basal position of the embryo weakly
manifest externally,
ETYMoLOGY.-The name Apoclada (Greek, apo,
separate, and clados, branch) alludes to the apparently
independent origin and insertion of the primary
(first-order) elements of the midculm branch
compiemen ts.
TYpE-sPECrEs.-Apoclada simplex McClure and L.
B. Smith.
RELATIONSHIPS.-Aside from certain morphological
features of the individual florets and a few
features of the vegetative structures that are shared
by other bamboo genera of the New World flora,
Apoclada’s only apparent special relationship to
another genus is one that vigorously stirs one’s
curiosity. I refer to the strong detailed resemblance
that links the fruits of Apoclada simplex (Figure
5 P-R) to the fruits of the in-other-respects very
different taxon, Rhipidocladum verticillatum (Figure
42 R-T).
NUMBER 9 9
DISTRIBUTION.-AS far as the available documentation
informs us, all four of the currently recognized
species of Apoclada are endemic to Brazil,
where they are restricted to low elevations near
the northern limits of the southern temperate zone.
Two of these species are represented only by specimens
collected from humid sites, and the other two
appear to be restricted to arid habitats.
Key to the Species of Apoclada
la. Plants of humid sites; culm internodes antrorse-pubescent: primary axes of branch complement
at midculm nodes more or less strongly unequal ................................................ 2
lb. Plants of and sites: culm internodes glabrous; primary axes of branch complement at
midculm nodes subequal, slender, appressed ................................................................... .3
2a. Culm internodes with walls of substantial thickness; primary axes of branch complement
at midculm nodes robust, divergent; flowering branches all leafy: inflorescences of
invariable form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................................ 4. A. simplex
2b. Culm internodes thin walled: primary axes of bra complement at midculm nodes
slender, appressed; flowering branches all leafless; inflorescences of highly variable
form ........................................................................................................... .3. A. diversa
Primary axes of midculm branch complements without basal buds: flowering branches
leafy; keels of palea not perceptibly winged: style densely hispid ...................
..................................... 2. A. canmvieira, new combination
buds; flowering branches leafless; keels of palea perceptibly winged: style glabrous.. .........
3a.
........................................
3b. Primary axes of midculm branch complements potentially proliferating from basal branch
1. dpocladu arenicola McClure, new species
FIGURES 1, 2
Culmi in speciminibus suppetentibus usque 1.1
m alti et 3 mm diametro, omnino glabri, internodiis
cavis, medianis (quorum paries crassitudinis
mediocris est) elongatis, versus cicatrices nodorum
ramiferorum vix vel leviter dilatatis, superficiei
omnino laevi et dura, primo opaca demum nitida.
Culmorum et ramorum primariorum vaginae deciduae
(in specimine suppetente deficientes, ita incognitae).
Rami primarii ad nodos medianos
culmorum prodientes plures, tenues, subaequales,
appressi, omnino papillati, praecipue versus basin
retrorse pubescentes, e nodis suis proximalibus proliferantes.
Foliorum vagina arcta, persistens, primo
praecipue apicem versus subtilissime et antrorse
scabriuscula et farinosula et papillata, alibi glabra
laevigata nitidaque, secus marginem externam ciliolata;
auriculis minutis vel rudimentariis vel obsoletis;
setis oralibus confertis, capillaceis, infime
cohaerentibus, superne liberis, pallidis, levibus, anfractisque,
denique cadentibus; ligula interiore
perbreve apice convexa, dorso canescenti, margine
dense ciliolata; ligula exteriore vix ulla, margine
primo dense ciliolata demum glabrescenti et vel
denticulata vel erosa vel integra; petiolo circa 1 mm
~~
.............................. 1. A. ardcoka, new species
longo, basi puberulo, superne glabro, primo glaucescenti
demum nitido; lamina angustissima, vel
planiuscula vel aculeiforrni rigidaque, usque ad 9
cm longa et circa (in statu siccato) 1 mm lata,
utrinsecus omnino glabra; nervis perpaucis; venulis
transversis haud manifestis.
Inflorescentiae ex culmo ab initio efoliato ortae;
nonnullae (saltem in specimine suppetente) ad
spiculam unicam redactae; vel solitariae vel 2-3
simul insertae a bractea communi subtentae; pro-
Phyllis nullis. Glumae transitionales vulgo 2 (interdum
1 vel 3) usque 3 cm inter se distantes,
naviculares, apice vel acutae vel acuminatae et
breviter aristatae, dorso carinatae plurinerves et
glabrae, substantia subhyalinae fragilesque, I ca 10
mm longa, 11 ca 11 mm longa, 111 ca 12 mm longa;
gluma transitionalis suprema nonnunquam florem
depauperatum subtendens ubi lemmati fertili similis.
Spiculae pedicellatae, usque 4 cm longae,
laxe pauciflorae, lanceatae, compressulae. Rachillae
segmenta tarde disarticulantia, 2-3 mm longa, prominenter
nervosa, clavata, unilateraliter complanata,
apice dilatata, vel omnino vel pro parte
puberula. Lemma pergamineum vel papyraceum,
usque ad 13 mm longum, lanceatum, plurinerve,
superne carinatum et apice breviter mucronatum.
Palea lemmate brevior, papyracea, oblonga, dorso

NUMBER 9 11
FIGURE I.-Apoclada arenicola McClure. A, Leafy vegetative
culm (midculm section), x 0.6; B, midculm bud complement,
x 6; c, branch complement (basal part), x 6; D,
leafy twig, x 1.8; E, apex of leaf sheath and base of leaf
blade x ca. 18; F, flowering culm, with rhizome, x 0.6.
Drawings A, c, F based on Chase 11886 (US) and B, D, E on
Chase 12007 (US).
latissime sulcata, in apicem variabilem vel hebete
acuta vel obtusa vel subtruncata et bi- vel quadridentata,
in plano mediano valde curvata, inter et
extra carinas Clare nervosa glabraque, secus carinas
prominenter alatas ciliis pallidis pulchre ornata.
Lodiculae vel opacae vel pellucidae vel diaphanae,
margine suprema ciliolatae alibi utrinsecus glabrae,
anterioribus 2 sub-semiovatis, posteriore lanceata.
Antherae usque 6 mm longae, brunneae, apice subpenicullatae.
Ovarium glabrum. Stylus glaber,
typice in stigmatibus 2 divisus, nonnumquam
B
(atypice) cum ram0 stigmatic0 tertio abortivo
praeditus. Caryopsis incognita.
Culms up to 1.1 m tall and 3 mm in diameter,
glabrous throughout, the midculm internodes
elongate, scarcely or only lightly flared toward the
sheath scar at branch-bearing nodes, hollow, with
a smooth, hard surface and walls of medium thickness.
Sheaths of the culm and sheaths of primary
branches deciduous (lacking in the available specimens).
Primary branches produced at midculm
nodes several, proliferating from buds at their basal
nodes, slender, subequal, appressed, papillate
throughout, retrorse-pubescent toward the base.
Leaf sheaths tight, persistent, obscurely antrorsescabrous,
perceptibly farinose and papillate near
the apex, elsewhere glabrous, smooth, and lustrous,
ciliolate on the outer margin; auricles minute, rudimentary
or obsolete; oral setae numerous, crowded,
TICURE 2.-Apoclada arenicola McClure, A, Spikelet, x 1.2; B, transitional glume I, X 7.2; c,
ransitional glume 11, x 7.2; D, transitional glume 111, x 7.2; E, lemma, x 7.2; F, floret,
x 7.2; G, palea, abaxial view of distal portion, x 7.2; H, palea, oblique view, x 7.2; I, lodicule
:omplement, x 15; J, stamen, x 7.2; K, gynoecium, x 30. All drawings based on Chase 11886
(US) '
.
12 SMITHSONIAN CONTRIBUTIONS TO BOTANY
hair-like, cohering basally, free, pale, smooth and
wavy above, falling away at length; inner ligule
very short, dorsally canescent, apically convex,
densely ciliolate on the margin; outer ligule a thin
line, its margin ciliolate at first, later glabrescent,
then seen as either denticulate, erose, or entire;
petiole ca 1 mm long, basally puberulous, glabrous
distally, lightly glaucous at first then lustrous; blade
very narrow, flattish to aculeiform and stiff, up to
9 cm long and (when dry) ca 1 mm broad, glabrous
throughout on both surfaces, the longitudinal
nerves very few, strong, with transverse veinlets not
at all apparent externally. Flowering culms leafless
from the first.
Inflorescences racemose (often reduced to a single
spikelet), solitary or emerging in twos or threes
and subtended at culm nodes and branch nodes by
a common bract; prophylla none. Transitional
glumes subhyaline, commonly 2 (sometimes 1 or
3) up to 3 cm distant from each other and from
the first fertile lemma, boat-shaped, acute or acuminate
and awned at the apex, dorsally keeled and
glabrous, I ca 10 mm long, 11 ca 11 mm long, 111
ca 12 mm long, the uppermost transitional glume
sometimes similar to the lemma and subtending a
depauperate flower. Spikelets lanceate, pedicellate,
up to 4 cm long, lax, compressed, comprising few
to several perfect flowers. Rachilla segments tardily
disarticulating, 2-3 mm long, curved, D-shaped
in cross-section, prominently nerved, flared toward
the apex, puberulous to subglabrous throughout.
Lemma pergamineous, up to 13 mm long, severalnerved,
keeled toward the acuminate apex and
terminating in a short awn, dorsally glabrous,
densely fringed with pale cilia on the outer margin.
Palea shorter than its lemma, papery, oblong, emarginate
and bi- or quadridentate at the broadly
rounded apex, broadly sulcate dorsally, prominently
nerved between and beyond the keels,
fringed with pale cilia on the keels, elsewhere
glabrous throughout. Lodicules (in the specimen)
either opaque or pellucid or diaphanous, ciliate at
the apex of the margin, elsewhere glabrous
throughout; the anterior 2 sub-semiovate, the posterior
one lanceate. Anthers up to 6 mm long,
yellow. Ovary glabrous, the style glabrous, typically
divided into 2 stigmatic branches, atypically
bearing (in addition) an abortive stigmatic branch.
Caryopsis unknown.
TYPE-CoLLECTIoN.-Brazil, Matto Grosso, between
Bonito and Rondonopolis, from a single
flowering plant among scattered colonies of plants
in the vegetative state growing in sandy soil, 8 iv
1930. Agnes Chase 11886 (holotype: US 1500498,
1500499). Additional specimens seen: between
Rondonopolis and Santa Rita do Araguaya, from
a plant in the vegetative state, representative of an
element common in the local flora for several
kilometers along the road, 13 iv 1930, Agnes Chase
12007 (US 1500468 and 1500469).
2. Apoclada cannavieira (Alvaro da Silveira)
McClure, new combination
FIGURE 3
Arundinaria cannavieira Alvaro da Silveira, 1919: 101, fig. 2.
TYPE-COLLECTION.-In campis arenosis siccisque in
Serra do Caparao (as published; ‘Serra do Cabral’
as indicated in the field notes that accompany the
type-collection) Minas Geraes, ubi sub nomine
‘cannavieira,’ vulgo cognita est. Floret Novembri.
Silveira 644 (holotype: Silveira herbarium), US
(duplicate from the type-collection).
3. Apoclada diversa
FIGURE 4
Apoclada diversa McClure and Smith, in Reitz, ed., 1967:62,
fig. 10z-e.
TYPE-COLLECTION.-Brazil, Santa Catarina, Cacador,
Rio Cacador, lugares limidos, 22 i 1946,
Swallen 8271 (holotype: US 2152479 and 2383418).
4. Apoclada simplex
FIGURE 5
Apoclada simplex McClure and Smith, in Reitz, ed., 1967:59,
fig. 10s-y.
FIGURE 3.-Apoclada cannavieira (Alvaro da Silveira) Mc-
Clure. A, Upper portion of flowering culm, x 0.6; B, apex
of leaf sheath and base of leaf blade, x 12; C, spikelet,
x 1.2; D, diagram of longitudinal section of spikelet, x ca.
2.4; E, transitional glume (sterile lemma), x 3.6; F, depauperate
floret attached to rachilla segment (transitional
glume [E], which subtended bud has been removed),
x 6; G, floret, x 2.7; H, lemma, x 2.7; I, palea, X 2.7: J,
stamen complement and lodicule complement, x 3.6; K,
lodicule complement, x 6; L, stamen, X 3.6; M, gynwcium,
x 14.4. All drawings based on Alvaro da Silveira $44 (US).
2
C
c
w,
Acl
0
W
0 2
J
NUMBER 9 15
FIGURE 4.-Apoclada diuersa McClure and L. B. Smith. A,
Upper portion of leafless flowering culm, x 0.6; B, bud complement
from node above midculm level, x 3; C, base of
branch complement, x 3; D, spikelet with antecedent part
of stem, x 1.2; E, bract that subtends flowering branches,
x 1.2; F, floret, x 7.2; G, lemma, x 7.2; H, palea, x 7.2;
I, lodicule complement, x 15; J, stamen, x 15; K, gynoecium,
x 15; L, fruit (juvenile), x 15. All drawings based on
Swallen 8271 (US).
TYPE-coLLEcrIoN.-Brazil, Santa Catarina, Campos
Novos, Palmares, pinhal, 900 m, 27 xii 1963,
Reitz and Klein 16355 (holotype: US 2434535).
Arthrostylidium Ruprecht
FIGURE 6
Arthrostylidiuin Ruprecht, 1839:27; 1840:117.-Hance, 1862:
235.-Munro 1868:39.-Doell, in Martius, 1880: 172.-Bentham,
in Bentham and Hooker, 1883: 1208.-Hackel, in
Engler and Prantl, ed., 1887:92.-Pilger, in Urban, ed.,
1900-1901 : 336.-Hackel, 1903a: 67.-Hitchcock, 1927b: 307;
1936:243.-McClure, 1957:199.
Plants unicespitose, unarmed. Rhizomes pachymorph.
Culms either self-supporting or clambering;
each midculm node bearing a single initial
(primary) branch bud, the prophyllum in some
species elongating simultaneously with the germination
of the bud, the primordium producing a
single segmented, terete primary axis, this dominant
over axes of higher orders that usually proliferate
promptly from conventionally subtended buds typically
present at its own proximal nodes; the base
of the primary axis typically extending downward
beyond the locus of insertion of the prophyllum or
lowermost circumcingent sheath to form a low,
downward-tapered promontory, this promontory
naked in a few species (as in the available specimens
of the type-species), elsewhere bearing usually
two distichously inserted buds or branches each
subtended by a small, noncircumcingent bract.
Branch complement (when developed) at midculm
nodes of either restricted or gremial insertion, the
primary member typically strongly to weakly dominant;
sometimes, however, apical growth in the
initial primordium fails to take place, even after
the precocious germination of buds at its proximal
nodes. Sheaths at midculm nodes with or without
a persistent girdle. Leaf blades with transverse
veinlets not at all manifest to more or less clearly
visible.
Inflorescences semelauctant, spicate racemes;
subtending bracts and prophylla always lacking,
pulvini at the base of the branches of the rachis
usually lacking; transitional glumes 2 or 3, the first
and second empty, the third a sterile lemma sometimes
subtending a rudiment; spikelets sessile or
subsessile, containing few to several perfect florets
and terminating in a depauperate sterile anthecium;
lemma of perfect florets subtending its palea
only basally at maturity, palea gaping antically,
broadly sulcate and 2-keeled dorsally; lodicules 3,
the anterior 2 asymmetrical and paired, the posterior
one smaller and symmetrical (exceptionally
none, as recorded for A. cacuminis). Stamens 3, the
filaments filiform, free. Stigmas 2. Mature fruitas
far as known (available examples are rare)-an
oblong mucronate, sulcate caryopsis, the pericarp
pergamineous, of even thickness throughout, or appreciably
thickened apically to form a distinct cap
or nodule basal to the stylar column; the sulcus
and the basal position of the embryo clearly manifest.
ETYMoLoGY.-The name Arthrosty Zidium (n,),
derived from the Greek, arthron, joint, and stylidium,
column, alludes to the readily disarticulating
nature of the rachilla, in spikelets of the lectotype
and numerous other (but not all) species of the
genus.
TypE-spEcIEs.-Hitchcock (1927b: 307) designated
Arthrostylidium cubense Ruprecht (1839:27) as the
type of the genus (see McClure 1957:199).
of the 43 trivial names that have previously been
associated with it, the image of the genus Arthrostylidium
still contains a few taxa that embody
conspicuously deviant features. Since in my present
perspective I characterize these features as “maverick”
(see glossary for definition), I have not given
them significance at the genus level. Some species
still retained in the genus (e.g., Arthrostylidium
schomburgkii) share such a feature with Aulon.emia
q ueko, Glaziophy t on m irab ile, R h ipidoclad um
harmonicum, and several species of Myriocladus.
I refer to a disproportionate elongation of at least
the first above-ground internode, and usually some
internodes of primary branches as well. These
elongated internodes are followed by one or more
aborted (obsolete) internodes, the successive ones
being separated from each other by normal nodes
RELATIONSHIPS.-EVen after the exchsion of 23

NUMBER 9 17
FIGURE 5.-Apoclada simplex McClure and L. B. Smith. A,
Portion of clump showing rhizomes and base of culms, x
0.12; 6, midculm internode, nodes and culm sheath, x 0.6;
c, apex of sheath from lower culm node, adaxial aspect,
x 1.2; D, apex of sheath from higher node, adaxial aspect,
x 1.2: E, midculm complement of leafy flowering branches, x ca. 0.6; F, variant of midculm branch complement, x
ca. 0.6; G, apex of leaf sheath and base of blade, x ca. 4.8;
H and I, spikelets, showing reduced laminiferous foliage leaf
standing in place of a first transitional glume, X 3; J,
sterile lemma (either standing empty or subtending a depauperate
flower) as a second transitional glume (that a n
be either a sterile lemma or a fertile lemma), x 6; K, floret
from middle of spikelet, x 6 L, palea, x 6; M, lodicule
complement, x ca. 18; N, stamen, x 15: 0, gynoecium, x
15; P, fruit, embryo side, showing weak indication of the
position of the embryo, x 7.2; Q, fruit, hilum side, showing
the sulcus, x 7.2; R, fruit, longitudinal section, x 7.2. Drawings
A, H based on Klein 4108 (US), B-F and I-R on Reitz
and Klein 16,355 (US), and G on Klein 4657 (US).
bearing normal sheaths and branch buds (or
branches) inserted distichously (cf. McClure, in
Maguire, Wurdack, et al., 1964:2). Available specimens
of Arthrostylidium fimbriatum show several
vegetative features that suggest introgression from
(or a possibly recessive genetic heritage shared by)
some members of the genus Myriocladus. I retain
this species in Arthrostylidium on the basis of the
characteristic features of the inflorescence. The
inflorescence in Arthrostylidium angustifolium
deviates from the spicate form of the inflorescence
of the type-species of the genus by substituting a
spicate raceme for each spikelet. I retain this species
in Arthrostylidium on the basis of a characteristic
feature complex of the midculm branch
complement. Arthrostylidium excelsum, A. sarmentosum,
and A. venezuelae produce spicate racemes
with a geniculate rachis (inflorescences technically
identical with those of Rhipidocladum harmonicum)
but the same characteristic feature complex
of their midculm branch complements keeps them
in Arthrostylidium.
In terms of the ontogeny and morphology of the
midculm branch complements, the taxa I retain in
Arthrostylidivm display a cline, one extreme of
which is found in A. cubense and the other in A.
sarmentosum. The central structure of this feature
complex is the promontory (Figure SD), a tapered
bulge that extends downward from the locus of
initiation of the primoridum of the solitary primary
bud at midculm nodes. In available specimens
of A. cubense the promontory lacks lateral
buds; in A. ekrnanii it sometimes bears one bud; in
A. capillifolium it sometimes bears two buds; in
A. urbanii it sometimes bears four buds. The buds
that are borne on this outwardly unsegmented promontory
are distichously inserted and subtended
by noncircumcingent bract-like sheaths, while the
buds that are borne more distally, on the proximal
nodes of the segmented, terete primary axis that
develops from the apical meristem of the initial
primordium are subtended by circumcingent
sheaths. As the number of buds on the unsegmented
promontory increases, the number of circumcingent
sheaths subtending buds on the
proximal part of the primary axis decreases until,
in Arthrostylidium sarmentosum, all the buds (excepting
the uppermost one) are subtended by noncircumcingent
bract-like sheaths. In some specimens
of Arthrostylidium sarmentosum from Venezuela
and Puerto Rico the development of the strongly
dominant primary axis of the midculm branch complement
is aborted immediately above the bud or
branch at its uppermost proximal node. In other
specimens, the dominance of the primary axis over
the proximal secondary axes is greatly reduced. In
such a specimen-as for example in Soderstrom
1051 from Trinidad-the fully developed midculm
branch complement is deceptively similar to that
characteristic of all taxa I include in the genus
Rhipidocladum. The persistence in arthrostylidioid
taxa of the promontory (a feature that is entirely
lacking in rhipidocladoid taxa), however, suggests
that we have, in the vegetative cline observed in arthrostylidioid
taxa, a case of evolution convergent
toward Rhipidocladum with respect to the described
feature complex. This convergence is weakly
paralleled by the appearance, mentioned above, of a
geniculate rachis in the inflorescence of Arthrostylidium
excelsum, A. venezuelae, and A. sarmentosum,
a feature shared by Rhipidocladum solely
in R. harmonicum.
As initially delimited by Ruprecht, Arthrostylidium
was already highly polymorphic. In the Latin
notes that follow his brief formal description he
(Ruprecht, 1840: 117) says: “Genus Chusqueas et
Arundinarias jungens.” For years, I read this as
meaning simply that Arthrostylidium is a genus
that falls between Chusquea and Arundinaria. Although
Ruprecht did not actually transfer any
18 SMITHSONIAN CONTRIBUTIONS TO BOTANY
NUMBER 9 19
FIGURE 6.-Arthrostylidium cubense Ruprecht (A-M) and Arthrostylidiurn
rnultispicaturn Pilger (N-Q). A, Rhizome and
lower part of culm, x 0.6; B, leafy flowering midculm nodes,
x 0.6; c, apex of leaf sheath and base of blade, x a. 3;
D, insertion of components of midculm branch complement,
x 3; E, inflorescence, x 6; F, spikelet, x 6; G, transitional
glumes (3), x 6; H, floret, x 6; I, lemma, x 6; J, palea,
x 6; K, lodicule complement, x 15; L, stamen, X 15; M,
gynoecium, x 15; N, fruit, embryo side, x 12; 0, fruit, hilum
side, x 12; P, fruit, oblique aspect, x 12; Q, fruit, longitudinal
section, x 12. Drawings A and H-M based on Brother
Leon 16214 ( U S ) , B-D on Ekrnan 12649 (US), E-G on Wright
3809 (US) and N-Q on Leonard 3817 (US).
names from either Chusquea or Arundinaria to
Arthrostylidium, he continues in a vein that conveys
the impression that he does, in fact, consider
that known members of these three genera actually
form a continuum. Doell seems to have taken this
view, since he was led, apparently by Ruprecht’s
comparison of Chusquea leptophylla to Arthrostylidium
trinii, to publish the combination Arthrostylidium
leptophyllum (Nees) Doell (in
Martius, 1880: 175). Ruprecht’s image of Arundin,
aria was clearly derived from Nees (1834) and not
from Michaux (1803).
It appears that the natural boundary (disjunction)
between Arundinaria and Arthrostylidium
was obscured for Ruprecht (1839)-and has remained
so for other taxonomists, down to the present-
by the image of the genus Arundinaria which
Nees produced (1834:478-483) when he incorporated
it in such morphologically (and generically)
divergent species as Arundinaria falcata, A. verticillata,
A. amplissima, A. macrostachys, A. pinifolia,
A. wightiana, and A. glaucescens. Every one
of these species is excluded from Arundinaria by
my circumscription of the genus. Hance (1862:235;
1876:340) comments upon the weakness of the published
bases for the differentiation of Arthrmtylidium
and Arundinaria from each other. Major
treatises by Munro (1868), Bentham (in Bentham
and Hooker, 1883), Hackel (in Engler and Prantl,
ed., 1887), and E. -G. Camus (1913) are unanimous
in their perpetuation of the earlier failure to establish
a clear distinction between the respective
groups of New World bamboos that truly represent
these two genera.
The Latin notes that follow Ruprecht’s brief formal
description of the genus Arthrostylidium
(1839:27) read in free translation as follows:
Natives of tropical America. This genus unites the Chusqueas
and the Armdinaria. Among the latter especially,
fruit-bearing flowers of A[rundinaria] amplissima and A.
wightiana absciss, while in the rest [of Arundinaria] and in
Chusquea, the flowers adhere tenaciously to the rachilla or
“stylidium.” A[rthrostylidium] maculatum, by the sexual difference
within the dimorphic spikelets, provides a full notion
of the genus, [a notion] gradually obliterated in the
rest [of the species] by way of A[rthrostylidium] cubense.
Apparently begun by Ruprecht in 1839, the
search for a natural boundary (disjunction) between
Arthrostylidium and “Arundinaria” has remained
fruitless for more than 130 years. The
sense of futility induced by this perennial failure
is revealed dramatically in several taxonomic papers
that carry the quest into the 20th centuryamong
them Pilger (in Urban, ed., 1900-1901:
336-337), and Hackel (1903a:67-70). Pilger (in
Urban, ed., 1907:289) gave renewed expression to
his bewilderment in the following words (translated
from the German):
It now seems to me doubtful whether the genus Arthrostylidiurn
has any justification, or should go into Arundinaria.
Hackel (1903[a]:67 ff) has already transferred most
of the genus Arthrostylidium to Arundinaria, leaving in
Arthrostylidium only the forms with a doniventral inflorescence
([as in] A. Prestwi Munro from the West Indies).
I do not believe that this difference is sufficient to
justify the genus; a future monograph may well have to
unite the two genera.
Hackel’s tentative proposal (1903a: 69) to clarify
the concept of the genus Arthrostylidium by limiting
its content to the taxa represented by A. trinii
Ruprecht, A. racemiflorum Steudel, A. presto&
Munro, A. pittieri Hackel-“and perhaps also A.
fim briatum Grisebach’l-has been universally and
persistently ignored. A wholly different image of
the genus was adumbrated by Hitchcock’s (1927b:
307) selection of Arthrostylidium cubense Ruprecht
as lectotype. But the boundary between Arthrostylidium
and Arundinaria was still left undefined.
Pilger and Hackel both revealed, in print, a mutually
held conviction that if the basis for a taxonomic
boundary between Arthrostylidium and
Arundinaria really exists it must be sought amongst
the morphological diversities that characterize the
reproductive structures of the respective members
of the complex. In his key to bamboo genera
Hackel (in Engler and Prantl, ed., 1887:92) proposed
the separation of the then current images of
Arundinaria and Arthrostylidium solely on the ba20
SMITHSONIAN CONTRIBUTIONS TO BOTANY
sis of the (unreliable because unstable) number of
transitional glumes (empty glumes alone or empty
glumes and sterile lemmas) borne at the base of
each spikelet.
The natural boundary between Arthrostylidium
Ruprecht and Arundinaria Michaux still awaits
explicit definition. In morphological terms, plants
typical of Arthrostylidium are distinguishable from
plants typical of Arundinaria by the possession of
pachymorph rhizomes, a unicespitose clump habit,
a spicate inflorescence with sessile or subsessile
spikelets, and binate stigmas; while plants typical
of Arundinaria possess leptomorph rhizomes, a diffuse
or pluricespitose clump habit, a paniculate or
racemose inflorescence with pedicellate spikelets,
and ternate stigmas. The respective areas of the
natural distribution of plants of the two genera are
disjunct, both in space and in respect to prevailing
temperature minima; plants of the genus Arthrostylidium
being confined to frost-free sites, while
the genus Arundinaria is made up of frost-hardy
plants.
DISmIBuTIoN.-The twenty named New World
species I have retained in the genus Arthrostylidium
are all endemic to regions with a mesophytic,
tropical or subtropical climate. Their aggregate
recorded geographic range extends from Brazil,
with one known species, to Venezuela, with four
known species, to Cuba, with eight known species,
and several other Caribbean islands, each with one
or two species. Pilger (in Urban, ed., 1900-1901:
337) stresses the narrowly limited known distribution
of some of the taxa he calls “island species.”
Annotated Checklist of New World Species
Included in the Genus Arthrostylidium
As here conceived and circumscribed, the genus
Arthrostylidium embraces the following named
taxa, all of which are endemic to the New World.
1. Arthrostylidium angustifolium Nash, 1903: 172.
2. Arthrostylidium cacuminis McClure, in Maguire,
Wurdack, et al., 1964:3.
3. Arthrostylidium capillifolium Grisebach, 1862:
531.
Arundinaria capillifolia (Grisebach) Hackel, 1903a:
69.
4. Arthrostylidium cubense Ruprecht, 1839:28,
pl. iv:fig. 13 (Figure 6 A-M).
Arundinaria cubense (Ruprecht) Hackel, 190Ja:69.
5. Arthrostylidium distichum Pilger, in Urban,
ed., 1900-1901:342.
6. Arthrostylidium ekmanii Hitchcock, 1936: 16.
7. Arthrostylidium excelsum Grisebach, 1864:529.
Arundinaria excelsa (Grisebach) Hackel, 1903a:69.
As represented by currently available herbarium
specimens, Arthrosty lidium excelsum does not appear
to be clearly differentiated (disjunct) from
Arthrostylidium venezuelae.
8. Arthrostylidium fmbriatum Grisebach, 1862:
9. Arthrostylidium haitiense (Pilger) Hitchcock
Arundinaria haitiensis Pilger, in Urban, ed., 1907:
288.
10. Arthrostylidium longiflorum Munro, 1868:41.
Arundinaria longiflora (Munro) Hackel, 1903a:69.
Guadua exalata Doell, in Martius, 1880:181.
11. Arthrostylidium multispicatum Pilger, in Urban,
ed., 1900-1901:341.
Arundinaria multispicata (Pilger) Hackel, 1903a:
69.
12. Arthrostylidium obtusatum Pilger, in Urban,
ed., 1900-1901 :340.
Arundinaria obtusata (Pilger) Hackel, 1903a:69.
13. Arthrostylidium pubescens Ruprecht, 1839:29,
Arundinaria pubescens (Ruprecht) Hackel, 1903a:
69.
14. Arthrostylidium reflexum Hitchcock and Ekman,
in Hitchcock, 1936: 19.
15. Arthrostylidium sarmen.tosum Pilger, in Urban,
ed., 1903:108; Chase, 1914:278, pl. 21.
16. Arthrostylidium scandens McClure, in Maguire,
Wurdack, et al., 1964:4.
17. Arthrostylidium schomburgkii (Bennett) Munro,
1868:41.
Arundinaria schomburgkii Bennett, in Schomburgk,
1841:562.
1 8. A r t h rosty lid iu m simp liciuscu lum (Pilger) Mc-
Clure, new combination.
Arundinaria simpliciuscula Pilger, 1920:29.
19. Arthrostylidium urbanii Pilger, in Urban, ed.,
1900-1901 :339.
531.
and Chase, 1917:399.
pl. iv: fig. 14.
NUMBER 9 21
20. Arthrostylidium venezuelae (Steudel) McClure,
Chusquea ? venezuelae Steudel, 1854:337.
?Arundinaria standleyi Hitchcock, 1927a:79.
As illustrated by currently available herbarium
specimens, Arundinaria standleyi appears to be
only weakly differentiated from Arthrostylidium
venezuelae by the sparse dorsal vesture of its lemmas
in the form of lustrous white appressed or
spreading deciduous hairs.
1942172.
Checklist of New World Species Here Excluded
from the Genus Arthrostylidium
1. Arthrostylidium ampliflorum. See Rhipido-
2. Arthrostylidium amphimum. See Aulonemia
3. Arthrostylidium angustiflorum. See Rhipido-
4. Arthrostylidium aridaturn. See Aulonemia
5. Arthrostylidium bartlettii. See Rhipidocladum
6. Arthrostylidium burchellii. See Colanthelia
7. Arthrostylidium eflusum. See Aulonemia ef-
8. Arthrostylidium geminatum. See Rhipidocla-
9. Arthrostylidium haenkei. See Aulonemia
10. Arthrostylidium harmonicum. See Rhipidocla-
11. Arthrostylidium leptophyllum is Chusquea
12. Arthrostylidium longifolium. See Bambusa (sg.
13. Arthrostylidium maculatum. See Aulonemia
14. Arthrostylidium maxonii. See Rhipidocladum
15. Arthrostylidium pittieri. See Rhipidocladum
16. Arthrostylidium prestoei. See Rhipidocladum
17. Arthrostylidium purpuratum. See Aulonemia
cladum ampliflorum.
amp lcssima.
cladum angustiflorum.
setigera.
bartlettii.
burchellii.
fusa.
dum geminatum.
haenkei.
dum harmonicum.
leptophylla Nees.
Guadua) longifolia.
parui flora.
maxonii.
pittieri.
prestoei.
purpurata.
18.
19.
20.
21.
22.
23.
Arthrostylidium queko (as quexo). See Aulme-
Arthrostylidium racemiflorum. See Rhipidocladum
racemiflorum.
Arthrostylidium spinosum. See Bambusa (Gua-
Arthrostylidium steyermarkii. See Aulonemia
Arthrostylidium subpectinatum. See Aulonemia
Arthrostylidium trinii. See Rhipidocladum parmia
queko.
dua) longifolia.
stey ermarkii.
su bpectinata.
viflorum.
ArundinarM Michaux
FICURIB 7-18
Arundinaria Michaux, 1803, 1:73.-McClure, 1957:200.
Miegia Persoon, 1805, 1:lOl.-McClure, 1957:205.
Ludolfia Willdenow, 1808:320.-McClure, 1957:204.
Macronax Rafinesque, 1808:353.-McClure, 1957:204.
Triglossum Fischer, 1812:vi.-McClure, 1957:210.
Pleioblastus Nakai, 1925: 145.-McClure, 1957:207.
Nipponobambusa Muroi, 1940:89.-McClure, 1960:194.
Nipponocalamus Nakai, 1942:350; 1951:326.-McClure, 1957:
206.
The literature that deals with Arundinaria-even
when the genus is viewed solely in terms of the few
New World taxa that I recognize as falling within
its natural boundaries-comprises many individual
items. Repeated consultation and attentive study
of this literature has convinced me that it neither
contains nor conveys the documented information
needed for the construction of a clear image of
the New World components and the natural
boundaries of the genus. For this reason, instead
of listing all or any major part of it here, I have
selected-and reserved for notice in the text-those
items that appear to have a clear pertinence to the
thread of the present account and to be least likely
to confuse the reader.
Plants of diffuse habit (both diffuse and pluricespitose
in some species); all axes unarmed. Rhizomes
leptomorph. Culms self-supporting, either
remaining solitary, or tillering from subterranean
buds-with or without the intercalation of a metamorph
axis, the internodes fistular, either terete
throughout or more or less markedly sulcateusually
so for only a short distance upward from
the locus of insertion of a bud or a branch complement,
rarely (as occasionally seen in plants of
SMITHSONIAN CONTRIBUTIONS TO BOTANY
Awndinaria gigantea ssp. gigantea) lightly so all
the way from one node to the next. Primary
branch buds at culm nodes solitary, each containing
but a single initial primordium. Branch complement
at midculm nodes in some species facultativerestricted
monoclade, (rarely, either temporarily or
permanently, lacking throughout) typically unrestricted
monoclade and always of restricted insertion,
with the primary member dominant. Leaves
(blades of leaf sheaths) petiolate, with transverse
veinlets clearly visible on both surfaces initially,
but sometimes becoming obscure in old leaves.
Inflorescences semelauctant, generally of open
racemose or paniculate branching, sometimes assuming
both of these forms, or even reduced to a
single spikelet, all in the same plant. Rachis either
deliquescent or excurrent, its branches typically not
prophyllate, the lowermost primary ones always,
the upper ones commonly, subtended by either a
small or rudimentary bract, or a line of hairs.
Transitional glumes at the base of each spikelet,
usually 2 and empty (the first one sometimes
lacking), rarely 3, the third (conventionally called
a sterile lemma) sometimes empty, more commonly
subtending a depauperate flower. Spikelets comprising
several to many perfect florets, and terminating
acuminately in progressively depauperate sterile
florets or empty anthecia. Rachilla segments potentionally
disarticulating just below the locus of
insertion of each fertile lemma. Lemma (when
subtending a functional flower) fully embracing
the palea only basally at maturity. Palea broadly
sulcate and 2-keeled dorsally, the margins not at
all or only slightly and partially overlapping.
Lodicules 3, the anterior two typically more or less
strongly asymmetrical and paired, the posterior one
symmetrical and usually smaller (at least either
shorter or narrower) than the anterior two. Stamens
typically 3, the filaments filiform, free.
Stigmas 3. Mature fruit an oblong, sulcate, glabrous
caryopsis, terminating apically in the persistent base
of the style; the pericap thin, pergamineous or
coriaceous, in some species appreciably thickened
at the apex of the fruit; the contours of the basal
embryo clearly revealed as a rule in the dry fruit
by the embryotegium.
ETYMoLoGY.-The name Arundinaria is derived
from the Latin, urundo (arundin-), reed or cane,
and the suffix, aria, that signifies belonging to.
From 1829 (Kunth, 1829:137) at least until 1861
(Bentham, 1861 :433) predominant usage attributed
the authorship of the genus Arundinaria to L. C.
Richard. In the absence of positive evidence to
the contrary, however, Michaux is now generally
recognized as the author of it.
TYPE-spECIEs.-Arundinaria gigantea (Walter)
Muhlenberg (1813).
As basionym, Arundo gigantea Walter (1788),
takes precedence over Festuca grandiflora Lamarck
(1791), and Arundinaria macrosperma Michaux
(1803) at the species level, since both Lamarck’s
and Michaux’s species are here included in a
comprehensive circumscription of Arundinaria gigantea
(ICBN 1966:Art. 57).
RELATIONSHIPS.-From the polymorphic neotropical
genus, Arthrostylidium (q.v.), numerous species
of which have at one time or another been misplaced
in Arundinaria, taxa characteristic of Arundinaria
are distinguishable by the following
invariable combinations of morphological features:
leptomorph rhizomes, diffuse or pluricespitose
clump habit, and ternate stigmas. Moreover, the
respective geographical distributions of the two
genera are different and disjunct, and are correlated
with distinctive ecological adaptations. As far as
known, plants of the genus Arthrostylidium are
confined to frost-free sites, while the genus Arundinaria
comprises plants that are frost-hardy.
Included by some authors in Arundinaria, the
recognized components of the Old World genus
Thamnocalamus are set off from this genus by the
combination of pachymorph rhizomes, unicespitose
clump habit, and well-developed, sheath-like bracts
by which the lower branches of each inflorescence
are subtended.
Of the known African species at present allocated
to Arundinaria, A. alpina K. Schumann (at least) is
excluded from this genus by its pachymorph rhizomes,
a unicespitose clump habit, and binate
stigmas; cf Yushania sg. Otatea, p. 116.
DISTRIBuTION.-The known New World components
of the genus Arundinaria constitute the polymorphic
type-species whose natural distribution is
limited to continental United States. They are
represented principally in North American, British,
and European herbaria by specimens gathered in
Alabama, Arkansas, Delaware, Florida, Illinois,
Indiana, Kentucky, Louisiana, Maryland, MissisNUMBER
9 23
sippi, Missouri, North Carolina, Ohio, Oklahoma,
South Carolina, Tennessee, Texas, Virginia, and
West Virginia.
A map of southeastern United States showing,
by counties, the observed incidence of stands of
native Arundinaria-under the name Arundinaria
tecta applied in a comprehensive sense-is presented
by West (1935:258, fig. 1). Owing to the difficulties
involved in reconciling incongruities and inconsistencies
between various published taxonomic
usages-difficulties augmented by the fragmentary
nature of most of the available preserved specimens-
it is not yet possible to construct authentic
maps of the precise distribution of stands of the
respective subspecific entities herein recognized. A
general idea of the pattern of their distribution,
however, may be derived from Gilly's map (1943,
fig. 1) by assuming a rough (but not precise) correspondence
between Gilly's "Mississippi-type'' and
my A. gigantea ssp. gigantea; between Gilly's "Atlantic-
type" and my subspecies tecta; and between
Gilly's "Intermediates" and my subspecies macrosperma
(cf. Figure 16).
Known Old World taxa that I recognize as
members of the genus Arundinaria are found in
China proper, Hainan Island, northern India,
Japan, Korea, Madagascar, Nepal, the Ryukyu
Archipelago, Sikkim, Taiwan, Tibet, and northern
Vietnam.
THE NEW WORLD COMPONENT OF THE GENUS
Awndinaria.-The only known New World bamboos
embraced by my circumscription of the genus
Arundinaria are those included in its polymorphic
type-species, Arundinaria gigantea (Walter) Muhlenberg,
sensu lato. The development of an improved
idea of the geographical boundaries of the
genus and its known content requires, therefore,
a critical examination of taxa that have hitherto
been assigned to it from the bamboo floras of the
Old World, as well as those of the New.
During the interval since Arundinaria was first
described and given monotypic status by Michaux
in 1803, the number of species assigned to the
genus increased spectacularly. The inventory of
names linked to it in the taxonomic literature
grew from the initially solitary entry, Arundinaria
macrosperma Michaux, to a total of about 376,
counting only binomials. The inclusion of trinomials
swells the number of its published nomenclatural
diversities to about 482. Of the New World
component of this total, I have excluded the taxa
represented by 50-odd names listed on page 36
et seq.
As a result of interim studies and revised judgments
published by other authors, many of the
Old World species originally (or at one time)
incorporated in the genus Arundinaria have subsequently
been made either the respective types,
or components, of the following genera: Brachystachyum,
Chimonobambusa, Indocalamus, Neosasamorphia,
Oreocalamus, Pseudosasa, Sasa,
Sasaella, Sasamorpha, Semiarundinaria, Sinarundinaria,
Sinobambusa, Thamnocalamus, and Yushania.
All of the bamboos that have since been
allocated to these genera are endemic to Old World
areas.
Annotated Checklist of New World Elements
Included in the Genus Arundinaria
(as herein defined)
Arundinaria gigantea (Walter) Muhlenberg, 1813:
14. Basionym: Arundo gigantea Walter, 1788:81.
Arundo tecta Walter, 1788:81.
"Festuca grandiflora Lamarck, 1791:191.
Arundinaria macrosperma Michaux, 1803, I:74.
* Triglossum bambusinum Fischer, 1812:6.
Arundinaria tecta (Walter) Muhlenberg, 1813: 14.
"Arundinaria bambusina (Fischer) Trinius, 1820:
"Miegia pumila Nuttall, 1837: 149 [illegitimate
"Arundinaria tecta p pumila Ruprecht, 1839:22.
"Arundinaria tecta y distachya Ruprecht, 1839:22.
*Awndinaria tecta 8? colorata Ruprecht, 1839:22.
"Arundinaria macrosperma a arborescens Munro,
"Arundinaria macrosperma p suffruticosus Munro,
"Armdinaria macrosperma tecta Wood, 1871:
"A,.undinaria gigantea tecta (Walter) Scribner, in
"Armdinaria tecta var. decidua Beadle, in Bailey,
"Given the present unsatisfactory state of their
description, their documentation, and their typification,
vis-8-vis the highly polymorphic character
of the populations here included under Arundi-
97.
name].
1868: 15.
1868:15.
404.
Kearney, 1893: 478.
1914:446.
24 SMITHSONIAN CONTRIBUTIONS TO BOTANY
naria gigantea, sensu lato, it is my judgment that
proposals for a more refined formal disposition of
the taxa represented by these names should be
deferred until comprehensive studies on a broad
disciplinary spectrum in field and laboratory have
improved present perspectives.
Proposed Taxonomic Revision
of the Type-species of Arundinaria
The relegation by Hitchcock (1935:29 [1951:27])
of the name Arundinaria macrosperma Michaux to
synonymy under Arundinaria gigantea (Walter)
Muhlenberg appears to be referable to an earlier
tentative decision expressed in the following paragraph
quoted from page 156 of a paper entitled
“Types of American Grasses” (Hitchcock, 1908).
Arundinaria macrosperma Michaux “Gramen altissimum
ramosum a Virginia ad Floridam & in occidentalibus juxta
fluviis ab Illinoensibus ad ostium Misissipi [sign for undershrub].”
The specimen is fragmentary and one can not be
certain which species of Arundinaria it represents. Michaux
probably included the large and the small canes in one
species. As he described the plants as being very high, we
may retain this name for the tall cane, as is done in our
manuals.
Hitchcock’s disposition of the matter of synonymy
appears to have been influenced by the reiterated
allusion to size-in the trivial name, gigantea, of
one taxon, and in the description of the other,
macrosperma, as “altissima.” In other words, it is
based upon an interpretation of Michaux’s description
of the monotypic genus Arundinaria, and not
upon an interpretation of the lectoholotype by
completing the image of its botanical source (cf.
Rickett and Camp, 1950). At the same time, however,
the taxon represented by Arundinaria tecta
(Walter) Muhlenberg was maintained by Hitchcock
as a distinct species.
In present perspectives it appears that the taxa
represented by the binomials Arundinaria gigantea
(Walter) Muhlenberg, Arundlnaria tecta (Walter)
Muhlenberg, and Arundinaria macrosperma Michaux
embody, respectively, three components of a
polymorphic array of populations (cf. Figures 16,
17). As shown by their purest available typifications,
Arun.dinaria gigantea and Arundinaria tecta are
clearly set off from each other by distinctive combinations
of strongly contrasting morphological and
ontogenetic features (Figure 15). The lectotype
and isotype of the taxon Arundinaria macrosperma
are seen as representing a plant that falls within
the series of variants (putative nothomorphs) that
form a cline bridging the morphological and ontogenetic
divergency that exists between Arundinaria
gigantea and Arundinaria tecta (Figures 17, 18).
Circumstantial evidence of the biological relationships
of these three taxa supports the proposal
that they be treated taxonomically as members of
a single polymorphic species.
When the circumscription of Arundinaria macrosperma
Michaux is interpreted as including the
type of Arundo gigantea Walter, the adoption of
the name Arundinaria gigantea (Walter) Muhlenberg
in place of Arundinaria macrosperma Michaux
for the type-species of the genus Arundinaria
(Hitchcock, 1935:29 [1951:27]) has the sanction of
the International Code of Botanical Nomenclature
(see Lanjouw et al, ed., 1966: Art. 57). This precedent
determines the binomial to be adopted when
the three taxa now under discussion are treated as
subspecies under one specific name.
In present perspectives, any specimen that bears
a recorded origin indicating nativity to continental
United States, and that shows features of both
subspecies gigantea and subspecies tecta-even
though it is otherwise incomplete-may be determined
as subspecies macrosperma.
The subspecific identification placed on incomplete
specimens that show only features characteristic
of either subspecies gigantea or subspecies tecta
should be followed by a query (?), since the name
of the subspecific taxon to which the character of
the missing structure (s) would carry the identification,
by way of the key, must remain uncertain
in such cases. Where, on account of the incompleteness
of a specimen, uncertainties arise as to its
correct subspecific disposition, it may with propriety
be filed tentatively as Arundinaria gigantea, sensu
lato. However, the possibility that more intensive
field studies in the United States may bring to light
native plants specifically distinct from Arundinaria
gigantea, as this taxon is here circumscribed, should
not be ignored.
Concerning the Typification
of Arundo gigantea, Arundo tecta, and
Arundin,aria macrosperma
Of all of the many previously published treatises
%UMBER 9
Key to the Herein Recognized Subspecies of Arundinuria gigantea, sensu lato
la. Rhizomes without air canals: typically of steadily diageotropic growth, i.e., normally not
turning up at the apex to form a culm. Plants of diffuse habit, the culms normally
solitary and not tillering. Midculm sheaths each shorter than the corresponding internode;
deciduous. Primary axis of midculm branch complements not constricted basally,
lacking a “neck” (see glossary). The lowest bud (or secondary branch) of midculm
pleioclade) branch complements inserted but a short distance above the base of the
first internode of the primary branch and (in contrast to those inserted at its succeeding
nodes) not subtended by a conventional sheath. The primary and secondary axes of
midculm branch complements relatively short and slender, all appressed basally and
then curving strongly away from the culm. Culms with their leafy branch complements
fully developed present a narrowly linear profile. Leaf blades pubescent on the abaxial
surface; subglabrous on the adaxial. Lemmas dorsally hirsute with antrorse hairs and
minutely hispidulous or scaberulous; the exposed exterior surface pale green (not tinted
with wine) and more or less conspicuously glaucous when fresh; transverse veinlets
clearly manifest externally. Lodicules typically subopaque, fringed with marginal cilia,
and suppfied with numerous vascular traces. Pistils lacking the uncinate rudimentary
stylar branch that is regularly present on the pistil in the neotype of subspecies tecta.
Fruits narrowly conical apically and terminating in the persistent base of the style
(Figures 7-11) ........................................ ........................... 1. subspecies gigantea
lb. Rhizomes with air canals forming a continuous and unbroken cylinder; apical growth
commonly diageotropic (horizontal) for only a short distance, then directed upward
to form a culm. Plants of diffuse habit (culms arising from lateral buds of the rhizomes
normally remaining solitary) and also pluricespitose as a result of the tillering of culms
(principally those that are apical to rhizomes). Midculm sheaths each longer than the
corresponding internode: not deciduous. Primary axis of midculm branch complements
constricted basally into an obconical “neck” consisting of usually 2 or 3 very short
internodes with budless sheath-bearing nodes (Figure 12~) , followed by more elongated
internodes each bearing basally a secondary member of the branch complement (or a
bud) subtended by a conventional sheath. The primary and secondary axes of midculm
branch complements elongate, appressed basally, then ascending and, when fully developed,
curving broadly away from the culm. Culms with their leafy branch complements
fully developed present a lanceolate profile. Leaf blades densely pubescent on
both surfaces. Lemmas dorsally glabrous or nearly so, the exposed exterior surface
tinted wtih wine, and not glaucous when fresh; transverse veinlets not at all manifest
or only barely perceptible. Lodicules thin, transparent, with glabrous margins and
vascular traces weakly to scarcely manifest. Pistils all showing the unciform rudiment
that becomes a conspicuous feature in mature fruits. Fruit dome-shaped apically,
terminating in a short beak, and regularly bearing (just below the base of the beak,
and directly above the tip of the sulcus) a usually more or less strongly unciform rudiment-
apparently representing a supernumerary style or stylar branch (Figures 12-14)
................................................................................................. 2. subspecies tecta, new status
lc. Rhizomes with air canals either present or absent; when present, either continuous or
discontinuous; other individual morphological and ontogenetic features appearing in
diverse recombinations of either identical or intermediate expressions of the same contrasting
features that (in characteristic combinations described above) clearly differentiate
subspecies gigantea and subspecies tecta . , , , , . , . , 3 . subspecies macrosperma, new status
26
*Edna Rema Walker (1906) reports and illustrates the occurrence of rudimentary stylebranches
in a number of species of grasses, including Oryza sativa. See also Arber, 1934, fig.
88. I have not found any published notice of the occurrence of this feature in any bamboos.
that have been concerned with the clarification of Munro’s monograph (1868)-Hitchcock‘s publicathe
taxonomy of the New’World components of the tions (1905, 1908) proposing lectotypes for Arundo
genus Arundinaria-including not only local floras gigantea Walter and Arundin.aria macrosperma
and manuals, but such taxonomic treatises as Michaux, are the only ones that are recognizable
26 SMITHSONIAN CONTRIBUTIONS TO BOTANY
as representing a disciplined conformity to the
spirit and the letter of the pertinent references in
the International Code of Botanical Nomenclature.
In the selection of a neotype for Arundo tecta,
and living plants (hypotypes, sensu Frizzel, 1933:
653) to supplement the original images of subspecies
gigantea, tecta, and macrosperma, I have followed
the stipulations of the International Code of
Botanical Nomenclature (1966) as set forth in note
3 under Art. 7; and in paragraphs 3, 4, and 5
under the heading, Guide for the determination
of types (Lanjouw, et al., ed., 1966:71). The protolog
pertaining to each of these three taxa lacks a precise
indication of type-locality. However, I believe that
the plants selected present details that authentically
supplement the sketchy images of these taxa
provided by their respective protologs. According
to Ewan (1969:198) “Thomas Walter’s new species
in his Flora Caroliniana (1788) are generally assumed
to be descriptions of plants that grew about
his plantation at the mouth of the Santee River.
But we must remember that John Fraser was
Walter’s advocate and ‘well spring in the wilderness’
who brought plants for Walter to study and
describe.”
In the present limited state of our knowledge of
the component living plants, and in the restricted
perspective afforded by their fragmentary representation
in extant herbarium specimens, the apparently
clinal nature (composition) of presumably
hybrid populations from which these three taxa
have been selected renders their more refined
systematic treatment infeasible as yet (cf Bocher,
1963: 11-12, 1967:258; Stebbins, 1940; Heiser, 1949;
and Allan, 1949).
1. Arundinaria gigantea ssp. gigantea
FIGURES 7-11, 1 5 ~ , 1 6 ~
HoLoTY PE.-Walter Herbarium (BM) ; Walter
s.n. The specimen is a mere fragment of a sterile,
leafy axis-apparently the tip of a young culmwith
the distal branch complement in an early
stage of development. A ticket attached to the
specimen bears the name, Arundo gigantea, but no
other annotation.
The feature-combination characteristic of Arundinaria
gigantea ssp. gigantea-as elaborated in the
descriptive key-is based on the following specimens:
Vegetative state: McClure (LU 15140) collected
18 April 1928 from a homogeneous,
apparently natural stand growing in and adjacent
to a swamp at the side of a small stream on the
Old Shaker Farm, near Lebanon, Warren County,
Ohio; and McClure 21321 (US) yielded by plants
from the same source maintained under cultivation
as MBG 2784. Flowering state: Winteringer
7214 (US) “Low ground west of Sandusky, Alexander
Co., Illinois. 7 July, 1951.” Fruiting state:
Bain 117 (US) “River banks, Tennessee. 1892;”
and Bush I98 (US) “Common in woods. Eagle
Rock, Missouri. May 31, 1898.”
Swallen 6717 (US 2078084-2078089)-i11ustrated
in Hitchcock, 1935 [1951, fig. I], as Arundinaria
giganten (Walter) Muhlenberg-and D. M. Moore,
s.n. (US 2206426-2206430) are relatively complete
specimens that appear at first sight to represent typical
A. giganten ssp. gigantea. On closer examination,
however, both of these collections show evidence
of minor introgression of genetic influence from
subspecies tecta. This is expressed inconspicuously
in such tectoid features as a noticeable vinous
tinting of the lemma and the palea, and the development,
sooner or later, of negative geotropism
in each rhizome, whereupon it turns upward apically
to form a culm from its growing point. Strict
application of the formula provided by the descriptive
key identifies the plants represented by
these two specimens as members of the clinal
series embraced by A. gigantea ssp. macrosperma.
They may be visualized, however, as occupying
positions morphologically and genetically (and
therefore taxonomically) very near to A. gigantea
ssp. gigantea.
2. Arundinaria gigantea ssp. tecta (Walter)
McClure, new status
FIGURES 12-14, 1 5 ~ , 16c
Arundinaria tecta (Walter) Muhlenberg, 1813:14: 1817:191.
Arundo tecta Walter, 1788:81.
FIGURE 7.-Arundinaria gigantea (Walter) Muhlenberg ssp.
gigantea. A, Basal section of mature culm with section of
rhizome attached, x 0.6; B, middle section of mature culm
at end of first season, x 0.6; c, base of midculm branch
complement, x 2.4; D, apex of leaf sheath with petiole and
base of leaf blade, x ca. 6; E, cross-section of rhizome, showing
lack of air canals, x ca. 12. All drawings based on
McClure Bamboo Garden specimen no. 2784 (US).

28 SMITHSONIAN CONTRIBUTIONS TO BOTANY
NEOTYPE.-Walter’s type has not been found.
However, the living plant to which his brief characterization
led me, has presented both the vegetative
and the reproductive stages of its ontogeny, and
many facets of its potential morphological diversity.
The feature-combination characteristic of A . gigantea
ssp. tecta-as elaborated in the descriptive keyis
based on McClure 22000 (US), a series of specimens
yielded by plants in a natural stand growing
in and adjacent to a swamp lying between Stony
Run Creek and the Pennsylvania Railroad, near
Friendship International Airport, Anne Arundel
County, Maryland, and by plants from the same
source maintained under cultivation, as MBG 2762.
They were collected over a period of years both
at the original site, and from plants taken from
the same source and maintained under cultivation,
since 1955. ‘This series of specimens constitutes the
exclusive documentation of the characterization of
subspecies tecta presented in the descriptive key.
Plants growing in the wooded swamp reach a
maximum height of about 2.5 m. These have
remained in a sterile (vegetative) state ever since
my observations were initiated in 1952. Other sterile
plants of the same taxon growing in upland clay
soil on the opposite side of the railroad (until
they were destroyed by earth fill) reached a maximum
height of about one meter. Plants growing
along the edge of the swamp push their rhizomes
into the original gravel ballast along the railroad,
where they produce culms 0.5 to 1.0 m tall. Here
and there among these latter, little patches of
culms in a depauperate condition may be found in
flower every year. However, I have never found
fruits produced in that situation. The reproductive
organs in every floret are always more or less completely
destroyed by the larvae of insects that pass
this stage of their life cycle within the anthecia.
Two plants among those maintained under cultivation
in my garden (about 20 miles from the
natural stand) flowered and fruited freely. The
production of fruits by these cultivated plants is
a circumstance attributed to the absence of the parasitic
insect at the latter site. In all observed cases
of flowering, whether in wild plants or in those
grown under cultivation, the culms that flowered
died in the same season, along with the rhizomes
that remained attached to them.
3. Awndinaria gigantea ssp. macrosperma
(Michau) McClure, new status
FIGURE 16a
Arundinaria macrosperma Michaux, 1803, I:74 [as to type].
TYPE (elig. Hitchcock, 1908: 156).- (P) “Herbier
de 1’Amerique Septentrionale d’Andre Michaux.”
“Gramen altissimum ramosum 5 [Michaux’s sym.
bol for undershrub]. Hab. a Virginia ad Flondam
et in occidentalibus [a phrase is here crossed out
and illegible] juxta fluvios ab Illinoensibus ad
ostium Misissipi -f [Michaux’s symbol for perennial
plant.]” The specimen is a leafy flowering
branch with leaf blades and florets detached.
ISOTYPE (nunc elig.).-A specimen that quite
clearly represents a duplicate from the same Michaux
collection as that which yielded the lectotype
is preserved at the Paris Herbarium. A printed
label on the sheet bears the following information:
MUSEUM D’HISTOIRE NATUWLE DE PARIS
herbier d’Antoine Laurent de Jussieu.
Donneau Museum par les enfants d’Andrien
de Jussieu en 1857. Catal. no. 2881.
The label (made in Paris by the elder Jussieu)
bears the following annotations:
Arundo gigantea Walter
M. Michaux le dit tres different de 1’Arundo.
il a tout le port du Nastus de l’inde et paroit
etre la meme plante, mais M. Michaux dit qu’il n’a
que 3 etamines et pour I’ordinaire 3 styles ou
peut-etre un style divisC profondement en 3. Caroline
et Rives du Mississipi jusqu’a 1’Ohio. Donne par M.
Michaux 1797.
Arundinaria Mich. en Dew.
Another specimen, preserved in the Richard
Herbarium (P) and labeled “ex Herb. E. Drake”
supplements slightly the fragmentary representation
(and the diversity) of the feature-combinations that
appear in the image of the taxon Awndinaria
macrosperma Michaux provided by the lectotype
and the isotype. The species label bears the inscrip-
FIGURE 8.-Arundinaria gigantea (Walter) Muhlenberg ssp.
gigantea. A, Midculm section of young culm, x 0.6; B, tip
section of young culm, x 0.6; c, culm sheath from midculm
node, with blade not reflexed, x 3; D, culm sheath from
higher level, with blade reflexed, x 3; E, Walter’s type of
Arundo gigantea (from British Museum photo), x ca. 0.4.
All drawings based on McClure Bamboo Garden specimen
no. 2784 (US).
NUMBER 9 29
SMITHSONIAN CONTRIBUTIONS TO BOTANY
tion “Arundinaria macrosperma [ Michaux] Caroline.
m.” With the possible exception of the word
“macrosperma”-which appears to have been inserted
by a different hand-this inscription is in the
handwriting of Michaux. The specimen gives the
plant from which it came a recognizable relationship
to subspecies tecta.
A sterile specimen, McClure 21320 (US), and a
fruiting specimen, McCZure 21664 (US), represent
a plant taken from the wild near the Ogeechee
River, about 12 miles south of Savannah, Georgia,
in the vegetative state, by David Bisset. I received
a division of it in 1942 and maintained it under
cultivation (as MBG 2803) until August 1948, when
it flowered, bore fruits and died. The inflorescences
match those of the lectotype and isotype of Arundinaria
macrosperma. The fruits of this plant are
similar in shape to the fruits of the neotype of
subspecies tecta but differ in their much larger
average size and in the infrequent incidence of the
unciform rudimentary style branch.
McClure 22018 (US) represents a plant secured
from the wild near Travelers Rest, South Carolina,
through the good offices of Robert A. Young, and
maintained under cultivation (as MBG 2782) from
1948 to the present. The plant, which has remained
steadily in the vegetative state, combines expressions
of morphological features characteristic, respectively,
of subspecies gigantea and subspecies
tecta, and some that are intermediate. The rhizomes
(as in ssp gigantea) lack air canals, but (as in ssp.
tecta) turn up apically to form a culm. The culms
(as in ssp. tecta) proliferate from subterranean
buds to form either culms or rhizomes. The primary
axis of midculm branch complements (as in
ssp. gigantea) lacks a basal “neck” but the component
members of the complement (as in ssp.
tecta) are more elongate and not so strongly curved
away from the culm. The leaf blades are, on the
average, larger than those of either typical subspecies
gigantea or typical subspecies tecta.
The revised taxonomic dispositions embodied in
the nomenclature proposed here for these three
subspecific components requires a postscript to my
paper (McClure, 1963a) entitled “A New Feature
in Bamboo Rhizome Anatomy.” The plant therein
called Arundinaria gigantea (Walter) Muhlenberg,
whose rhizomes lack air canals (cf. Figure 7 ~ ) , is
herein classified (and described in the key) under
the name Arundinaria gigantea ssp. gigantea. Its
documenting number is incorrectly given (McClure,
1963) as MBG 2792; however, the origin of the
plant is correctly given there; its correct number is
MBG 2784. The plant (MBG 2762) therein called
Arundinaria tecta (Walter) Muhlenberg, whose
rhizomes show continuous air canals (cf. Figure
1 3 ~ ) is herein classified (and described in the key)
as Arundinaria gigantea ssp. tecta. The unnamed
specimens (Biltmore 1405, and Radford and Wood
6879-A) cited therein and described as having air
canals discontinuous in the rhizomes, are now
recognized as falling within my circumscription of
Arundinaria gigantea ssp. macrosperma.
In the limited perspective provided by their
fragmentary representation in extant specimens
of the conventional sort, these three subspecies are
usually not clearly differentiated from each other
in herbaria, For the identification of complete
plants or complete specimens, however, the descriptive
key is believed to be adequate and reliable.
As they appear in nature, and as they are characterized
in the descriptive key (p. 25), plants of
subspecies macrosperma probably comprise the
numerically dominant element of Arundinaria gigantea,
sensu lato. This subspecies is here interpreted
(circumscribed) as embracing plants (all
apparently of hybrid origin) that embody, respectively,
diverse recombinations of either identical or
intermediate expressions of the same contrasting
features that (in characteristic combinations)
clearly differentiate subspecies gigantea and subspecies
tecta, the putative genetical parents.
The union, under one specific name, of the
morphologically and genotypically distinct but obviously
(at least ostensibly) interbreeding members
of sympatric populations has the sanction of precedent.
In the present case it has other defenses as
well. It affords a framework for facilitating the
practical disposition (identification and filing) of
the fragmentary specimens of this species that
FIGURE 9.--Arundinaria gigantea (Walter) Muhlenberg ssp.
gigantea. A, Young leafy sterile culm with section of rhizome
attached, x 0.6; B, young leafless flowering culm, x 0.6; c,
whole sheath on midculm node of young culm shoot, x 0.6;
D, apex and blade of culm sheath from upper node of young
culm, showing reflexed sheath blade, x ca. 1.8. All drawings
based on D. M. Moore s.n., Arkansas, USA, 2 August 1947
(US) ’

32 SMITHSONIAN CONTRIBUTIONS TO BOTANY
predominate numerically in herbaria. Assigning
subspecific status to the three components portrayed
in the descriptive key provides a means of
giving greater depth of focus to floristic and ecological
field studies such as those carried out by West
(1935). For example, West’s identifications and
descriptions of the geographic areas within which
the existence of “canebrakes” is recorded by him
are extremely valuable. Apparently, however, no
effective effort was made, either to document the
observed “ecological forms” by means of annotated
voucher specimens or to correlate their distribution
with that of the nine different “habitats” (ecological
complexes) and the eight different floristic associations
(“forest types”) within which they were
observed to occur. Hopefully, such ecological and
floristic emphasis will also encourage collectors to
cultivate an awarneess of the potential taxonomic
importance of morphological and ontogenetic diversities
that await documentation, and to exercise
the discipline necessary to give due attention to
them in preparing specimens, whether these are
intended for identification only or for permanent
preservation.
West (1935:255) maintained that the Arundinaria
gigantea and Arundinaria tecta of the manuals
are indistinguishable. “It is believed that the two
so-called species of the manuals are in reality
ecological forms of the same species , . . when both
types are planted side by side in the greenhouse
or in the open, they soon become indistinguishable.”
I have had under observation for at least ten
years, both in the wild and in cultivation under
essentially identical conditions, plants I recognize
as representing Arundinaria gigantea, Arundinaria
tecta, and Arundinaria macrosperma, respectively.
They have retained their distinguishing features as
described in the key and as documented by voucher
specimens cited here under the corresponding subspecies.
Plants from different natural stands of either
subspecies gigantea or subspecies tecta rarely match
each other in all details of their ontogeny and their
gross morphology. The persistence of observed
minor differences between a number of variants of
both subspecies gigantea and subspecies tecta when
the plants were grown for a number of years under
essentially identical conditions appears as evidence
that the genotypic heterogeneity that characterizes
hybrid populations may embrace the whole content
of this polymorphic species-lightly so in the most
divergent taxa (ssp. gigantea and ssp tecta) and
more profoundly so in the diverse components of
subspecies macrosperma that form a cline between
the first two subspecies.
Observations Concerning the
Taxonomic Utility of Certain Structures
and Their Contrasting Features
I find that certain features that have been used
by others to differentiate taxa identified, respectively,
as Arundinaria gigantea and A. tecta are
taxonomically undependable at both the species
level and the subspecies level.
MAXIMUM CULM HEIGHT.-The impression given
by many local floras and incidental treatises is that
plants of two recognizable bamboo taxa native to the
United States may be distinguished from each other
by the maximum height attained by their culms.
Maximum culm heights given for plants referred
to in the literature-either as gigantea or as macrosperma-
range from 7.6 to 12 meters (Nuttall, 1837:
149; Mitford, 1896:165; Hitchcock, 1936: 19 [1951:
271) and those given for plants-referred to either
as tecta or as macrosperma range up to 4.6 meters
(Nuttall, 1818, I:39; Mohr, 1901: 103). However,
I have not found published culm heights above 5.5
meters documented in any case by reference to
field notes accompanied by voucher specimens.
Among the field notes accompanying specimens of
more than 300 collections examined personally, I
found less than a dozen that bear an indication of
the height of the culms of the plant from which
the specimen was taken. And in no case have I
found field notes recording a maximum height
greater than 5.5 meters, regardless of how the
specimen was identified.
The existence today of canebrakes containing
culms at least 7.6 meters tall is not in doubt. During
FIGURE lO.--drundinaria gigantea (Walter) Muhlenberg ssp.
gigantea. A, Leafy culm tip, x 0.6; B, branch complement
from node high on culm, with leafy and leafless flowering
branches (with many spikelets shattered), x 0.6; c, base of
midculm branch complement, x 2.4; D, apex of leaf sheath
with petiole and base of leaf blade, x ca. 4.8. All drawings
based on D. M. Moore s.n., Arkansas, USA, 2 August 1947
(US).

34 SMITHSONIAN CONTRIBUTIONS TO BOTANY
the second world war I saw in a hardware store
in Savannah, Georgia, 7.6-meter-long culms of a
native bamboo on sale as fishing poles. I also have
a photograph taken in 1905 or 1906 which shows
a stand of native bamboo in Louisiana, the culms
of which are estimated (on the scale of a man on
horseback included in the picture) to be at least
7.6 meters tall. In the absence, however, of documentation
by means of adequate voucher specimens,
it is impossible to assess the taxonomic
pertinence of these published and unpublished
observations. In the following paragraph C. A.
Brown (1929:317) discusses what he considers to
be the ecological significance of differences in culm
size observed in different stands-all of which he
includes in a comprehensive “Arundinaria tecta.”
One might inquire as to the reason for the difference in
size between the small woody canes the size of a lead pencil
and only a few feet [ca. lm] tall and the large woody canes
one to one and a half inches [2.5-3.8 cm] in diameter and
up to thirty feet [9m] tall. It has been noticed that there
are many patches of this small type. The only explanation
that can be offered is that there must be some difference in
soil or moisture conditions which hinders the growth of
these plants.
West(1935:255 and fig. 1) follows Brown in recognizing
only one genotype, in adopting for it the
name Arundinaria tecta (in a comprehensive
sense), and in suggesting that plants producing
tall culms owe their superior stature to a phenotypic
response to favorable environmental conditions,
especially those conditions that are related
to edaphic factors. It appears that plants with very
tall culms (illustrated by Harper 1928, fig. 18) have
been reported only from the Gulf States. The answer
to the question “Does culm size have any
taxonomic significance?” remains to be discovered
by means of additional field studies, documented
by comprehensive, fully annotated specimens.
There is no significant difference between subspecies
gigantea and subspecies tech with respect to
the observed maximum heights of culms found in
stands of plants of these two taxa which are growing
under similar ecological conditions at the northern
limits of their currently known distribution.
I have had these stands under intermittent observation
for 40 years and 16 years, respectively.
In the light of morphological and ontogenetic evidence
that hybridization has taken place between
plants representing these two taxa, it seems possible-
even probable-that at least some of the plants
with very tall culms may embody what is commonly
known as hybrid vigor.
APPENDAGES OF CULM SHEATHS AND LEAF
SHEATHS.-The morphological expression of the appendages-
ligules (inner and outer), auricles, oral
setae, and blades-apical to culm sheath and leaf
sheaths is often more or less variable even as betlveen
specimens taken from the same plant. This
variability is due, in large part, to ontogenetic factors
related to the age of the whole plant, the age
range as between young and old culms in the same
plant, and the point of origin within the plant and
on the culm from which the specimen was taken
(see hlcclure, 1966b:6, par. 1).
dence, patterns of distribution, and persistence of
trichomes in the bamboos are subject to influence
by so many (generally unrecorded) variables that
it is often difficult to find consistent patterns of
vesture in a given structure, as represented by preserved
specimens. Pertinent variables that are generally
ignored during the collection of bamboo
specimens are age (or stature) of the plant; environment
and relative vigor of the plant; position
on the culm and relative age of the culm from
which a given specimen is taken. Weathering, spontaneous
abscission, or abrasion, may erase trichomes
that were originally present on some structures.
cently (Hitchcock, 1935:29, key) supposedly distinct
differences in the loci of insertion of
inflorescences was commonly offered as a means
for the differentiation of two elements of the New
World component of Arundinariu recognized (but
inadequately characterized) under the names A. gi-
VESTURE OF VEGETATIVE ~TRumuREs.-The inci-
LOCI OF INSERTION OF INFLORESCENCES.-Until re-
FIGURE l l .-Arundinaria gigantea (Walter) Muhlenberg ssp.
gigan tea. A, Spikelet with pedicel and inflorescence branch
subtended by a small bract, x 1.8; B, base of spikelet showing
transitional glumes I and 11 and sterile lemma, x 1.8; c,
transitional glume I (normal) , x 3.6; D, transitional glume
11 (variant), x 3.6; E, transitional glume 1x1 (sterile lemma),
variant, x 3.6; F, floret, x 3.6; G, fertile lemma (normal), x
3.6; H, fertile lemma with stipule-like appendages, x 3.6; I,
stipule-like appendage, X 10.8; J, palea, x 3.6; K, lodicule
complement showing dorsal surface, x 7.2; L, lodicule complement
from another flower showing adaxial surface, x 7.2;
kr, gynoecium and androecium, x 7.2; N, fruit, hilum side,
x 7.2; 0, fruit, embryo side, x 7.2. All drawings based on
D. M. Moore, s.n., Arkansas, USA, 2 August 1947 (US),
NUMBER 9 35
J
36 SMITHSONIAN CONTRIBUTIONS TO BOTANY
gantea (or A. macrosperma) and A. tecta. It appears
that C. A. Brown (1929:316) was the first to
focus attention upon the confusion that may result
from reliance upon this feature as a source of contrasting
characters to support a traditional pattern
of taxonomic segregation. I have observed that this
is one of the features whose expression-in Arundinaria
as well as in many other genera-may
change progressively during successive stages of the
flowering of a plant or may differ as between very
young and very old parts of the plant. Its reliability
for the differentiation of taxa of any category
must be tested by means of exhaustive and sustained
field studies.
TRANSITIONAL GLUMES INCIDENT AT THE BASE OF
EACH SPIKELET.-variable also are the transitional
glumes at the base of the spikelets. Gilly (1943:
301, figs. 2-A, 2-M) illustrated, as characteristic (typical)
of his Atlantic type and his Mississippi type,
empty glumes that are distinguished, respectively,
by size, shape, vesture, marginal ciliation, and venation.
These differences, which really exist, are ostensibly
of genotypic origin. However, I have
found the empty glumes to be not consistently uniform,
in their number or in their morphological
expression, either in duplicates from the collections
cited by Gilly, or in the plants I have chosen as
typical, respectively, of the subspecies I distinguish
under Arundinaria gigantea, sensu lato.
LoDIcuLEs.-Michaux (1803, I:74) gave the genus
Arundinaria but 2 lodicules, which he designated
as “appendices.” He, and those who subsequently
have quoted this number, apparently overlooked
the posterior member of the trio that I have always
found to be present in members of this genus. On
account of its slightly more distal insertion, however,
the posterior lodicule often remains attached
to the palea and hidden inside of it, when a floret
is dissected. Members of the lodicule complement
in Arundinaria (and in other genera as well) are
subject to such a diversity of unpredictable variabilities
that it is often very difficult to discover
features that embody expressions sufficiently reliable
for descriptive purposes.
THE STYLAR AxIs.-In the neotype of Arundinaria
gigantea ssp. tecta, the fusion of two of the
three stylar branches sometimes extends upward to
a point noticeably beyond the level at whidh the
first one became free. The appearance produced
by this deviation from the usual manner in which
branches of the style are disposed may have lent
deceptive support to a description that characterizes
the genus Arundinaria as embracing taxa having
two styles (or stylar branches) and taxa having
three styles (or stylar branches). Published 139
years ago by Nees (1834:478) this incorrect characterization
of Arundinaria is still widely copied in
the current literature. The error appears to have
had its origin in Nees’ allocation to the genus
Arundinaria (where the number of stylar branches
or stigmas is always 3) of several species of other
genera (see below the list entitled New World
species excluded from the genus Arundinaria as
herein defined) where the number of stylar
branches or stigmas is always 2.
The array of morphological, anatomical, and
ontogenetic features whose taxonomic utility I have
explored may not complete the list of potentially
useful ones. It remains to be seen whether comprehensive
studies in anatomy and other disciplines,
correlated with intensive field studies, will improve
the present taxonomic perspectives on this heterogeneous
complex of closely related but distinguishable
entities between which biological (genetic)
isolation apparently has not yet been fully established.
Reeder (1957, fig. 29) illustrated, in cross-section
and sagittal section, the embryo of a fruit from a
plant of bamboo native in Ann Arundel County,
Maryland, documented under MBG 2762, and identified
as Arundinaria tecta. Could (1960:873) reported
for a plant from a stand of bamboo native
in Walker County, Texas, identified as Arundinaria
gigan.tea, a chromosome count of “n equals 24.”
Checklist of New World Species Excluded
from the Genus Arundinaria
(as herein defined)
The apparently general acceptance of the con-
FIGURE 12.-Arundinaria gigantea ssp. tecta (Walter) McClure.
A, Young culm shoot with segment of rhizome, x 0.6; B,
lower part of sterile leafy culm in the second season of its
growth, with solitary leafy branches fully developed, x 0.6;
c, midculm section of culm in its second season of growth,
showing persistent sheaths and developing branch complement,
x 0.6; D, base of unrestricted monoclade midculm
branch complement, x 1.2; E, seedling with anthecium still
attached, x 1.2. All drawings from McClure Bamboo Garden
specimen no. 2762 (US).
NUMBER 9 37
38 SMITHSONIAN CONTRIBUTIONS TO BOTANY
ventional circumscriptions of the genus Arundinaria
that were published after that of Nees (1834)
has postponed the recognition that, of the approximately
fifty distinct neotropical taxa that have at
one time or another been assigned to this genus,
not one falls within its natural boundaries.
1. Arundinaria acuminata. See Yushania acumi-
2. Arundinaria amplissima. See Aulonemia am-
3. Arundinaria aristulata. See Aulonemia aristu-
4. Arundinaria ? attenuata Doell; species sedis
5. Arundinaria burchellii. See Colanthelia bur-
6. Arundinaria cannavieira. See Apoclada canna-
7. Arundinaria capillifolia is Arthrostylidium
8. Arundinaria cubensis is Arthrostylidium cu-
9. Arundinaria ? decalvata Doell; species sedis
nata.
p lissima.
lata.
mihi incertae etiam nunc manet.
chellii.
vieira.
capillifolium Grisebach.
bense Ruprecht.
mihi incertae etiam nunc manet.
10. Arundinaria deflexa. See Aulonemia deflexa.
11. Arundinaria distans. See Colanthelia distans.
12. Arundinaria effusa. See Aulonemia effusa.
13. Arundinaria excelsa. See Arthrosty lidium venezuelae.
14. Arundinaria fimbriata is Arthrostylidium fimbriatum
Grisebach.
15. Arundinaria Pabellata; species sedis mihi incertae
etiam nunc manet.
The name Arundinaria flabellata (Fournierj Mc-
Clure (in Maguire, Wurdack, .et al., 1964:182) is
based on Guadua ? Pabellata Fournier 1881:131.
The plant is known only by nonflowering specimens
(Liebmann 132 [“131”]) really too fragmentary
for confident generic allocation at present.
Since it clearly does not belong in Arundinaria (as
the genus is defined herein) the taxonomic disposition
of this plant awaits the realization of its thorough
study and documentation, preferably at the
type-locali ty.
16. Arundinaria glaziovii. See Aulonemia glazi-
17. Arundinaria glaziovii var. macroblephara. See
ovii.
A u 1 on em ia ramosissima.
18. Arundinaria goyazensis. See Aulonemia goya-
19. Arundinaria haenkei. See Aulonemia haenkei.
20. Arundinaria haitiensis is Arthrostylidium haitiensis
(Pilger) Hitchcock and Chase.
21. Arundinaria herzogiana. See Aulonemia herzogiana.
22. Arundinarict hirtula. See Aulonemia hirtula.
23. Arundinaria humillima. See Aulonemia hu-
24. Arundinaria leptophylla is Chusquea lepto-
25. Arundinaria longiflora is Arthrostylidium
26. Arundinaria longifolia. See Bambusa (Gua-
27. Arundinaria macrostachya. See Colanthelia
28. Arundinaria maculata. See Aulonemia parui-
29. Arundinaria microclada is Chusquea abieti-
30. Arundinaria mirabilis is Glaziophyton mira-
31. Arundinaria mucronata. See Aulonemia aris-
32. Arundinaria multiflora is synonym of Arundi-
33. Arundinaria multispicata is Arthrostylidium
34. Arundinaria obtusata is Arthrostylidium obtu-
35. Arundinaria parviflora. See Rhipidocladum
36. Arundinaria patula. See Aulonemia patula.
37. Arundinaria pinifolia is Chusquea pinifolia
38. Arundinaria pittieri. See Rhipidocladum pit-
39. Arundinaria prestoei. See Rhipidocladum preszensis.
millima.
phylla Nees.
longiflorum Munro.
dua) longifolia.
macrostachya.
flora.
folia Grisebach.
bile Franchet.
t u la ta,
naria trianae, q. v. infra.
multispicatum Pilger.
satum Pilger.
parviporum.
(Nees) Nees.
tieri.
toei.
FIGURE 13.-Arundinaria gigantea ssp. tecta (Walter) McClure.
A, rhizome turned upward apically and terminating in a
simple inflorescence, x 0.6; B, leafy and leafless flowering
branches from tip of rhizome, x 0.6; c, apex of sheath and
reflexed blade from midculm node, x 3; D, apex of leaf
sheath showing petiole and base of blade, x 6; E, crosssection
of rhizome, showing one of the peripheral air canals,
x ca. 12. All drawings based on McClure Bamboo Garden
specimen no. 2762 (US).
NUMBER 9 39
40 SMITHSONIAN CONTRIBUTIONS TO BOTANY
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
Arundinaria pubescens is Arthrostylidium pu-
Arundinaria queko. See Aulonemia queko.
Arundinaria radiata. See Aulonemia radiata.
Arundinaria ramosissima. See Aulonemia ra-
Arundinaria rhizantha. See Colanthelia rhizan-
Arundinaria schomburgkii is Arthrostylidium
Arundinaria setifera. See Aulonemia haenkei.
Arundinaria setigera. See Aulonemia setigera.
Arundinaria simpliciuscula. See Arthrostylid-
Arundinaria sodiroana. See Aulonemia sodi-
Arundinaria standleyi. See Arthrostylidium
Arundinaria trianae. See Aulonemia trianae.
Arundinaria ulei. See Aulonemia ulei.
Arundinaria urbanii. See Arthrostylidium ur-
Arundinaria verticillata. See Rhipidocladum
Arundinaria viscosa. See Aulonemia viscosa.
bescens Ruprecht.
m osiss ima .
tha.
schomburgkii (Bennett) Munro.
ium simpliciusculum.
roana.
venezuelae.
banii.
verticillatum.
Athroostachys Bentham
FIGURES 19, 20
Athroostachys Bentham, in Bentham and Hooker, 1883: 1208,-
Hackel, in Engler and Prantl, 1887:93.-McClure, 1957:200.
Achroostachys [sic] Bentham, 1881: 134 [nomen nudum].-
McClure, 1957:199.
Plants unicespitose, unarmed. Rhizomes pachymorph.
Culms clambering (teste Gardner), the internodes
terete and fistular. Primary branch buds
at culm nodes apparently solitary (not seen). Initial
branch primordia at each midculm node more
than one. Branch complement at a midculm node
unrestricted pleioclade, the initial component axes
(three in Glaziou 13325; illustrated in Figure 1 9 ~ )
each arising from an apparently independent initial
primordium, all inserted at approximately the same
level, subequal or the middle one noticeably dominant.
Leaf blades with transverse veinlets not
manifest externally.
Inflorescences semelauctant, solitary, contracted,
short-peduncled panicles of capitate or subcapitate
form, each inserted at or near the apex of a
usually leafy twig, and subtended by a laminiferous
sheath; the primary branches of the percurrent
rachis inserted distichously in obscurely secund
orientation, very short, not prophyllate, either unbranched
or bearing one or two subequal short
branches; the lowermost primary branches each
subtended by a bract, the uppermost one or more
not so subtended, the branches of higher orders
mostly so subtended; the bracts varying from the
liguliform caudate unawned foliar organ up to 3
cm long that subtends the lowermost primary
branch, down to short, awned scales 0.5 mm long
that subtend branches of the higher order. Transitional
glumes 2, empty, approximate. Spikelets
pedicellate, each containing a single perfect flower,
the rachilla potentially disarticulating just below
the locus of insertion of the fertile lemma, distally
prolonged behind the palea and bearing a depauperate
sterile anthecium. Lemma of perfect florets
fully embracing its palea only basally at maturity.
Palea gaping antically, narrowly sulcate and 2-
keeled dorsally. Lodicules 3, the anterior 2 asymmetrical
and paired, the posterior one symmetrical
and much smaller. Stamens 3, the filaments filiform,
free. Stigmas 2. Mature fruit unknown.
ETY MoLocY .-The name A throostach ys (derived
from the Greek, athroos, crowded, and stachys,
spike) ostensibly alludes to the condensed paniculate
branching of the capitate inflorescence.
TYPE-sPECIES.-AthroostachyS capitata (Hooker)
Bentham (in Bentham and Hooker, 1883:1208).
Merostachys capitata Hooker (1840:pl. 273-274);
Munro (1868:50); Doell (in Martius, 1880:216);
Ekman (1913:64). Chusquea fimbriata Steudel
(1854:338). Cotypes of Merostachys capitata:
Gardner 136 (BM); Tweedie 1324 (BM); type of
Chusquea fimbriata: Riedel s.n. (P).
As a result of my own experience, I can understand
the difficulty evidently encountered by Bentham
in determining the categories to which to
FIGURE 14.-Arundinaria gigantea ssp. tecta (Walter) McClure.
A, Inflorescence of racemose branching, x 0.6; B, infloresence
of paniculate branching, x 0.6; c, immature spikelet, x 1.8;
D, abnormally elongate spikelet, X 1.5; E, floret, X 7.2; F,
fertile lemma, x 7.2; G, palea, x 7.2; H, lodicule complement,
x 15; I, stamen, x 7.2; J, gynoecium x 7.2; K, fruit,
hilum side, x 7.2; L, fruit, embryo side, x 7.2; M, fruit,
embryo side, the embryotegium inconspicuous, X 7.2; N,
fruit, longitudinal section, x 7.2. All drawings based on
McClure Bamboo Garden specimen no. 2762 (US).
NUMBER 9 41
L
U
42 SMITHSONIAN CONTRIBUTIONS TO BOTANY
assign the successive sections of each ultimate
branch of the inflorescence and the several sheathing
structures that precede the fertile lemma in this
bamboo. Having identified as a first empty glume,
the bract that subtends the pedicel of a spikelet,
and having equated the pedicel, so subtended, to
an elongated rachilla, Bentham (in Bentham and
Hooker, 1883: 1209) wrote as follows: “spiculis
. . . revera sessilibus, sed ob rachillam supra
glumam infimam productam pedicellatae apparent.”
That is to say “the spikelets are really
sessile, but they appear to be pedicellate on account
of the elongation of the rachilla beyond the first
glume.”
Doell encountered similar difficulties. In his description
(in Martius, 1880:216) of the type-species
of the genus Athroostachys (under Merostachys
capitata Hooker) Doell not only referred to the
spikelets as sessile, he gave them two rudimentary
empty glumes and two sterile lemmas!
Munro (1868:50) recognized the spikelets of
Merostachys capitata as pedicellate, but he saw the
two empty glumes as sterile lemmas-by convention
called, at that time, “flosculi steriles.”
In his original description of Merostachys capitutu,
Hooker confessed uncertainty as to its generic
affinity, but his recorded observations do not contain
the errors noted above.
Under Merostachys capitata, but without designating
types, Doell (in Martius, 1880:217) described
two variants of Athroostachys capitata, distinguishing
them on the basis of width of leaf blades: (Y
latifolia (presumably representing the typical form
of the species) and @ angustijolia. Munro (1868:
50) intimates that the type of Chusquea frnbriata
Steudel represents a plant with the character of the
latter variant. He adds, however, that it is not
clear whether the sterile, leafy branches of the
Riedel specimen belong with the flowering branches
or not.
RELATIONS HIPS.-^ do not find clear evidence of
a close affinity between Athroostachys and any other
known genus. The strong resemblance of the very
prominent radiate oral setae, borne on its leaf
sheaths (Figure I~D), to those of many species of
Merostachys may have exerted an influence in
favor of the initial allocation of the type-species
of Athroostachys to the genus Merostachys. This
superficial resemblance, however, plus noticeable
similarities in a few technical features of the reproductive
apparatus, does not add up to a clear
indication of close phylogenetic affinity between the
type-species of Athroostachys and members of the
genus Merostachys.
Drsm1suTIoN.-Recorded only from Brazil, the
single known species of Athroostachys has been collected
in the immediate vicinity of the city of Rio
de Janeiro; near Jacarehy in the state of Parani;
and near Villa Maria in the state of Mato Grosso.
Atractantha McClure, new genus
FIGURES 21-23
The defects of conventional procedures prevalently
followed in the documentation of bamboos
for taxonomic purposes are impressively demonstarted
by the extreme incompleteness of the first
two (and the only available) collections representing
this very distinct, hitherto undescribed genus.
Representation, in herbaria, of the individual component
taxa apparently has not been improved in
comprehensiveness during the thirty years that
have passed since the first specimen was brought
to light!
Plantarum habitus non adhuc relatus. Rhizomata
non adhuc visa (?pachymorpha). Culmi in
speciminibus suppetenibus haud satis representati;
internodiis teretibus, vel inanibus vel efistulosis.
Ramorum complementarum (in speciminibus suppetentibus)
axes componentes primum proditi
plures, ad circa libram inserti, unusquisque ex primordio
distinct0 enascens (gemmis intactis primordios
primos includentibus non adhuc visis); uno
alios plus minusve valde dominanti; ramorum com-
FIGURE 15.-Young culms and roots of Arundinaria gigantea
(Walter) Muhlenberg ssp. gigantea (A) and A. gigantea ssp.
tecta (Walter) McClure (B) compared. Aa, Base of culm and
section of rhizome, with roots, x 0.3; Ab, young culm, full
height at age 3 months, showing branches already well
developed, x 0.15; AC, cross-section of root from rhizome,
to show air canals, x 15; Ad, cross-section of root from culm
base, to show air canals, x 15; Ba, base of culm and section
of rhizome, with roots, x 0.6; Bb, young culm, full height,
with branches not yet developed, x 0.3; BC, cross-section of
root from rhizome, to show air canals, x 30; Bd, cross-section
of root from culm base, to show air canals, x 30. Drawings
Aa-Ad based on McClure Bamboo Garden specimen no. 2784
(US), Ba-Bd on McCture Bamboo Garden specimen no.
2762 (US).
NUMBER 9 43
44 SMITHSONIAN CONTRIBUTIONS TO BOTANY
plemento ubi plene evoluto composito-pleioclado.
Foliorum lamina venulis transversis extus vel infirme
vel haud manifestis.
Inflorescentiae iterauctantes, vel capitatae vel
diff usae, earumdem ramorum insertione aut disticha
aut sympodiali, rhachidum ramificationibus
omnibus et bracteatis et prophyllatis, omnis rhachidis
segment0 terminali pro pedicel10 spiculae consistenti.
Glumae transitionales ad basin spicularum
vulgo praeditae hic nullae. Spiculae deciduae, infra
et proxime locum lemmatis insertionis disarticulantes,
unaequaeque flosculam perfectam
solitariam habentes, rhachilla post paleam prolongata
in setam gracilem anthecio rudimentario
terminatam. Lemma fertile maturitate saltem paleam
suam basi tantum circumplectans. Palea apice
bifurcata, dorso sulco angustissimo instructa, antice
leviter imbricata. Lodiculae 3, anteriores 2
sese aequales, plus minusve valde asymmetricae
atque geminatae, posteriore symmetrica et anterioribus
aliquanto minore. Stamina 3, filamentis filiformibus
liberisque. Stigmata typice 2 (atypice et
rarissime 3). Fructus maturus adhuc non inventus.
Plants of as yet unrecorded habit. Rhizomes not
yet seen, ?pachymorph.l Culms not adequately represented
in the available specimens; the internodes
terete, either fistular or efistular. Branch complements
(at least those seen) containing several initial
components, each arising from a distinct initial
primordium.2 The branch complement at midculm
nodes unrestricted pleioclade when fully developed,
one of the primary components more or less
strongly dominating the others, all inserted in close
order at approximately the same level. Leaf blades
with transverse veinlets weakly to not at all manifest,
externally.
‘The pachymorph character of the rhizomes (which are
neither represented in the available specimens nor described
by the collector) may be surmised from the strong resemblance
of the inflated base of the primary axis of a branch
complement (exemplified in the isotype specimen of A. radiata)
to a pachymorph rhizome (see McClure 1966b358,
last paragraph).
‘It has not been p s i b l e to ascertain with certainty from
the available specimens whether the independent primary
primordia incident at each culm node are originally enclosed
in a common prophyllum as one primary bud, or whether
each independent initial primordium corresponds to an independent
primary bud.
FIGURE 16.-Arundinaria gigantea, sensu lato. Semidiagrammatic
sketches showing the extremes and intermediate expressions
of the principal features that characterize the
recognized components of the Arundinaria gigantea complex,
including subspecies gigantea (A) , subspecies macrosperma
(B), and subspecies tecta (c), as to growth habit
(a), rhizome in cross-section (b), spikelet (c). fertile lemma
(d) , and transitional glumes (e) , Subspecies gigantea is
characterized by the combination of expressions Aa, Ab, AC,
Ad, and Ae. Subspecies tecta is characterized by the combination
of expressions ca, cb, cd, and ce of the same
features. Nothomorphs of the presumed hybrid swarms (of
which subspecies gigantea and subspecies tecta are the
putative parents or grandparents) herein given the status
subspecies macrosperma, show diverse combinations OE the
same contrasting expressions of these features, or of intermediate
expressions of them. Some of these intermediate
expressions are illustrated by Ba,- sd, and ~e (transitional
glumes) in the intermediate expression shown in the lectotype
of Arundinaria macrosperma Michaux. The transitional
glumes often vary more or less widely in size and shape
within any given specimen, whether of otherwise typical subspecies
gigantea or typical subspecies tecta, or of the polymorphic
subspecies macrosperma. Moreover, in any given
specimen of any one of the subspecies the lowermost transitional
glume may be lacking entirely, or it may be inserted
at some distance below the second one. A third transitional
glume sometimes appears in the guise of a “sterile lemma.”
Inflorescences iterauctant, of either capitate or
diffuse form, their branching pattern either distichous
or sympodial, each axis the bracteate and
prophyllate rachis of a pseudospikelet, the terminal
segment of each rachis serving as the pedicel of an
abscissile spikelet. Transitional glumes at the base
of each spikelet none.3 Spikelets each made pedicellate
by the segment of the rachis it terminates,
each spikelet typically containing but a single perfect
flower, and disarticulating promptly at maturity
just below the locus of insertion of the
lemma, the rachilla prolonged behind the palea in
a bristle-like structure bearing a minute rudiment
of a sterile anthecium. Lemma in functional florets
embracing its palea only basally at maturity. Palea
apically bifurcate, lightly imbricate antically, dorsally
sulcate, the sulcus very narrow or subtubular.
Lodicules 3, the anterior 2 more or less strongly
asymmetrical, paired and equal, the posterior one
symmetrical and somewhat smaller (at least either
shorter or narrower). Stamens 3, filaments filiform,
The ostensible (approximate) position of the transitional
glumes is here (and in Elytrostachys, q.v.) occupied by what
I now classify as bracts.

46 SMITHSOXIAN CONTRIBUTIONS TO BOTANY
ss
Arundinaria gigantea, sensu Lato
9iyantea ssp. tecta
N o t h o m o r p h s
? (ssp. macrosperma) ~~~ ?
FIGURE 17.-Arundinaria gigantea, sensu lato. Diagrammatic
representation of the putative phylogenetic relationships of
the principal New World components of the genus Arundinaria.
The arrows suggest putative contemporary potential
for further introgression or hybridization. The question
marks and the broken lines raise the question, “Are there
any existing stands of ‘pure’ gigantea and ‘pure’ tecta?”
Arundinaria
in rhizomes \ A i r canals continuous I canals not present
/ in rhizomes
t
Air canals discontinuous
in rhizomes
FIGURE 18.-Arundinaria gigantea, sensu lato. Schematic portrayal
of some of the quantitative and qualitative relationships
visualized as existing between the strongly divergent populations
herein categorized as subspecies gigantea and subspecies
tecta, respectively, and the populations of intermediates
(putative nothomorphs) herein given taxonomic status as
subspecies macrosperma, based on Arundinaria macrosperma
Michaux (as to nomenclatural lectotype) . The apparent
absence of either a clear morphological discontinuity, or an
absolute barrier to the exchange of genes, as between the
populations comprising these three groups of plants, they
are herein interpreted as constituting a polymorphic species,
and given the formal, comprehensive name, Arundinaria gigantea,
sensu lato.
free. Stigmas typically 2 (atypically and very rarely
3 in A. radiata, q. v,), Mature fruit not yet found.
ETYMoLoGY.-The name A tractantha (f.) Greek
atraktos, spindle, and anthos, flower, alludes to the
spindle-like form of the anthecia.
TYPE-SPECIES.-A tractantha radiata McClure.
RELATIoNSHIPS.-The obvious natural affinity of
the known bamboos of the genus Atractantha is
toward bamboos of the genus Elytrostachys, with
which they share the following features: inflorescences
are pseudospikelets of a peculiar form characterized
by rachises with long terminal segments,
each of which serves as the pedicel of an abscissile
spikelet; the absence of the transitional glumes
commonly found at the base of each spikelet, and
the presence, in their stead, of a bract inserted on
each of the two nodes that precede the elongated
terminal segment of each rachis; each spikelet typically
containing but a single perfect floret, which
is followed by a rudiment of a sterile anthecium
borne on the tip of a bristle-like prolongation of
the rachilla. Insofar as currently available specimens
inform us, bamboos of the genus Atractantha
differ from bamboos of the genus Elytrostachys in
the following respects: midculm branch complements
containing several axes of primary order,
each developed from one of several distinct initial
branch primordia each with its own prophyllum;
leaf sheaths with auricles and oral setae not only
much less conspicuously developed but of different
morphological configuration; empty bracts preceding
the base of the terminal segment of each rachis
none or one; the slender, spindle-like form of the
anthecia; and the number of stamens in each flower
limited to 3. While the recorded geographical
range of bamboos of the genus Atractantha is limited
to the state of Bahia, Brazil, the recorded
FIGURE 19.-Athroostachys capitata (Hooker) Bentham. A,
Basal portion of young plant in the vegetative state, showing
two culms and the rhizomes to which they are terminal,
x 0.G; B, midculm node and branch complement showing
three primary branches of independent origin, derived from
the primordial meristem enclosed within the prophyllum
of a solitary primary bud, x 3; c, tip of a leafy flowering
branch with its terminal inflorescence, X 0.6; D, apex of
leaf sheath which subtends a flowering branch and the
base of its blade, x 0.9; E, base of a flowering branch
(schematic). Drawing A based on Glaziou 13325 (P), B, on
Weddell 152 (P) , and c, D, E on Glaziou 20155 (F) ,
NUMBER 9 47
48 SMITHSONIAN CONTRIBUTIONS TO BOTANY
geographical range of bamboos of the genus Elytrostachys
extends from Venezuela to Nicaragua.
bution of the two known species of the genus Atractantha
is limited to the state of Bahia, Brazil. It
appears that nothing concerning the ecological features
of their respective natural habitats was re-
DISTRIBUTIoN.-The recorded geographical distri- corded by the collectors of the only available
specimens.
Key to the Known Species of Atratantha
la. Leaf blades up to 10.5 cm long and up to 1.0 cm broad; leaf sheaths hispid and hispidulous
at first; internodes of leafy twigs hirsute; inflorescences diffuse, their ultimate
branches curved; rachises of sympodial insertion, each bearing a bud at a single proximal
node; bud-subtending bracts laminiferous ................................ 1. A. falcatu, new species
lb. Leaf blades up to 19.5 cm long and up to 3.5 cm broad; leaf sheaths glabrous from the
first; internodes of leafy twigs glabrous; inflorescences capitate, their ultimate branches
straight; rachises of distichous insertion, each bearing a bud at each of"2 proximal
nodes; bud subtending bracts without leaf blades .................... 2. A. radiafu, new species
1. Atructantha fulcutu McClure, new species
FIGURE 21
Plantarum habitus non adhuc relatus. Culmi in
specimine suppetenti neque satis representati neque
descripti, itaque ignoti. Vaginae culmi et vaginae
ramorum primariorum in specimene suppetenti deficientes.
Vagina foliorum arcta, initio hispida hispidulaque
demum subglabrescens; versus apicem
tantum nervis interdum debiliter manifestis; auriculis
vel subnullis vel sat evolutis glebosisque; setis
oralibus paucis, e basi ad apicem sensim angustatis,
omnino glabris vel basin versus sparsissime et antrorse
scaberulis, fragilissimis, vel mox vel postea
effractis; ligula interiore subnulla, truncata; ligula
exteriore truncata, margine primo sparse et minutissime
ciliolata denique glabrescenti integraque;
petiolo vix 2 mm longo, pagina adaxiali hispidulo
pagina abaxiali hirsuto; lamina usque 10.5 cm
longa et 1.0 cm lata, lineari-lanceata, apice acuta,
basi rotundata, basin versus utrinsecus sparse hirsuta,
alibi utrinsecus primo antrorse hispidula,
denique sensim glabrescenti; costa mediana et tribus
paribus nervorum secundariorum pagina
abaxiali sat prominulis; pagina adaxiali inter
omnes nervos propinquos profunde et angustissime
sulcata, prope marginem externam costa alata penitus
instructa; venulis transversis extus haud manifestis.
Inflorescentiae ramos ramulosque vel foliigeros
vel aphyllos terminantes, ob rhachidibus ramificationibus
insertionis sympodialis falcatae diffusaeque.
Bracteae e quoque rhachidi proditae
pleraeque 2, altera gemmam subtendenti altera
vacua, approximatae, laminiferae, paucinerves,
substantia indurata sed fragili, dorso sparsim puberulae.
Spiculae usque 15 mm longae, maturitate
cadentes; ubi unaquaque spicla delapsa relictum
est segmentum terminale rhachidis a bractea persistenti
fere omnino vestitum. Glumae transitionales
nullae. Anthecia papyracea, glabra nitidaque,
nervis extus vix vel haud manifestis. Lemma usque
15 mm longum, lanceatum, attenuate acuminatum.
Palea lemmate brevior, secus margines sulci angustissimi
antrorse ciliata. Lodiculae diaphanae,
angustae, margine apicem versus sparsim ciliolatae.
Antherae apice emarginatae, basi profunde divisae,
ubi desiccatae brunneae. Ovarium vel omnino
glabrum vel apice tantum hispidulum. Stylus antrorse
hispidulus. Fructus maturus adhuc non visus.
FIGURE 20.-Athroostachys capitata (Hooker) Bentham. A,
Diagram of the branching system of the inflorescence; B,
a primary branch of the inflorescence, terminating in a
spikelet, and bearing two secondary branches each terminating
in a single somewhat immature spikelet, x ca. 3; c,
a secondary branch subtended by a bract and bearing a
solitary spikelet, x ca. 3; D, transitional glumes of a mature
terminal spikelet, x ca. 3; E, hermaphrodite floret and (at
left) depauperate (sterile) anthecium of a mature terminal
spikelet, x ca. 3; F, component structures of a secondary
branch of the inflorescence (schematic) ; G, floral parts in
immature state (gynoecium not visible), x ca. 12-i H, mature
lodicule complement (the narrow symmetrical one is the
posterior member), x ca. 18; I, stamen, x ca. 12; J, gynoecium,
unmodified, x ca. 12; K, gynoecium, immature
fruit stage, x ca. 12. All drawings based on Glaziou 20155
(F).
NUMBER 9 49
K
50 SMITHSONIAN CONTRIBUTIONS TO BOTANY
Plants of unrecorded habit. Culms neither rep
resented nor described in the available specimen.
Culm sheaths and sheaths of the primary axes of
branch complements lacking in the specimen. Leaf
sheaths tight, hispid and hispidulous at first, at
length subglabrous; nerves sometimes weakly discernible
toward the apex only; auricles either subobsolete
or somewhat developed and lumpy; oral
setae few, tapered, glabrous throughout or sparsely
and antrorsely scaberulous basally, very fragile,
sooner or later broken off; the inner ligule truncate,
not at all or barely exserted; the outer ligule
truncate, sparsely ciliolate on the margin at first,
at length glabrescent and entire; petiole scarcely 2
mm long, hispidulous on the adaxial surface, and
hirsute on the abaxial; the blade up to 10.5 cm
long and 1.0 cm broad, linear-lanceolate, apically
acute, basally rounded, sparsely hirsute on both
surfaces at the base, elsewhere antrorsely hispidulous
at first, then gradually glabrescent on both surfaces;
the conventional midrib and three pairs of
secondary nerves moderately prominent on the
abaxial surface, narrowly and deeply sulcate between
all adjacent nerves of the adaxial surface,
with a winglike rib from base to tip near the outer
margin and, at the same time, the midrib obsolete
toward the apex on this surface; transverse veinlets
not at all manifest on either surface.
Inflorescences terminating leafy and subleafless
branches of all orders; of diffuse, falcate branching,
the rachis branches of sympodial insertion. Bracts
at the base of each rachis commonly 2, approximate,
laminiferous, few-nerved, of indurate but
brittle substance, sparsely puberulent dorsally, one
subtending a bud, the other empty. Spikelets up
to 15 mm long, falling away at maturity, each at
length leaving the ca 5 mm long terminal segment
of its rachis almost completely clothed in a persistent
bract. Transitional glumes at the base of
each spikelet none. Anthecia papyraceous, glabrous
and lustrous, without externally discernible nerves.
Lemma up to 15 mm long, lanceate, attenuately
acuminate. Palea shorter than the lemma, the sulcus
very narrow, with its edges marked by a row of
antrorse cilia. Lodicules diaphanous, narrow,
sparsely ciliate on the margin apically. Anthers
dark brown when dry, sagittate basally and emarginate
apically, the tip of each half curving outward.
Ovary glabrous throughout or antrorsely hispidulous
apically. Style antrorsely hispidulous. Mature
fruit not yet seen.
ErYMoLocY.-The trivial name, falcata, alludes
to the curvature that characterizes the branches of
the inflorescences. This curvature is construed as
induced by the sympodial nature of the branching
of the individual rachises.
HoLorYPE.-In the U.S. National Herbarium
(no.2040296) collected in a “carrascal” [forest of
small trees] at Esplanada, Bahia, Brasil, sometime
during 1950-51, by Gerald0 Pinto (no. 0681).
2. Atructantha radiata McClure, new species
FIGURES 22, 23
Plantarum habitus collectore neglectus. Culmi
in specimine suppetenti neque satis representati
neque descripti, ita ignoti. Vaginae culmi et vaginae
ramorum primariorurn in specimine suppetenti
deficientes. Vagina foliorum arcta, dorso glabra,
opaca, nervis vix aut haud manifestis; auriculis vel
nullis vel ad lineam reductis; setis oralibus crebris,
superne saltem anfractis, omnino glabris, propter
substantiam fragilissimam mox efiactis; ligula interiore
ca 1.0 mm longa, dorso granulata, apice
asymmetrica, margine integra; ligula exteriore arcuata,
angusta, margine primo minute ciliolata
demum glabrescenti; petiolo usque 5 mm longo,
gracili omnino glabro; lamina usque 18 cm longa
et 3.5 cm lata, lanceata, apice acuminata, basi inaequaliter
vel rotundata vel subdeltoidea, utrinsecus
glabra vel faciei abaxiali pro parte subtiliter puberula;
nervis pluribus, validis, tertiariis a secundariis
haud facile distinguendis, venulis transversis
perpaucis inter se late distantibus.
Inflorescentiae capitatae, in ramis ramulisque foliiferis
insertae ubi et terminales et laterales, rhachidum
ramificationes insertione distichae. Rhachidum
segmentum terminale usque 10 mm longum,
FIGURE 21.--A tractantha falcata McClure. A, Culm internode
with branch complement of leafy and leafless flowering axes,
x 0.6; B, base of branch complement, x ca. 3; c, apex of
leaf sheath and base of leaf blade, x ca. 6; Da and Db variations
in structures forming terminal portion of a flowering
twig, diagramatic; E, laminiferous bract from base of a
branch of the inflorescence; F, floret, x 6, G, lemma, x 6; H,
palea (right) and prolongation of the rachilla (left), x 6;
I, lodicule complement, x 15; J, stamen, x 6; K, gynoecium,
x 30. All drawings based on Pinto 0681 (US).
NUMBER 9 51
H
52 SMITHSONIAN CONTRIBUTIONS TO BOTANY
NUMBER 9 53
apice usque 1.0 mm diametro dilatatum, omnino
glabrum laevigatum nitidumque. Bracteae gemmam
subtendentes in nodis basalibus omnis rhachidis
vulgo 2, glabrae, substantia delicatissima,
infima 2-3 mm, superiore usque 5 mm longa, non
laminiferae. Spiculae usque 19 mm longae, omnes
maturitate caducae, rhachidi sua tum conspicua
relicta apicaeque nuda. Glumae transitionales
nullae. Anthecia papyracea, dorso glabra nitidaque,
nervis extus haud manifestis. Lemma usque 17 mm
longum, lanceatum, attenuate acuminatum. Palea
lemmate parum longior, seas latera sulci angusti
ciliolis antrorsis crebris ornata. Lodiculae minimae
angustae diaphanae acuminatae secus marginem
superne ciliolatae. Antherae apice obtusae basi
conspicue sagittatae, ubi desiccatae vinaceae. Ovarium
superne pilosum. Stylus vel conspicue vel
inconspicue hispidulus. Stigmata typice 2 (atypice
et rarissime 3). Fructus maturus adhuc non visus.
Plants of unrecorded habit. Culms neither adequately
represented nor described in the available
specimens. Culm sheaths and sheaths of the primary
axis of branch complements lacking in the
specimen. Leaf sheaths tight, glabrous, dull, the
nerves scarcely or not at all visible externally;
auricles either lacking or reduced to a line; oral
setae crowded, curved distally, glabrous throughout,
very fragile and soon broken off; the inner
ligule ca 1.0 mm long, dorsally granulate, apically
asymmetrical, the margin entire; the outer ligule
arcuate, narrow, the margin minutely ciliolate at
first, then glabrescent; petiole up to 5.0 mm long,
slender, glabrous throughout; blade up to 18 cm
long and 3.5 cm wide, lanceate, acuminate, rounded
or subdeltoid at the asymmetrical base, either
glabrous on both surfaces, or partly and minutely
puberulous on the abaxial surface; many-nerved,
the nerves strong, the tertiary ones not easily distinguished
from the secondary ones; transverse veinlets
few and far removed from each other.
Inflorescences capitate, terminal and lateral to
leafy branches and leafy twigs, the branches (ra-
FIGURE 22.-Atractantha radiata McClure. A, Leafy flowering
branch, x 0.6; B, apex of leaf sheath and base of leaf blade,
x 3; c, base of branch complement, x 4.8; D, E, variants of
flowering axes, x 0.6. All drawings based on Frdes I9947
(US) *
chises) of distichous insertion. Bracts at the base
of each rachis typically 2, approximate, glabrous
and of very delicate substance, not laminiferous,
soon disintegrating, the lowermost one 2-3 mm,
the uppermost up to 5 mm long; each subtending
a bud. Spikelets up to 19 mm long, readily falling
away at maturity, each leaving conspicuous and
naked the apex of the terminal segment of its
rachis; the latter glabrous, smooth and lustrous, up
to 10 mm long and (at its slightly flared apex) up
to 1 mm in diameter. Anthecia papery, glabrous
and shining throughout, the nerves not at all visible
externally. Lemma up to 17 mm long, lanceate,
attenuately acuminate. Palea a little longer than
the lemma, the margins of the sulcus close together
and densely fringed with antrorse cilia. Lodicules
diaphanous, very small, narrow, the margin apically
ciliate. Anthers sagittate basally, obtuse apically,
dark purple when dry. Ovary apically pilose. Style
conspicuously or inconspicuously hispidulous. Stigmas
typically 2, atypically and very rarely 3. Mature
fruit not yet seen.
ETYMoLocY.-The trivial name, radiata, alludes
to the radiate orientation of the rachises of the
capitate inflorescence.
HOLOTYPE.-h the U. S. National Herbarium
(nos. 1910764, 1910765, and 2146781) collected at
“Estrada de Bom Gosto a Olivenca, Estado da Baia,
Brasil, March 15, 1943,” by Ricardo de Lemos
Froes (no. 19947). Additional material examined:
an isotype from the same collection on loan from
the herbarium of the Instituto Agronomico do
Norte, Belem, Para, Brazil (IAN no. 15373).
Aulonemia Goudot
FICURE~ 24-26
Aulonemia Goudot, 1846:75.
Matudacalamus Maekawa, 1961:344.
Plants unicespi tose, unarmed. Rhizomes pachy
morph. Culms of small or medium stature and
self-supporting, to very tall and upright, the apex
erect, nodding, pendulous or scandent; each midculm
node bearing a single initial (primary) branch
bud, the prophyllum elongating simultaneously
with the germination of the bud. Branch complement
at midculm nodes typically comprising but a
single initial (primary) axis, this dominant over
axes of higher orders that may proliferate promptly
54 SMITHSONIAN CONTRIBUTIONS TO BOTANY
rs H
NUMBER 9 55
or tardily from usually functional buds borne at
its own proximal nodes. Sheath at midculm nodes
typically lacking a well-marked (conspicuous) basal
girdle; the auricles and oral setae (varying from
species to species) entirely lacking to well developed
and scaberulous to entirely glabrous. Leaf blades
with transverse veinlets not at all to weakly to
sometimes strongly manifest externally.
Inflorescences semelauctant, forming typically
open panicles, the typically solitary, pulvinate or
epulvinate primary ramifications with subtending
bracts generally small, obsolete or lacking entirely.
Spikelets few-flowered to many-flowered, terminating
in a depauperate sterile anthecium. Transitional
glumes 2 or 3; when three the uppermost one
usually recognizable as a “sterile lemma.” Lemma
in functional florets embracing its palea only basally
at maturity. Palea broadly sulcate and 2-
keeled dorsally, the margins not at all or barely
imbricate. Stamens 3, the filaments filiform, free.
Stigmas 2. Mature fruit (as far as known) an
oblong or subfusiform, sulcate, apically mucronate
caryopsis, the pericarp pergamineous, glabrous, of
uniform thickness throughout or sometimes appreciably
thicker apically, the sulcus and embryotegium
clearly manifest.
ETYMoLoGY.-The name Aulonemia, a Latinized
combination of the Greek words, aulos (ancient
name of a flute made from a shin bone) and nemos
(a forest with pasture for cattle) alludes to (1) the
musical instruments (called, in the vernacular,
“queko”) made by the native inhabitants from
elongated culm internodes of plants of the typespecies,
and (2) the browse made accessible to
herbivorous animals by the pendulous leafy culm
tips of plants of the same species (cf. Goudot,
1846:75, et seq.; R. W. Brown, 1954:112, 554). The
terminal element, nemos, makes the natural gender
of the word Aulonemia neuter; however, the
vernacular trivial name, queko, does not reveal a
FIGURE 23.-Atractantha radiata McClure. A, Primary pseudospikelet
(early stage of development), x 15; B, two
pseudospikelets, showing the prophyllum at the base of the
one on the left, x 3; c, cluster of pseudospikelets, with two
pedicels naked, x 3; D, diagram of a typical pseudospikelet;
E, floret, x 9; F, lemma, x 9; G, palea (left) and prolongation
of the rachilla (right), x 9; H, diagrammatic crosssection
of palea, x 18; I, lodicule complement, x 15: J,
stamen, x 6; K, gynoecium, x 15. All drawings based on
Froes 19947 (US).
choice of gender by Goudot. In the absence of
any other precedent, Aulonemia was given the
feminine gender by McClure and L. B. Smith (in
Reitz, ed., 1967:47) when they attached to it the
trivial name, lancipora.
TypE-spEcIEs.-Aulonemia queko Goudot.
RELATIoNSHIPs.--The array of 24 species here incorporated
in Aulonemia imparts to the genus an
admittedly polymorphic content. Some of the arms
of marked diversity recognizable within the genus
as I have circumscribed it involve species that show
individual features suggesting the presence of genetic
elements shared by several other genera. Aulonemia
humillima alone is superficially grass-like
in appearance. My study of fragmentary specimens
and brief field notes provided by the solitary extant
collection of Aulonemia hirtula suggests to me
that it probably joins Aulonemia queko in being
distinguishable from the other known species of
the genus by the possession of relatively large culms
with greatly elongated internodes, each followed by
two or more nodes, the adjacent ones separated
from each other by an aborted subobsolete internode.
This peculiarity-without apparent linkage
with any other stable attribute-also appears as a
stable attribute of each plant in Glaziophyton mirabile,
in a few species of Arthrostylidium (cf.
McClure, in Maguire, Wurdack, et al., 1964:2), and
quite generally in the known species of Myriocladus.
Aulon,emia eflusa apparently is joined by Aulonemia
deflexa in combining other vegetative
features that suggest a phylogenetic heritage shared
by members of Myriocladus, viz., the frequently delayed
germination of primary branch buds at culm
nodes and/or the branch bud at the proximal nodes
of primary branches (a feature ostensibly related
to the strongly indurated vegetative foliar appendages)
and the closely appressed culm branches, each
comprising conspicuously elevated numbers of short
internodes and crowded foliiferous nodes. Aulonemia
subpectinata shows a facultative but not obligate
tendency to delay the germination of midculm
branch buds, and to postpone proliferation from
buds at the proximal nodes of the initial branch at
midculm nodes. Aulonemai laxa and A. viscosa
share this tendency. These two species also share
with other species of Aulonemia, and with some
species of both Aythrostylidium and Bambusa, a
more or less conspicuous vinous tinting or macula56
SMMITHSONIAN CONTRIBUTIONS TO BOTANY
tion of either vegetative structures alone or reproductive
structures alone, or both. To accommodate
one of these last two species, Maekawa has proposed
(196 1 : 344) the recognition of a monotypic genus,
Matudacalamus. In the perspective resulting from
my studies in the field and in various herbaria,
Matudacalamus laxus Maekawa appears to be a
member of one of the several nondisjunct arms of
diversity that make up the polymorphic natural
genus, Aulonemia. Maekawa expresses the opinion
that the “wavy, smooth” oral setae of the leaf
sheaths, and the “one-at-a-node” culm branches are
two critical features that set off Matudacalamus
from other genera he mentions as related to it.
However, neither one nor both of these featuresnor
any other combination of morphological features
at present available to me-appears to supply
either a reliable (stable) basis, or a justification, for
segregating Matudacalamus from Aulonemia at the
genus level. The mingling of racemose branching
with the paniculate branching of the inflorescences
of Aulonemia purpurata gives it a striking resemblance
to the branching typical of the inflorescence
in Colanthelia cingulata. The type-specimen of A.
purpurata, however, does not include any feature
that shows other affinities in that direction.
DIsTRIBu’rIoN.-The named species here incorporated
in Aulonemia gives the genus an aggregate
recorded geographical and altitudinal range extending
from Brazil, with 10 species at 800-2800 m,
through Bolivia, with one species at about 3200 m;
Peru, with four species at 1500-3000 m; Ecuador,
with one species at 1800-3000 m; Colombia with
three species at 2800 m; Venezuela, with three species
at 1900-2760 m; Guyana, with one species at
about 2700 m; Costa Rica, with one species at 1800-
2700 m; to Mexico, with one species at about
2200 m.
Annotated Checklist of
Bamboos Here Recognized
as Belonging in the Genus Aulonemia
1. Aulonemia amplissima (Nees) McClure, new
combination.
Arundinaria amplissima Nees, 1834:479.
Arthrostylidium amplissimum (Nees) McClure, in
Steyermark, et al., 1951:33.
My dissections of florets from the type-collection
reveal the number of stigmas to be regularly 2-
instead of 3 as recorded by Nees (1834:479) and as
reiterated by Ruprecht (1839:27). The number of
stigmas is correctly given by Munro (1868:26) and
by later authors.
2. Aulonemia aristulata (Doell) McClure, new
Arundinaria aristulata Doell, in Martius, 1880: 165.
Sieglingia aristulata (Doell) Kuntze, 1898, I11 (3):
341.
Arundinaria mucronata Munro ex E.-G. Camus,
19 12: 244.
3. Aulonemia deflexa (N. E. Brown) McClure,
combination.
new combination.
4. Aulonemia eflusa (Hackel) McClure, new combination.
Arundinaria deflexa N. E. Brown, 1901:75.
Arundinaria eflusa Hackel, 1903a:71.
Arthrostylidium eflusum (Hackel) McClure, in
Steyermark, et al., 1951:31.
5. Aulonemia glaziovii (Hackel) McClure, new
combination.
6. Aulonemia goyazensis (Hackel) McClure, new
combination.
Arundinaria glaziovii Hackel, 1903a: 72.
Arundinaria goyazensis Hackel, 1903a:71.
7. Aulonemia haenkei (Ruprecht) McClure, new
combination.
Arthrostylidium haenkei Ruprecht, 183927.
Arundinaria haenkei (Ruprecht) Hackel, 1903:69.
Arundinaria setifera Pilger, 1905: 145.
8. Aulon,emia herzogiana (Henrard) McClure,
new combination.
Arundinaria herzogiana Henrard, in Herzog, 1921 :
75.
9. Aulonemia hirtula (Pilger) McClure, new combination.
Arundinaria hirtula Pilger, 192 1 :445.
FIGURE 24.-Aulonemia queko Goudot. A, Leafy flowering
branch, x 0.6; B, junction of leaf sheath and leaf blade,
x ca. 1.8; c, insertion of lowermost branch of inflorescence,
x ca. 6; D, spikelet, x 1.5; E, transitional glumes I, 11, and
III, x 3.6; F, sterile lemma, x 3.6; c, floret, x 3.6; H, fertile
lemma, x 3.6; I, palea, x 3.6; J, lodicule complement, X
7.2; K, stamen, x ca. 6; L, gynoecium, x ca. 12: M, stigmatic
processes, greatly enlarged. .411 drawings based on Goudot
no. I (P).
2
C
58 SMITHSONIAN CONTRIBUTIONS TO BOTANY
10. Aulonemia humillima (Pilger) McClure, new
Arundinaria humillima Pilger, in Engler and
11. Aulonemia laxa (F. Maekawa) McClure, new
Matudacalamus laxus F. Maekawa, 1961:345.
12. Aulonemia panriflora (Presl) McClure, new
Guadua parviflora, J. S. Presl, in K. B. Presl, 1825-
Bambusa parviflora (Presl) J. A. and J. H. Schultes,
in Roemer and Schultes, 1830, 7 (2):1350.
Arthrostylidium maculatum Ruprecht, 1839:28.
Arundinaria maculata (Ruprecht) Hackel, 1903a:
69.
13. Aulonemia patula (Pilger) McClure, new combination.
Arundinaria patula Pilger, 1898:719.
14. Aluonemia purpurata (McClure) McClure,
Arthrostylidium purpuratum McClure, 1942: 170.
15. Aulonemia queko Goudot, 1846:76, pl. 4 (Figures
24-26).
Arthrostylidium queko (Goudot) Hackel, in Engler
and Prantl, 1887:93 [sphalm. quexo].
Arundinaria queko (Goudot) Hackel, 1903a:74.
Available specimens from collections supplementing
the holotype represent this taxon as a
species embracing plants somewhat diverse in respect
to mature stature and habit, size, shape,
texture, and vesture of leaf blades, disposition of
the peripheral branches of the inflorescence, and
degree of development of awns and dorsal pricklehairs
on the lemmas. The wide altitudinal range of
the natural distribution of plants, which I recognize
as belonging to this species, suggests that some of
the phenotypic diversities noted may be ecotypic
in origin. More detailed studies in field and laboratory
may, however, reveal the existence of genotypic
bases for some of the observed diversities.
In 1950 I had an opportunity to visit the typelocality
of Aulonemia queko near La Trocha in the
region of Colombia called Quindio. I found no
sign of either the original forest, or the bamboo
described by Goudot as Aulonemia queko. The
combination.
Prantl, 1906: 100.
combination.
combination.
1835, I (4-5):257.
new combination.
mountainside is clothed instead with a beautiful
carpet of Pennisetum clandestinum Hochstetter ex
Chiovanda, a grass native to Africa. I eventually
decided that specimens of a plant I had collected
in Ecuador (McClure 21429) and in Peru (McClure
21450) represent Aulonemia queko, in spite of
minor deviations with respect to some of the
features given this species by Goudot. This ended
my own quarter-century of puzzlement over the
riddle presented by Goudot’s deliniation of Aulonemia
queko. It appears that Hackel’s action (in
Engler and Prantl, 1887) in making Aulonemia a
synonym of Arthrostylidium was induced by Ruprecht’s
(1839) broad circumscription of Arthrostylidiunz.
Again, Hackel’s (1903a) transfer of
Alonemia queko to Artmdinaria apparently was
suggested by Nees’ (1834) image of Arundinaria.
The anguished efforts of Hackel (1903a) and Pilger
(in Urban, 1900-1901) to find bases for distinct
and stable generic images of the diverse elements
that had been, at one time or another, placed in
Arthrostylidium and Arundinaria are brought into
focus elsewhere (p. 19).
I identify as representing a species of Elytrostachys
the specimen Fendler 2496 (a leafy twig in
the vegetative state, preserved at Kew) cited by
Munro (1868:46) in his treatment of Aulonemia
queko. It seems likely that Munro’s modification of
Goudot’s spelling of the specific epithet from queko
to quexo misled Hackel (in Engler and Prantl,
1887). In 1903a, however, Hackel restored the original
spelling in the combination, Arundinaria
queko.
16. Aulonemia radiata (Ruprecht) McClure and
Arundinaria radiata Ruprecht, 1839:25.
17. Aulonemia ramosissima (Hackel) McClure,
Arundinaria 1-amosissima Hackel, 1903a:74.
Arundinal-ia glaziouii var. macroblephara E.-G.
18. Aulonemia setigera (Hackel) McClure, new
L. B. Smith, in Reitz, ed., 1967:56.
new combination.
Camus, 1913, I:41.
combination.
FIGURE P5.-8ulonemia queko Goudot. A, Seedling plant,
x 0.6; B, twinned culm sheaths, x ca. 0.6: c, sterile leafy
branch, x 0.6; D, apex of leaf sheath and base of leaf blade,
x 1.5. Drawing A based on McClure 214298 (US), B, c, D on
McClure 21429 (US).
NUMBER 9 59

NUMBER 9 61
Arundinaria setigera Hackel, 1903a:73.
Arthrostylidium aristatum Glaziou ex E.-G. Camus,
19. Aulonemia sodiroana (Hackel) McClure, new
Arundinaria sodiroana Hackel, 1930a:70.
On the basis of the similarity of the inflorescences
in their respective type-specimens, Arundinaria
sodiroana was by Hitchcock and Chase made a
synonym of Arundinaria patula. Plants of the two
taxa may be distinguished, however, even in the
nonflowering state by the distinctly prominent
fimbriation of the outer margin of the leaf sheaths
present in A. sodiroana but lacking in A. patula.
20. Aulonemia steyermarkii (McClure) McClure,
new combination.
Arthrostylidium steyermarkii McClure, in Steyermark,
et al., 1951:31.
21. Aulonemia subpectinata (0. Kuntze) McClure,
new combination.
Arthrostylidium subpectinatum 0. Kuntze, 1891,
II:760.
22. Aulonemia trianae (Munro) McClure, new
combination.
Arundinaria trianae Munro, 1868:25.
Arundinaria multiflora Doell, in Martius, 1880: 166.
23. Aulonemia ulei (Hackel) McClure and L. B.
Arundinaria ulei Hackel, 1903a:75.
Species sedis mihi etiam nunc incertae manet.
24. Aulonemia viscosa (Hitchcock) McClure, new
Arundinaria viscosa Hitchcock, 1927a: 79.
1913, I:67.
combination.
Smith, in Reitz, ed., 1967:57.
combination.
Checklist of New World Species Here
Excluded from the Genus Aulonemia
FIGURE 26.-Aulonemia queko Goudot. A, Upper culm nodes
with buds and base of branch bud, x 0.6; B, buds enlarged,
x 1.2; c, bud in advanced development, x 1.2; D, base of
young branch complement, x ca. 1.2; E, tip of leafy branch
terminating in an inflorescence, x 0.6; F, spikelet, x 2.4;
c, transitional glume I, x 6; H, transitional glume 11, x 6; I,
sterile lemma, x 6; J, fertile lemma, x 6 K, floret, showing
palea, x 6; L, mature fruit (hilum side), x ca. 15; M, mature
fruit (embryo side), x ca. 15; N, mature fruit, longitudinal
section, x ca. 15; 0, lodicule complement, x ca. 6;
P, gynoecium (without stigmas), x ca. 6. All drawings based
on McClure 21429 (US).
1. Aulonemia cingulata. See Colanthelia cingu-
2. Aulonemia intermedia. See Colanthelia inter-
3. Aulonemia lancipora. See Colanthelia lancilata.
media.
flora.
Bambusa Schreber, Subgenus Guadua (Kunth)
Hackel
FIGURES 27-29
Genus Cuadua Kunth, 1822a: 150; 1822b:254.-Ruprecht,
1839:38 [1840: 1281.-Munro, 1868:70.-Doell, in Martius,
1880: 176.-Bentham, in Bentham and Hooker, 1883: 1210.-
McClure, 1957:203.
Bambusa (Guaduae) Nees, in Martius, 1829:532.
Bambusa Sect. I1 Guadua, Hackel, in Engler and Prantl,
Bambusa (Guadua) Hackel, 1903c:194.
1887:95.
Plants typically unicespitose and typically armed.
Rhizomes pachymorph, the rhizome neck in some
species more or less elongated (exceptionally, as
in McClure 21438 and 21438-A, to 5 m or more).
Culms self-supporting below and apically nodding,
pendent, or clambering; internodes cylindrical or
terete, in some species shallowly to more or less
deeply sulcate above the point of insertion of a
bud or a branch complement; hollow, occasionally
subsolid (teste Doell, in Martius, 1880: 186), exceptionally
(as commonly in Bambusa amplexifolia)
multifistular in the lower part of the culm. Branch
buds at culm nodes typically solitary, lacking in the
midculm range (in plants of some taxa at mature
stature). Midculm branch complements lacking in
mature plants of some species (as in some strains
of Guadua angustifolia; see Bambusa guadua),
facultative restricted monoclade in others; in most
species unrestricted monoclade, of restricted insertion
(the insertion rarely gremial, as in Guadua
spinosa; see Bambusa swalleniana) the component
axes more or less strongly unequal, the primary
axis always dominant. Branches (at least those at
lower nodes of the culms) typically thorny in all
known species (see exception noted under Bambusa
amplexifoZia). Leaves (leaf sheath blades) with
transverse veinlets not at all visible to more or
less clearly manifest externally.
Inflorescences iterauctant, developing by way of
pesudospikelets, the primary pseudospikelets either
pedunculate or sessile, and forming sparse to more
62 SMITHSONIAN CONTRIBUTIONS TO BOTANY
or less densely crowded, sometimes capitate aggregations,
lateral or terminal to leafy or leafless axes.
Spikelets typically comprising several perfect florets
(only 2 in Bambusa capitata) and either tapering
apically through progressively reduced sterile
anthecia, or terminating abruptly in a single depauperate
anthecium; the rachilla potentially
disarticulating immediately below the locus of
insertion of each fertile lemma, and in some species
also below a sterile lemma. Transitional sheathing
organs inserted between the uppermost bud-subtending
bract and the first fertile lemma 0-2 (-3).
[In some species of Bambusa subgenus Guadua,
and in other genera in which branches of the inflorescences
form pseudospikelets, sheathing structures
occasionally reported as “sterile lemmas”
sometimes prove to be bracts that are inserted at
a level below the base of the spikelet proper, and
subtend prophyllate buds that are easily mistaken
for abortive florets. Cf. Munro 1868, pl. 5: fig. 4
(such a bud) and fig. 5 (a floret) both referred to
by Munro (1868: 153) as “flosculi fertilis.”] Lemma
in functional florets fully embracing its palea only
basally at maturity. Palea sulcate and 2-keeled
dorsally, the keels typically more or less prominently
winged (not more noticeably winged in Burchell
7642 [cotype of Guadua refracta Munro] than in
some species of the Old World genus Bambusa),
the margins typically not at all or only barely
imbricate. Lodicules typically 3, the anterior 2
asymmetrical and paired (sometimes connate}, the
posterior one symmetrical and smaller. Stamens 6
(3 in Guadua spinosa; see Bambusa swalleniana),
the filaments filiform, free. Style terminating in 2
or 3 (rarely 4) stigmas or (exceptionally) undivided.
Mature fruit an oblong or lagenoid, mucronate or
subrostrate, in some species strongly dorsiventral
caryopsis, the pericarp pergamineous or coriaceous,
somewhat thickened apically, the sulcus and the
basal position of the embryo clearly manifest.
The name Guadua is the Latinized version of a
vernacular term used by aboriginal natives of
Ecuador and Colombia to designate plants of this
subgenus.
TYPE-sPECIEs.-McClure (1957:203) designated
Guadua angustifolia Kunth as type-species of the
genus Guadua. This species therefore becomes the
type of subgenus Guadua of the genus Bambusa,
but under the name Bambusa guadua Humboldt
et Bonpland (Figure 27).
RELATIoNsHIPs.-Bamboos hitherto allocated to
the genus Guadua have no close relatives among
known members of other New World genera. However,
their common technical image is very similar
to that of members of the Old World genus Bambusa.
Kunth (1822a:150) failed to cite any morphological
feature to differentiate his genus Guadua
from Schreber’s genus Bambusa. The first promising
effort in this direction was made by Munro
(1868:76) in‘ these words: “The upper palea in
Guadua has the keels much more distinctly and
broadly winged than in any Bambusa.” However,
the trivial taxonomic significance Munro attached
to this feature is clearly apparent in the following
sentence: “This genus is kept up by most authors;
but I am unable, in the absence of a perfect fruit,
to point out any good distinction, except a wellmasked
one in geographical distribution, between
it and Bambusa.” Munro then lists the features
proposed by other authors as distinctive of Guadua
as a genus, and shows that not one of them provides
Guadua with a clearly distjunct morphological relation
to Bambusa. My own observation is that
the keels of the palea are not more noticeably
winged in Burchell 7642 (cotype of Guadua refracta
Munro) than in some species of the Old
World genus, Bambusa. It also appears that subgenus
Guadua and subgenus Bambusa are not
sharply disjunct, with respect to features of the
gross morphology of either their vegetative or
reproductive structures or with respect to their
geographical endemism.
By 1829 Nees (in Martius, p. 532) had already
made “Guaduae” an uncategorized division of the
genus Bambusa. This proposal was more fully
developed by Nees in Linnaea (1834:465).
In 1887 Hackel (in Engler and Prantl, p. 95)
proposed the formalization of the status of Guadua
as Section I1 under Bambusa; in 1903 he (1903~:
194) continued the subordination of Guadua under
FIGURE 2T.-BamBusa guadua Humboldt and Bonpland,
sensu lato. A, culm sheath, adaxial aspect, x 0.3; B, same,
detail at base of blade, enlarged; c, upper portion of culm
sheath, abaxial aspect, x 0.3; D, portion of first-order branch
bearing a leafy twig with thorns, x 0.6; E, apex of leaf
sheath and base of blade, x ca. 0.6; F, branch bud at culm
node, x ca. 0.6; G, short-shoot thorny branch at lower node
of culm, x ca. 0.6. Drawings A-D based on Gutierrez and
Rarkley 17 C 609 (US), and E-G on Fosberg I9831 (US).
NUMBER 9 63
64 SMITHSONIAN CONTRIBUTIONS TO BOTANY
the genus Bambusa and proposed the elevation of
its status from section to subgenus.
DISTRIBUTION.-AS documented at present, the
recognized New World components of this arm of
the genus Bambusa exhibit an aggregate recorded
distribution extending from Mexico to all countries
of Central and South America, excepting Chile.
They reach their greatest spontaneous development
under mesophytic conditions in frost-free areas at
elevations not known to greatly exceed 1500 m.
Some species show a remarkable tolerance of, or
ability to thrive in, soil that is saturated with water.
I have observed that, in the general vicinity of
Chinchina, Department of Caldas, Colombia, plants
of Bambusa guadua occupy, spontaneously, practically
every available type of ecological niche.
Guadua philippinensis Gamble, an Old World
species of diverse, as yet incompletely known,
affinities, is known only by the type-collection (C. V.
Piper 475). This collection was made 15 May 1911,
at Mati, District of Davao, island of Mindanao. No
information concerning the ecological setting of the
plant has come to light.
Agrostologists who find it difficult to accept
Hackel’s disposition of the genus Guadua here
adopted may be helped by the following words of
Davis and Heywood (1963:106) quoted from their
treatment of the category “genus”: “When in doubt
as to whether to accord generic rank to a group,
there is much to be said for the subgenus as a
suitable category; it draws attention to the group
in the classification, and at the same time allows
people to continue to use the old binomial.” Final
emphasis added.
Annotated Checklist of Binomials
Recognized Herein as Belonging to
the Subgenus Guadua
NEW WORLD BAMBOOS
Improved perspectives resulting from the development
of more complete images of the taxa
represented by some of the binomials in the suggested
synonymies may give occasion for future
revision in some cases.
1. Bambusa aculeata (Ruprecht) Hitchcock [as
Guadua aculeata Ruprecht ex Fournier, 1881: 130.
Guadua aculeata var. liebmanniana E.-G. Camus,
Bambos], 1913:387 (Figure 28 k, 1).
1913, I:112.
Guadua inermis Ruprecht ex Fournier, 1881.129.
Guadua intermedia Ruprecht ex Fournier, 1881: 130.
Bambusa aculeata appears to represent a northern
extension of Bambusa guadua (Guadua
angustifolia) clearly distinguishable from the latter
only by means of vegetative features. I have not
been able to distinguish Guadua inermis from
Guadua aculeata by means of inflorescences in their
respective type-collections. Contrary to an assertion
in the original description of Guadua inermis, the
lemmas in the type-specimen are entirely glabrous,
just as they are in the type of Guadua aculeata. If
Fournier’s descriptive term “inermis” and the
vernacular name “caiia mansa” quoted by him are
authentic, then Guadua inermis may prove to be
an unarmed form of Guadua aculeata from the
northern range of that species. I do not at present
see any basis for giving a distinctive taxonomic
status to either Guadua inermis or Guadua
aculeata var. liebman.niana.
In 1877 Fournier (p. 198) published a paper
entitled “De la modification des envelopes florales
des GraminCes suivant le sexe de leurs fleurs” in
which he referred to the spikelets of Guadua
aculeata Ruprecht as embodying “a form of
polygamy.” When, however, Fournier cited Guadua
aculeata as the source of an illustrative example,
this binomial had not yet been validated or given
taxonomic content by the publication of a formal
description.
For descriptions of this same phenomenon as
observed in bamboos of the genera Schizostachyum
and Gigantochloa, among others, see McClure
(1934:544ff; 1966b:100, 117) and Arber (1934:133).
Instead of being considered as a reliable source of
taxonomic characters, however, this morphological
deviation should perhaps be seen simply as representing
progressive transitional morphogenetic
FIGURE 28.--Bambusa guadua Humboldt and Bonpland
(A-J) and B. aculeata Ruprecht ex Fournier (K-L) . A, Leafy
flowering twig, x 0.6; B, sterile leafy twig, x 0.6: c, pair of
pseudospikelets, x 1.2; D, diagram of longitudinal section
of pseudospikelet; E, bract and bud subtended by it at base
of pseudospikelet, x 1.8; F, floret, x 3; G, lodicule complement,
x ca. 5.4; H, stamen, x 6; x, gynoecium, x ca. 6;
J, diagram of cross-section of floret; K, fruit, embryo side,
x ca. 6; L, fruit, hilum side, x ca. 6. Drawing A based on
Cobin 1167 (US), B-J on McCEure 21228 (US), and K, L on
iMcClure 21556-A (US) .
NUMBER 9 65
66 SMITHSONIAN CONTRIBUTIONS TO BOTANY
changes in sexual expression that become potential
during the later stages of the ontogeny of indeterminate
(iterauctant) inflorescences in several
bamboo genera.
2. Bambusa amplexifolia (Presl) Schultes f., in
Roemer and Schultes, 1830:1348.
Guadua amplexifolia Presl, 1830:256.
Plants of Bambusa amplexifolia are very thorny
throughout most of the known range of the species
(which extends from Venezuela and Colombia to
Mexico) but progressively less thorny forms appear
from El Salvador northward, and a completely
unarmed form is found in the State of Sinaloa
(specimen at US under McClure 21200). No mention
of thorns appears in the original description
of this species, the type of which was collected by
Haenke from an unrecorded locality in Mexico.
Cf. similar notes regarding Guadua inermis under
Bambusa aculeata.
3. Bambusa barbata Trinius, 1835:627.
Nastus barbatus (Trinius) Ruprecht, 1839:41, pl. 17.
4. Bambusa capitata Triflius, 1835:626; 1836,
Schizostachyum capztatum (Trinius) Ruprecht,
Guadua ? capitata (Trinius) Munro, 1868:81.
Chromosome counts of 2n=ca. 36 and 2n=46
were reported by Gould and Soderstrom (1967,
under Guadua) for Brazilian collections of this
species.
Guadua distorta (Nees) Ruprecht, 1839: 131, pl. 16;
111: PI. 337.
1839 [1840: 136-137, pl. 17, fig. 46.1
5. Bambusa distorta Nees, 1834:470.
fig. 59.
6. Bambusa glaziouii Hackel, 1903c: 194.
glaziowii], 1913, I: 108.
combination.
Guadua glaziouii (Hackel) E.-G. Camus [as
7. Bambusa glomerata (Munro) McClure, new
Guadua glomerata Munro, 1868:79.
Non invalidatus by Bambusa glomerata Royle ex
iMunro (1868:147) [nom. nud. ex sched.] as synonym
of Dendrocalamus strictus Nees. Cf. ICBN,
1961, Art. 64.
8. Bambusa guadua Humboldt et Bonpland,
1808:63, pl. 20 (Figures 27, 28a-j, 29).
Guadua angustifolia Kunth, 1822b: 253.
Nastus guadua (Humboldt et Bonpland) Sprengel,
‘3. Bambusa latifolia Humboldt et Bonpland,
Guadua latifolia (Humboldt et Bonpland) Kunth
The description under the name Guadua latifolia
in Munro (1868:78), and the illustration under the
same name in Doell, in Martius, (188O:pl. 49), are
based on Spruce 1954, a collection that appears to
represent a species quite distinct from the plant
represented by the nomenclatural type of G.
latifolia.
Nastus latifolia [!I (Humboldt et Bonpland)
10. Bambusa longifimbriata (E.-G. Camus) Mc-
Gziadzra Zongifimbriata E.-G. Camus, 1913, I: 113.
A specimen from the type-collection of this
species (Glaziou 5717) (US from C, where it is
labeled Guadua refracta Munro) differs from the
type of the latter species by having the keels of the
palea prominently winged.
11. Bambusa longifolia (Fournier) McClure, new
The basionym of the combination Arthrostylidium
longifolium (Fournier) E.-G. Camus (1913, 1:68)
is Arundinaria ? longifolia Fournier (1881: 131) the
type of which is a sterile Liebmann specimen cited
by Fournier as “Jicaltepec, Aprili, Lieb.” In his
description of Arthrostylidium longifolium, Camus
(1913, I:68) combines with his verbatim French
translation of Fournier’s description of Arundinaria
longifolia a description, in French, of a flowering
1825, vol. 2:113.
1808:67, pl. 21.
1822: 254.
Sprengel, 1825, 11: 113.
Clure, new combination.
combination.
FIGURE 29.-Barn busa guadua Humboldt and Bonpland, sensu
lato. A, Base of culm with its rhizome and the precociously
developed necks that help the root system support the culm,
which may be as much at 37 meters tall and 20 cm in
diameter, x ca. 0.06; B, tangential slice from internode and
nodes v and VI of a culm ca. 3.5 cm in diameter, x 0.6; C,
sector of cross-section of culm wall at internode IV, x 0.6;
D, apex of culm sheath, adaxial aspect, x 0.6; E, basal part
of first-order branch from midculm node, with buds still
intact, x ca. 0.6; F, second-order branch with buds still
intact, x ca. 0.6; 6, third-order branch showing spines and
foliage, x ca. 0.6; H, thorn complement with leafy twig
arising from proximal node of thorn, x ca. 0.6; I, J,
examples of extreme forms of leaves, x ca. 0.6; K, apex of
leaf sheath and base of blade, enlarged. Drawings A, B based
on McClure 21232 (US) and C-K on McClure 21215 (US).
NUMBER 9 67
68 SMITHSONIAN CONTRIBUTIONS TO BOTANY
specimen which he illustrates and cites as “Las
Sedas. Altitude 2000 m. (Pringle).” Available
material from the type and other collections of
Arundinaria longifolia Fournier matches specimens
of Arthrostylidium spinosum Swallen (1938:6)
whose type I identify as a species of Bambusa
subgenus Guadua. The name now becomes Bambusa
longifolia. The Pringle specimen cited by Camus
(of which I have three sheets available at US)
turns out to be a duplicate of Pringle 6742.
This I identify as belonging to the newly described
subgenus Otatea of the genus Yushania. Meanwhile
numerous specimens of bamboos pertaining to this
new subgenus are filed in herbaria around the
world under the name “Arthrostylidium longifolium
(Fournier) E.-G. Camus.”
12. Bambusa macrostachya (Ruprecht) McClure,
new combination.
Guadua macrostachya Ruprechb, 1839:39, pl. 15:
Guadua dioica Steudel, 1854: 334, teste Doell, in
Martius, 1880: 182.
13. Bambusa maculosa Hackel, 1903c: 196.
Guadua maculosa (Hackel) E. -G. Camus, 1913,
I: 106.
14. Bambusa paniculata (Munro) Hackel, 1903c:
195.
Guadua paniculata Munro, 1868:85.
Bambusa munroi Hackel, 1909b3374.
Non invalidatus by Bambusa paniculata Willdenow
ex Munro (1868:123) [nom. nud. ex sched.]
as synonym of Nastus borbonicus Gmelin (cf. ICBN
1966, Art. 64).
15. Bambusa paraguayana (Doell) Bertoni, 1918:
159.
Guadua paragunyana Doell, in Martius, 1880: 179.
16. Bambusa refracta (Munro) McClure, new
Guadua refracta Munro, 1868:84.
17. Bambusa spinosissima Hackel, 1903c: 197.
Guadua spinosissima (Hackel) E.-G. Camus, 1913;
I: 112.
18. Bambusa superba (Huber) McClure, new combidation.
Guadua superba Huber, 1904:479.
19. Bambusa swalleniana McClure, new name.
Guadua spinosa (Swallen) McClure, 1954:82 (not
fig. 35.
combination.
Bambusa spinosa Roxburgh, 1832: 198.)
Arthrostylidium spinosum Swallen, 1938:6.
20. Bambusa tagoara Nees, in Martius, 1829:532.
Guadua tagoara (Nees) Kunth, 1834:611.
21. Bambusa tessmannii (Pilger) McClure, new
Guadua tessmannii Pilger, 1924: 124.
combination.
22. Bambusa tomen.tosa (Hackel and Lindman)
Guadua tomentosa Hackel and Lindman, in
McClure, new combination.
Lindman, 1900:20, pl. 12.
23. Bambusa trinii Nees, 1834:469.
Guadua trinii (Nees) Nees ex Ruprecht, 1839:40,
pl. 15, fig. 38.
Guadua trinii var. p scabra Doell, in Martius,
1880: 179.
Bambusa tacuara Arechavaleta, 1897:550, pl. 72;
teste Parodi, 1936:239.
24. Bambusa iiograndensis Dutra, 1938:147, fig. 1.
Guadua riograndensis (Dutra) Herter, 1941:49.
Guadua ribbentropii Herter, 1940: 148 (based on
Bambusa tacuara Arechavaleta).
25. Bambusa venezuelae (Munro) McClure, new
Guadua venezuelae Munro, 1868:86.
combination.
26. Bambusa uirgata Trinius, 1835:624.
Guadua uirgata (Trinius) Ruprecht, 1839:40.
27. Bambusa weberbaueri (Pilger) McClure, new
Guadua weberbaueri Pilger, 1905: 152.
combination.
OLD WORLD BAMBOOS
28. Bambusa philippinensis (Gamble) McClure,
Guadua phi lip pinensis Gamble, 191 3:203.
new combination.
Checklist of
Names of Congeneric New World Taxa
of Unresolved Specific Status
1. Bambusa chacoensis Rojas, 1918:157.
2. Guadua fascicularis Doell, in Martius, 1880: 186.
3. Guadua lindmani E. -G. Camus, 1913:1113.
4. Guadua polyclados Doell, in Martius, 1880: 182.
Guadua sp., Lindman, 1900:22, pl. Ilc.
NUMBER 9
Annotated Checklist of
New World Species Here Excluded
from the Subgenus Guadua
1. Guadua exalata Doell (in Martius, 1880:181)
is Arthrostylidium longiflorum Munro (1868:
4 1).
2. Guadua ? flabellata Fournier (1881: 131).
Arundinaria flabellata (Fournier) McClure (in
Maguire, Wurdack, et al., 1964: 162), species
sedis mihi etiam nunc incertae; fortasse ad
genus Yushaniam subgenus Otateam attinit.
Guadua pallescens Doell (in Martius 1880: 186)
syn. Bambusa pallescens (Doell) Hackel (1908:
160), identified by Bentham (in Bentham and
Hooker, 1883: 1210) as Bambusa pallida Munro,
is Bambusa tuldoides Munro (1868:93), an Old
World species naturalized in Brazil.
Guadua parviflora J.S. Presl (in K. B. Presl,
1830:257). See Aulonemia paroiflora (Presl)
McClure.
Guadua perligulata Pilger (in Diels, 1937:8) is
Chusquea perligulata (Pilger) McClure, new
combination.
Chusquea Kunth
FICURFS 30, 31
Chusquea Kunth, 1822a: 151; 1822b:254; 1829:138; 1833:427.-
Nees, 1834:484.-Endlicher, 1836: 102.-Ruprecht, 1839:30
[1840: 1201.-Munro, 1868:52.-Doell, in Martius, 1880: 194.-
Bentham, in Bentham and Hooker, 1883: 1209.-Hackel, in
Engler and Prantl, 1887:93.-Arechavaleta, i8!37:542.-
McClure, in Swallen, 1955:86; 1957:202.
Rettbergia Raddi, 1823: 17.-Nees, in Martius, 1829:535.-Nees,
1834:486.-McClure, 1957:207.
Dendrugrostis Nees [as subgenus], 1834:487.-McClure, 1957:
202.
Plants of unicespitose or multicespitose or diffuse
habit; either unarmed throughout or (as in C.
Jendleri Munro; cf. McClure, 1966b, fig. 15) the
culms armed by a ring of short, sharp, aborted roots
at several of their lower, above-ground nodes.
Rhizomes either wholly pachymorph (as in most
known species) or wholly leptomorph (as in C.
simpliciflora Munro) or both pachymorph and
leptomorph in the same plant (as in C. fendleri).
Culms either self-supporting below and broadly
arched or clambering above (as in C. scandens
Kunth), or wholly dependent upon external support
from other vegetation (as in C. simpliciflora
69
Munro); the internodes pithy to the very center
and (as in Zea mays L.) lacking a natural lumen,
cylindrical or terete, in most species shallowly
sulcate for some distance above the locus of insertion
of a complement of buds or branches-rarely
(commonly in some variants of C. pinifoliu Nees)
sulcate or noticeably flattened all the way from
one node to the next. [In some species, as in C.
pittieri Hackel, the pith in old culms may become
shrunken or broken down, leaving an irregular
central passage without a membrane-lined inner
surface. This same change may explain discrepant
descriptions of C. uruguayensis Arechavaleta, whose
culms are described by Arechavaleta (1897:546) as
solid, and by Parodi (1941:334) as hollow.] Branch
complement at midculm nodes unrestricted pleioclade,
the initial components-developed from
separate primary buds of two size categories in
constellate insertion, the smaller ones usually many
(rarely only 2)-comprising axes of two size categories
(major and minor) the major ones solitary
and (when developed) strongly dominant (similar
in form to the mother culm and, particularly in
scandent species, often approaching it in size) the
minor ones (rarely only 2) usually numbering
several to many and, when many hiding the large
central bud from sight when it does not germinate.
Leaves (leaf sheath blades) with transverse veinlets
in some species (as in C. lanceolata Hitchcock)
clearly evident, in others obscure, and in most
species not at all manifest, externally.
Inflorescences semelauctant, terminal to leafy or
leafless branches, or to the culm itself, typically
with a strong, variously branched central axis,
usually paniculate, rarely either racemose or
capitate; prophylls and bracts usually lacking entirely,
bracts subobsolete in a few species (e.g., C.
bambusaeoides (Raddi) Hackel). Transitional
glumes at the base of the spikelet usually 4 (-5, as
sometimes in C. bambusaeoides): I and II “empty
glumes” (these in some species subobsolete), III,
IV and (-v) “sterile (empty) lemmas.” Spikelets
typically containing but one perfect floret-occasionally
2 in some species (e.g., C. andina
Philippi, teste Munro (1868:58); and C. tenella
Nees, teste Nees (1834:493); specimens under
McClure 21292 (C. oxylepis (Hackel) Ekman) commonly
show an aborted floret in the axil of transitional
glume IV (“sterile lemma” II)-the rachilla
70 SMITHSONIAN CONTRIBUTIONS TO BOTANY
NUMBER 9 71
FIGURE 30.-Chusquea scandens Kunth. A, Leptomorph rhizome
turning up to form a culm, x 0.6; B, lower segment of
a small culm with buds, x 0.6; c, midculm node showing
complement of branch buds, front view, x 1.2; D, same, in
profile, x 1.2; E, apex of culm sheath, abaxial view, x 1.2;
F, apex of culm sheath, adaxial view, showing ligule; G,
midculm node with fully developed branch complement,
x 0.6; H, auxiliary branch, showing leaves at apex, branch
sheath, and a complement of branches emerging extravaginally,
x 0.G; I, apex of leaf sheath with petiole and base
of leaf blade, x 12. All drawings based on McClure 21414
(US).
very short and not prolonged behind the palea of
the terminal perfect floret. [My personal impression
in this regard agrees with that expressed by
Bentham (in Bentham and Hooker, 1883:1209) in
Latin: “According to Doell, the rachilla in C.
tenella and other species [of this genus] is prolonged
beyond the apical floret a condition which
we have not seen in our specimens.” (Emphasis
added by F. A. M.)] Fertile lemma fully embracing
its palea only basally at maturity. Palea gaping
antically, in most species dorsally convex, narrowly
sulcate and more or less noticeably 2-keeled toward
the usually emarginate or bidentate or bimucronate
apex, rarely (as in C. capitata Nees) very similar
to its lemma, i.e., compressed and l-keeled (not
sulcate) dorsally and apically entire. Lodicules 3,
the anterior two appreciably asymmetrical and
paired, the posterior one symmetrical and smaller.
Stamens 3, the filaments filiform, free. Stigmatic
branches 2, one of them rarely (as sometimes in C.
scandens Kunth, for example) divided above the
point of its union with the other one. Mature
fruit (rarely found) a small, oblong or fusiform
cuspidulate or bicorniculate caryopsis with a
pergamineous pericarp of uniform thickness, the
sulcus and the embryotegium showing different
degrees of conspicuousness in different species.
ETYMoLoGY.-The name Chusquea is the Latinized
version of a vernacular term used by aboriginal
natives of Ecuador and Colombia to designate
plants of this genus.
TYPE-sPECIEs.-ChUsqUeU scandens Kunth (1822b:
254); Nastus chusque Humboldt, Bonpland, and
Kunth (1816, I:201).
genus Neurolepis share a spikelet structure (Figure
41 c-L, N) that is basically similar to that universal
in Chusquea, members of the latter genus diverge
RELATIoNsHIPS.-Although all members of the
conspicuously from Neurolepis by their bambusoid,
deciduous leaf blades, their branched culms with
efistulose internodes, and their more prominently
developed rhizomes.
Plants of the closely related genus Swallenochloa,
are distinguished from plants of the genus Chusquea
by the relatively weak development of their
rhizomes: the usually stiff, erect culm habit; the
hollow structure of the culm internode; the typically
horizontal and level (not constellate) insertion
of branch buds at culm nodes, and the correspondingly
different branching habit of the culm; the
more or less strongly xeromorphic (leathery) texture
of the leaf blades. The very narrow profile of
the inflorescences of Swallenochloa is not diagnostic,
however.
The existence of noticeable affinities or similarities
between members of the genus Chusquea and
members of the genus Arundinaria is suggested by
Trinius (1835:617), Baillon (1894:253), and Beadle
(in Bailey, 1914:449). As far as the reproductive
apparatus is concerned, similarities between members
of these highly dissimilar genera are limited to
some aspects of the lodicules and the stamens.
As far as elements of the vegetative apparatus are
concerned, the only noticeable similarities are
shown by those species of the respective genera
that share the following three features: (1) leptomorph
rhizomes; (2) leaf blades with conspicious
transverse venation; and (3) frost-hardiness. These
three features are not known to be associated in
any species of any one of the other bamboo genera
of the New World. I do not construe their association
together in exceptional species of Chusquea
to be evidence of a close phylogenetic relationship
or affinity between the two genera, Chusquea and
Arundinaria.
Nees (1834:467, 468) gave Chusquea Kunth four
subgenera: (1) Platonia (Kunth) with C. (Platonia)
elata (Kunth) listed as type; (2) Chusquea with
C. scandens Kunth listed as type; (3) Rettbergia
(Raddi) with C. (Rettbergia) gaudichaudii Kunth
listed as type; and (4) Dendragrostis Nees with C.
(Dendragrostis) pinifolia Nees listed as type. Subsequent
authors have not maintained taxonomic
status for the subgenera proposed by Nees. The
originally monotypic content of Platonia Kunth
(non Platonia Martius) was restored by Meisner

NUMBER 9 73
FIGURE 31.-Chusquea scandens Kunth (A-M) and C. oxylepis
(Hackel) Ekman (N-P). A, Leafy flowering complement of
branches of secondary order, x 0.6; B, inflorescence terminating
a primary branch, x 0.6; c, terminal ramifications of
inflorescence with persistent transitional glumes (“empty
glumes” I and 11), x 7.5; D, spikelet with its pedicel, x 7.5;
E, transitional glumes 111 and IV (”sterile lemmas”), x 7.5;
F, lemma, x 7.5; G, palea, x 7.5; H, lodicule complement,
x 9; I, stamen, x 7.5; J, gynoecium, x 9; K, terminal
branches of inflorescence, x 7.5; L, fruit, embryo side, x
7.5; XI, fruit, hilum side, x 7.5; N, fruit, embryo side, x 7.5;
0, fruit, hilum side, x 7.5; P, fruit, embryo side, x 7.5; Q,
diagram of spikelet. Drawing A-J based on McClure 21414
(US), K on Bonpland 3386, type-species (P) , L, M on Fosberg
22386 (us), and N-P on McClure 21292 (us).
(1843, I:426; 1843, II:325) to its original status as
genus, with a new name, Neurolepis.
As I visualize the natural boundaries of the
genus, Chusquea embraces taxa that’give it a polymorphic
composition. One arm of this diversity
induced Raddi to publish Rettbergia as a monotypic
genus. As observed by Nees (in Martius,
1829:536) a number of the features attributed to
the type-species of Rettbergia by Raddi (1823:18,
pl. 1: fig. 1) suggest faulty observation on the part
of its author. Subsequent study of numerous available
specimens collected from the type-locality
confirms the opinion-first expressed tentatively by
Nees (in Martius, 1829:536) and tacitly affirmed
by Kunth (1830:331), Ruprecht (1839:34), Munro
(1868:67), and Doell (in Martius, 1880: 198)-that
the type-species of Rettbergia does not fall outside
the natural boundaries of the genus Chusquea.
However, the trivial name of the type-species of
Rettbergia (R. bambusoides [as bambusaeoides]
Raddi) was not formally incorporated in the genus
Chusquea until its claim to priority over C. gaudichaudii
Kunth was given expression in the combination
C. bambusoides (Raddi) Hackel (in Wettsteii,
ed., 1908:81).
Chusquea bambusoides is joined by C. capituli-
Pora, C. sellowii, and C. capitata to form a clearly
recognizable but apparently not clearly disjunct
arm of diversity, in which C. capitata stands most
sharply distinguished from other members of the
genus by the form of the palea, which here approaches
that of the lemma. Another arm of diversity
is rendered clearly disjunct from Chusquea
proper principally by distinctive features of the
vegetative apparatus. It is represented by Chusquea
depauperata, C. spicata, C. subtessellata, and C.
weberbaueri which now constitute the published
content of the genus Swallenochloa, q.v.
Chusquea pinifolia (Nees) Nees (1834:4WO),
Awndinaria pinifolia Nees (in Martius, 1829:525),
represents a taxon of special interest. Its polymorphic
character finds expression in the following
validly published names: C. baculifera Alvaro da
Silveira (1919:99), C. heterophylla Nees (1835:488)
with three variants named by Doell (in Martius,
1880:207), and C. pinifolia var. heterophylla (Nees)
Hackel (in Wettstein, ed., 1908:82). Since, however,
they apparently have not been shown to represent
morphologically disjunct entities, they are currently
regarded as synonyms of a highly polymorphic
C. pinifolia. In the habit of the plants, the form of
the midculm branch complement, and the form of
the inflorescences, some of the divergent forms of
C. pinifolia bear superficial resemblance to members
of the genus Swallenochloa. Their basic
morphology, however, keeps plants of this taxon
clearly within the natural limits of the genus
Chusquea, their aggregate range in distribution
extending from slightly above sea level to elevations
(in the midtropics) reaching 3200 m, in generally
mesophytic sites between about 47OS latitude and
Mexico, and on islands adjacent to this area.
DIsTRIsuTIoN.-Documented in herbaria by specimens
of more than 100 described species, the genus
Chusquea is represented by one or more of these
in the indigenous flora of every country of the
New World from Mexico to Chile and Agentina,
and that of many of the islands as well.
The aggregate altitudinal coverage recorded for
its members carries the genus from approximately
sea level to the lower limit of perpetual snow.
According to Munro (1868:61) Jameson (in the
field notes to a collection under Jameson 9) recorded
the altitudinal range of Chusquea fendleri
in Ecuador as “a litore usque ad 12,000 ped. s. m.”
Chusquea scandens Kunth was collected by W. H.
Camp (E-1778 and E-1824, US) in Ecuador at an
elevation of 11,000 ft (3300 m) and C. andina
Philippi was collected by Germain in Chile “ad
limitem nivis perpetuae” (Munro, 1868:58).
Ranges of tolerance and of requirements, in
terms of ecological factors, are not adequately
recorded either in the available field notes or in
the literature; but the entire gamut covered by
members of the genus Chusquea must be very
comprehensive. According to Brade (1956) as re74
SMITHSONIAN CONTRIBUTIONS TO BOTANY
viewed by F. Markgraf (1958) the zone occupied
by Chusquea pinifolia (Nees) Nees is recognizable
on Itatiaia at elevations ranging from 2000 to 2400
meters. [Hackel (in Wettstein, ed., 1908:82) gives
the altitudinal range of C. pinifolia on Itatiaia
as 1300 m to 2750 m.] The meteorological station
at 2200 meters recorded an annual rainfall (as the
average for a period of 20 years) of 2417 mm
(roughly 95 in) and a temperature range between
an average low of 7” and an average high of 11 O C;
extremes recorded: -6O and 35” C. Brade lists the
following taxa that appear along with Chusquea
pinifolia as characteristic of the vegetation of this
high tableland: Cortaderia modesta, Paepalanthus,
Xyris, Sisyrinchium, Clematis, Fuchsia, Buddleia
speciosa, Fragaria chiloensis, Valeriana, Mikania,
and Erigeron.
Weberbauer (191 1 [ 1945: 1461) summarizes his
notes on the ecological role of Chusquea in the
vegetation of Peru as follows (trans. by FAM):
Chusquea is a genus characteristic of the cloud forest
(“Ceja de IaMonta‘iia”); in the north it extends over to the
western side of the Andes. In its true center of distribution
it sends out extensions that reach down as far as the tropical
forest. The shrubby species of the genus Chusquea exert a
conspicuous influence on the physiognomy of the plant formations
of the cloud forest, and here and there occupy large
areas, suppressing other plants. Their slender branches, from
whose nodes emerge masses of twigs and dense foliage, support
themselves on other woody plants, from which they
hang down in beautiful curves.
It should be borne in mind that Weberbauer’s
concept of the genus Chusquea included the species
treated herein under Swallenochloa, q. v. These
are referred to by him as “shrubby species” (vide
infra.). On the other hand, his concluding sentence
refers to species I retain in Chusquea.
Annotated Checklist of
Bamboos Here Recognized
as Belonging to the Genus Chusquea
1. Ch usquea abietif olia Grisebach, 1864: 529.
Arundinaria (Arthrostylidium?) microclada Pilger,
in Urban, ed., 1907:289.
Apparently this taxon is known only by vegetative
material from the type-locality; the nomenclatural
type-collection remains undesignated. Pilger’s
uncertainty as to the correct generic disposition of
this taxon is patent.
2. Chusquea acuminata Doell, in Martius, 1880:
204.
Chusquea tenuis Glaziou ex E.-G. Camus, 1913,
I:90.
3. Chusquea afinis Munro ex E.-G. Camus, 1913,
I:80, pl. 60: fig. B.
4. Chusquea andina R. A. Philippi, 1858:103.
[In rough notes McClure compares this species
with Chusquea culeou Desvaux, perhaps indicating
that it is synonymous with it. Munro (1868:58-59)
expressed the opinion that this species is very
probably only an alpine form of C. cu1eou.-T.R.S]
5. Chusquea anelythra Nees, 1834:491.
6. Chusquea anelytroides Ruprecht ex Doell, in
Martius, 1880: 206.
7. Chusquea argentina Parodi, 1941:339, pl. 24:
8. Chusquea bambusoides (Raddi) Hackel, in
Rettbergia bambusaeoides Raddi, 1823: 18, 57, pl.
Chusquea gaudichaudii Kunth, 1829: 138. Nomen
Chusquea gaudichaudii Kunth, 1830:331-332, pl.
Nastus bruneus A. N. Desvaux, 1831:211.
Nastus bruneus and some other collections have
the awns of the empty glumes antrorse ciliate or
antrorse scabrous on the dorsal keel of the awn.
This is a variable character.
fig. 4.
Wettstein, ed., 1908:81.
1: fig. 1.
nudum.
78.
9. Chusquea bambusoides var. minor McClure
and Smith, in Reitz, ed., 1967:25.
10. Chusquea bilimekii Fournier, 1881: 132.
11. Chusquea capitata Nees, 1834:489.
Rettbergia capitata Nees was cited as a synonym
of C. capitata in Munro (1868:69) and Doell (in
Martius, 1880: 196). Nees had regarded Rettbergia
as synonymous with Chusquea and published the
species as C. [K (ettergia)] capitata.
12. Chusquea capituliflma Trinius, 1835:613.
The spikelets sometimes contain two florets in
this species.
13. Chusquea capituliflora var. pubescens McClure
and Smith, in Reitz, ed., 1967:28.
14. Chusquea carinata Fournier, 1881: 132.
15. Chusquea culeou Desvaux, in Gay, 1854:450.
NUMBER 9 75
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
Chusquea culeou forma longiramea Parodi,
1941 :343.
Chusquea cumingii Nees, 1835:487.
Chusquea paroifolia R. A. Philippi, 1864:299.
Chusquea decolorata Munro ex Parodi, 1945:
Chusquea deficiens Parodi, 1941:335-338, figs.
2,3, and pl. 22.
Chusquea delicatula Hitchcock, 1927b:309-3 10.
Chusquea discolor Hackel, 1903b: 155.
Chusquea dombeyana Kunth, 1832:553, pl. 191.
Chusquea fasciculata Doell, in Martius, 1880:
Chusquea fendleri Munro, 1868:61.
Chusquea fernandeziana R. A. Philippi, 1873:
577-578. [May be the same as C. Zigulata
Munro.]
Chusquea galeottiana Ruprecht ex Munro,
1868: 59.
65-66.
202-203, PI. 54.
Chusquea galeottiana Ruprecht, in Galeotti,
1842:246. Nomen nudum.
27. Chusquea gracilis McClure and Smith, in Reitz,
ed., 1967:43-44, pl. 8i-k.
28. Chusquen heydei Hitchcock, 1927a:80-81.
29. Chusquea huantensis Pilger, 1920:29-30.
30. Chusquea ibiramae McClure and Smith, in
Reitz, ed., 1967:40-42, pl. 8d-f.
31. Chusquea inamoena Pilger, 1905: 150-151.
32. Chusquea jamesonii Steudel, 1854:337.
33. Chusquea juergensii Hackel, 1909a:325-326.
34. Chusquea lanceolata Hitchcock, 1935: 145-146.
35. Chusquea lehmannii Pilger, 1899:35-36.
Chusqziea pilgeri E. -G. Camus, 1913, 1:83. [Type
cited as “C. leibmanni [sic!] Pilger . . non Fournier,”
error for C. Zehmannii Pilger which is valid,
as is C. liebmannii Fournier.]
36. Chusquea leptophylla Nees 1835:489.
Arthrostylidium trinii sensu Steudel, 1854:336,
in part, non Ruprecht, 1839:119.
Arthrostylidium (?) leptophyllum (Nees) Doell,
37.
38.
39.
40.
41.
42.
in Martius, 1880:175-176.
Chusquea liebmannii Fournier, 1881: 132.
Chusquea ligulata Munro, 1868:62.
Chusquea linearis N. E. Brown, 1901:76. [Probably
equals C. p i n i f o h (Nees) Nees.]
Chusquea longifolia Swallen, 1940:210.
Chusquea longipendula Kuntze, 1898, I11 (3):
Chusquea lorentziana Grisebach, 1874:249-250.
348-349.
43. Chusquea macrostachya R. A. Philippi, 1896:
44. Chusquea mexicana Hackel, 1902:256.
45. Chusquea meyeriana Ruprecht ex Doell, in
Martius, 1880:203-204.
C. meyeriana is closely related to C. acuminata
Doell but differs in the pubescent leaf blades,
prominent external ligule, reflexed inflorescence
branches, and smaller spikelets.
46. Chusquea mimosa McClure and Smith, in
Reitz, ed., 1967:37.
47. Chusquea montana R. A. Philippi, 1864:298-
299.
48. Chusquea muelleri Munro, 1868:65.
49. Chusquea nelsonii Scribner and Smith, 1897:
16.
50. Chusquea nigricans R. A. Philippi, 1865:323-
324.
51. Chusquea oligophylla Ruprecht, 1839: 124, pl.
7: fig. 23.
52. Chusquea oxylepis (Hackel) Ekman, 1913:65,
pl. 4: fig. 6 (Figure 3 1 ~ - ~ ) .
Chusquea bambusoides (Raddi) Hackel subsp.
oxylepis Hackel, in Wettstein, ed., 1908:81-82.
53. Chusquea palenae R. A. Philippi, 1896:350.
54. Chusquea pallida Munro, 1868:65.
55. Chusquea paruipora R. A. Philippi, 1896:349-
350.
56. Chusquea perligulata (Pilger) McClure, new
combination.
Guadua (7) perligulata Pilger, in Diels, 1937:57-58.
57. Chusquea peruuiana E. -G. Camus, 1913, I:88.
Chusquea ramosissima Pilger, 1905: 145, not Lindman
1900:24.
Chusquea sandiensis Pilger, 1920:29.
58. Chusquea picta Pilger, 1905:151.
59. Chusquea pinifolia (Nees) Nees, 1834:490.
Arundinaria pinifolia Nees, in Martius, 1829:525.
.? Ludolfia pinifolia (Nees) A. Dietrich, 1833, II:25.
Chusquea heterophylla Nees, 1834:488.
Chusquea heterophylla var. elonguta Doell, in
Martius, 1880:207.
C h usq uea heterophy lla var. microphylla Doell,
in Martius, 1880:207.
Chusquea heterophylla var. squumosa Doell,
in Martius, 1880:207.
Chusquea pinifolia var. heterophylla (Nees) Hackel,
in Wettstein, ed., 1908 [1906:21].
60. Chusquea pittieri Hackel, 1903b: 153-154.
350-351.
76
Chusquea maurofernandeziana
1892: 64, nomen nudum.
C h usq uea maurof ernanderiana
Camus, 1913, I:86, pl. 56c.
Hackel ex Pittier,
Hackel ex E.-G.
61. Chusquea polyclados Pilger, 1905: 147.
62. Chusquea pubescens Steudel, 1854:337.
63. Chusquea pubispicula Pilger, 1905: 148-149.
64. Chusquea purdieana Munro, 1868:56.
65. Chusquea quila Kunth, 1830:329-330, pl. 77.
Nastus prolifer Desvaux, 1831:211.
Coliquea quila Steudel, in Bibra, 1853: 115.
Chusquea quila var. laxiflora Desvaux, in Gay,
Chusquea intermedia Steudel, in Lechler, 1857:52,
Chusquea quila var. longipila E.-G. Camus, 1913,
66. Chusquea ramosissima Lindman, 1900:24.
C h usquea phacellophora Pilger, 1923:456.
67. “Chusquea” rollotii Berry, 1929: 2-3.
This is a fossil species described from fragmentary
remains in lithified gray shale of the late Tertiary
(Colombia).
68. Chusquea scandens Kunth, 1822:254 (Figures
Nastus chusque Humboldt, Bonpland, and Kunth,
Bambos chusque Poiret, 1817:494.
Chusquea jamesonii Steudel, 1854:337.
Chusquea quitensis Hackel, in Sodiro, 1889:484,
Chusquea quitensis Hackel, 1903b: 154.
Chusquea quitensis var. patentissima Hackel,
Chusquea meyeriana var. patentissima (Hackel)
1854:447.
nomen nudum.
I:198, fig. 61a, nomen nudum.
30, 31 A-M).
1816, I:201.
nomen nudum.
1908:161.
69.
70.
71.
72.
E.-G. Camus, 1913, 1:94.
Chusquea sclerophylla Doell, in Martius, 1880:
200.
Chusquea sellowii Ruprecht, 1839 [1840:125,
pl. 11: fig. 261.
Chusquea serrulata Pilger, 1898: 71 9-720.
Chusquea simpliciflora Munro, 1868:54.
Chusquea simplicifolia Munro ex Hemsley, in
Godman and Salvin, ed., 1885:587, error for
C. simpliciflora Munro.
73. Chusquea sneidernii Asplund, 1939:797-799,
fig. 2.
SMITHSONIAN CONTRIBUTIONS TO BOTANY
74. Chusquea spadicea Pilger, 1899:35.
75. Chusquea spencei Ernst, 1872:262.
76. Chusquea spinosa Fournier, 1881: 131.
77. Chusquea straminea Pilger, 1905: 147-148.
78. Chusquea sulcata Swallen, 1940:209.
79. Chusquea swallenii McClure and Smith, in
Reitz, ed., 1967:44-45, 48, figs. a-c.
80. Chusquea tarmensis Pilger, 1905:151.
81. Chusquea tenella Nees, 1834:492-493.
82. Chusquea tenella var. latifolia Dutra, 1938: 146.
83. Chusquea tenuiflora R. A. Philippi, 1859:206.
Chusquea ciliata R. A. Philippi, 1864:299.
84. Chusquea tenuiglumis Doell, in Martius, 1880:
199-200.
85. Chusquea tenuiglumis var. laxiuscula Doell, in
Martius, 1880:ZOO.
86. Chusquea tenuiglumis var. subcytindrica Doell,
in Martius, 1880: 199-200.
87. Chusquea tonduzii Hackel, 1903b: 155.
88. Chusquea tuberculosa Swallen, 1931: 14.
Chusquea hispida McClure, 1942: 179, fig. 7.
89.
90.
91.
92.
93.
94.
95.
96.
1.
2.
3.
4.
5.
6.
7.
8.
Chusquea uliginosa R. A. Philippi, 1859:207.
Chusquea unifiora Steudel, 1854:337.
Chusquea urelytra Hackel, 1903b: 158.
Chusquea uruguayensis Arechavaleta, 1897:
546-547.
Chusquea valdiviensis E. Desvaux, in Gay,
1854: 446.
Chusquea virgata Hackel, 1903b: 156-157.
Chusquea ? wettsteinii Hackel, in Wettstein,
ed., 1908 [1906:21-221.
Chusquea wilkesii Munro, 1868:63.
Checklist of
Species Here Excluded from
the Genus Chusquea
Chusquea amplopaniculata Steudel (1854:337)
is Dinochloa scandens (Blume) 0. Kuntze.
Chusquea aristata. See Neurolepis aristata.
Chusquea depauperata. See Swallenochloa de-
Chusquea fimbriata. See Athroostachys capitata.
Chusquea glomerata Munro (1868:50) (as synonym
of Merostachys capitata Hooker). See
Athroostachys capitata.
Chusquea humilis. See Swallenochloa spicata.
Chusquea simplissima. See Swal1en.ochloa spi-
Chusquea spicata. See Swallenochloa spicata.
pauperata.
cata.
NUMBER 9 77
9. Chusquea subtessellata. See Swallenochloa sub-
10. Chusquea tessezlata. See Swallenochloa tessel-
11. Chusquea venezuelae. See Arthrostylidium
12. Chusquea weberbaueri. See Swallenochloa
t esse 1 la ta.
lata.
venezuelae.
weberbaueri.
Invalid Species of Chusquea
1. Chusquea caamanoi Sodiro (1881:ll). Nomen
nudum.
Colanthelia McClure and E. W. Smith, new genus
FIGURE 32
Plantae unicespitosae, inermes. Rhizomata pachymorpha.
Culmi habitu staturaque late diversi, aut
parum alti et erecti vel “decumbentes” (teste Hackelii
sub Arundinaria rhizantha) aut “altissimi scandentes”
(teste Trinii sub Arundinaria distans),
nodis valde elevatis et anguste cristatis, nodo medio
unoquoque juventute gemmam solitariam gerenti,
prophyllo gemmae et gemma ipsa in germinatione
simul elongatis. Ramorum complementum ad nodos
culmorum medianos axem ordinis primarii solitarium
continens, idem basi vulgo mox proliferatum,
ramis ordinis superioris axem primarium
plus minusve valde superantibus. Vaginae internodia
culmorum foventes basi cingulis latis post
delapsum vaginarum persistentibus praeditae. Foliorum
laminae venulae transversae extra vulga haud
manifestae.
Inflorescentiae semelauctantes, in speciebus plerisque
vel infirme paniculatae vel racemosae, formis
intermediis variis in ipsa planta interdum praesentibus;
prophylla omnino haud ulla, bracteis
dempta infima vel parvis vel obsoletis.
Glumae transitionales vulgo 2, rarissime vel 1
vel 3. Spiculae raro pauciflorae (ut in Colanthelia
lanciflom) pleraeque pluriflorae, angustissimae et
fragilissimae, apice in anthecio sterili plus minusve
valde depauperato terminantes. Lemma fertile in
maturitate paleam suam basi tantum circumplectans.
Palea dorso 2-carinata et late sulcata, marginibus
vix vel haud imbricatis. Rachillae segmenta
ob fragilitas articulorum (nodorum) suorum in
maturitate facilissime disarticulantia. Lodiculae
typice 3, duae anticae plus minusve asymmetricae
atque geminatae, postica symmetrica et vulgo
minore. Stamina 3, filamentis filiformibus liberisque.
Stigmata 2. Fructus non adhuc suppetens.
Plants unicespitose, unarmed. Rhizomes pachymorph.
Culms of small to medium stature and erect
or decumbent, to very tall and scandent, the nodes
prominent with a narrow crest, each midculm node
giving rise to but a single initial bud; the prophyllum
of the bud growing while the bud germinates,
the primordium producing a single (segmented,
terete) primary axis, this dominant over the axes
of secondary order usually proliferating promptly
from buds typically present at its own proximal
nodes. Sheath at midculm nodes provided with (and
abscising from) a conspicuous persistent girdle
(basal expansion zone). Leaf blades with transverse
veinlets as a rule not manifest externally.
Inflorescences semelauctant, either paniculate or
racemose (sometimes showing in the same specimen
forms intermediate or intermixed between these
two forms), part or all of the inflorescences (or
even whole flowering branches) in a given specimen
sometimes reduced individually to barely more
than a solitary spikelet. Transitional glumes typically
2, rarely 1 or 3. Spikelets usually pedicellate,
typically many-flowered (few-flowered in Colanthelia
lanciflora), very narrow and fragile, terminating
apically in a more or less strongly depauperate
sterile anthecium. Fertile lemma fully embracing its
palea only basally at maturity. Palea 2-keeled and
broadly sulcate dorsally, gaping ventrally. Rachilla
segments (owing to their fragility at the nodes)
easily disarticulating at maturity. Lodicules typically
3, the anterior 2 more or less asymmetrical
and paired, the posterior one symmetrical and
usually smaller. Stamens 3, the filaments filiform
and free. Stigmas 2. Fruit not yet available.
ETYMoLoGY.-The name Colanthelia (f.) is coined
from the Greek kolos, shortened, and anthele,
“plume or panicle of a reed” (R. W. Brown, 1954).
It alludes to a tendency common to all of the
currently recognized species of the genus. There
is a noticeably to strongly manifested tendency
toward a progressive reduction of the reproductive
structures from weak panicles to simple racemes
and, in some cases, even from whole flowering
branches all the way to little more than solitary
spikelets.
78 SMITHSONIAN CONTRIBUTIONS TO BOTANY
NUMBER 9 79
FIGURE 32.-CoZanthelia cingulata (McClure and L. B. Smith)
McClure. A, Rhizome and culm bases, x 0.6; B, branch complement
from midculm range, in flowering state, x 1.2;
c, culm sheath from above midculm range, with developing
branch complement, x 0.9; D, apex of leaf sheath and base
of leaf blade, x 7.2; E, paniculate inflorescence terminating
a leafy flowering branch, x 0.6; F, progressively depauperate
leafy and leafless flowering branches, x 0.6; G, flowering
branch terminating in an inflorescence with mixed paniculate
and racemose branching, x 0.6; H, spikelet with its
pulvinate pedicel, x 2.4; I, transitional glumes, x 6; J,
floret, x 6; K, lumma, x 6; L, palea, x 6; M, lodicule complement,
x 12; N, stamen, x 6; 0, gynoecium, x 12. Drawings
A-E, G-o based on Reitr and KEein 9679 (US), and F on
Duse'n 18011 (US).
TYPE-sPEcIEs.-Colanthezia cingulata (McClure
and L. B. Smith) McClure.
RELATIONSHIPS.-Affinity toward Aulonemia is
weakly suggested by the nature of the midculm
branch complement; this possible affinity is reinforced
by the occurrence of paniculate inflorescences
in some species. Among available examples,
a specimen of Aulonemia haenkei (US 1256334)
from Peru shows an occasional flowering branch
reduced to little more than a solitary spikelet. The
feature to which the name Colanthelia alludes also
comes to light here and there in members of other
genera. Among bamboos of the Old World genera
a few similarly reduced flowering branches sometimes
appear in specimens of Chimonobambusa
marmorea. Other attributes (e.g., racemose inflorescences)
suggest affinity toward Arthrostylidium.
Such divergent attributes are connected by intermediate
expressions to form clines, both from one
species to another and (in some cases) within the
same specimen. In the available material representing
the genus Colanthelia, the morphological
gamut of the reproductive apparatus reaches from
leafy to leafless branches, each terminating in either
a panicle, a raceme, a combination between those
two forms, or a single spikelet. Pertinence to the
genus Colanthelia (among the known bamboo
genera of the New World) is made plain in its
known members (even in their vegetative state) by
the overall delicacy of the plant, with distinctively
small leaf blades, combined with pachymorph
rhizomes, prominent narrow-crested midculm
nodes, a wide girdle at the base of the sheath at
each node, and the unarmed complements with
the primary element clearly dominant. With the
exception of Colanthelia lanciflora, all known members
of the genus have more or less distinctively
long and narrow spikelets with more or less clearly
exposed slender, elongate segments.
DISTRIBUTION.-AS far as their natural distribution
is a matter of record, all of the known species
of Colanthelia are confined to Brazil. Three species
have been reported from Santa Catarina, and
one each from Minas Gerais, Paran$ Rio Grande
do Sul, and S5o Paulo. The recorded altitudinal
range extends from 30 m for Colanthelia cingulata,
to 550 m for C. intermedia, to 710 m for C. gracillima,
and to 1600 m for C. lanciflora.
Annotated Checklist of
Recognized Species of the Genus CoZantheZia
1. Colanthelia burchellii (Munro) McClure, new
combination.
Arthrostylidium burchellii Munro, 1868:43.
Arundinaria burchellii (Munro) Hackel, 1903a:69.
2. Colanthelia cingulata (McClure and L. B.
Smith) McClure, new combination (Figure 32).
Aulonemia cingulata McClure and L. B. Smith, in
Reitz, ed., 1967:50.
3. Colanthelia distans (Trinius) McClure, new
combination.
Arundinaria distans Trinius, 1835:621.
Trinius (1836, III:622) states, incorrectly, that a
terminal rudiment is lacking in Arundinaria distans
Trinius, q.v.
4. Colanthelia intermedia (McClure and L. B.
Aulonemia intermedia McClure and L. B. Smith,
in Reitz, ed., 1967:52.
5. Colanthelia lanciflora (McClure and L. B.
Aulonemia lanciflora McClure and L. B. Smith, in
Reitz, ed., 1967:47.
6. Colanthelia macrostachya (Nees) McClure, new
combination. Arundinaria macrostachya Nees,
1834:48 1.
7. Colanthelia rhizantha (Hackel) McClure, new
combination. Arundinaria rhizantha Hackel,
1909a: 323.
Smith) McClure, new combination.
Smith) McClure, new combination.
Elytrostachys McClure
FIGURES 33-35
Elytrostachys McClure, 1942: 173, figs. 4-6; 1957:ZOZ.
Plants unicespitose; unarmed. Rhizomes pachymorph.
Culms self-supporting below, typically weak
80 SMMITHSONIAN CONTRIBUTIONS TO BOTANY
and pendulous or clambering above; the internodes
cylindrical, hollow. Branch buds at culm nodes
solitary, each containing but a single initial primordium.
Branch complements at midculm nodes
of either restricted or gremial insertion, unrestricted
monoclade, consisting typically of one
strong, central axis and a tuft of more slender ones
arising from lateral buds at proximal nodes of the
initial primordium, these latter buds sometimes
germinating precociously while the apical growth
of the initial primordium remains inhibited. Leaves
(leaf sheath blades) typically with transverse veinlets
not at all or only weakly manifest externally;
oral setae few, rigid, erect or spreading, slender
except at the bulbous base, scabrous.
Inflorescences iterauctant, terminating leafy or
leafless axes, of diffuse form, each axis the bracteate
and prophyllate rachis of a pseudospikelet, the terminal
segment of each rachis serving as the pedicel
of a spikelet. Transitional glumes at the base of
each spikelet none. [The conventional position of
the transitional glumes is here (and in Atractantha,
q. v.) occupied by what I now classify as bracts.
When, in 1942, I first described and illustrated the
genus Elytrostachys, I identified and labeled these
structures as being glumes. However, in the perspective
achieved more recently through a re-study
of the structures in this transitional zone in the
inflorescence of species of Elytrostachys in comparison
with their counterparts in species of A tractantha,
I now see them in both genera as empty bracts
that are terminal to the series of bracts that subtend
buds, rather than as empty glumes that precede
the lemmas. The elongated terminal segment
of the rachis that follows the two short internodes
on which the emtpy bracts are inserted serves as the
pedicel of the spikelet that is terminal to it. Having
noted this, we can confirm. the lack of empty
glumes by recalling that the natural loci of their
insertion would be found immediately preceding,
and close to, that of the first lemma.] Spikelets
each made pedicellate by the segment of the rachis
it terminates, and each typically containing but a
single perfect flower (sometimes 2 in E. czavigera),
promptly disarticulating at maturity immediately
below the locus of insertion of a fertile lemma, the
rachilla prolonged behind the terminal perfect palea
in a bristle-like segment bearing a minute rudiment
of a sterile anthecium. [When the spikelet
contains but a single perfect floret, the prolongation
of the rachilla is bristle-like and bears a minute
rudiment; but when the spikelet contains more
than one perfect floret, the prolongation of the
rachilla behind the palea of the terminal perfect
floret may be more robust than bristle-like, and
then will bear a sterile anthecium more substantial
than rudimentary.] Lemma in functional florets
fully embracing its palea only basally at maturity.
Palea gaping antically, dorsally canaliculate in
spikelets containing but a single perfect floret,
broadly sulcate in the lower floret when the spikelet
contains two perfect florets. Lodicules 3, subequal
or unequal, the anterior 2 asymmetrical and paired,
the posterior one symmetrical and smaller. Stamens
6, the filaments filiform, free. Stigmatic branches
2. Fruit a fusiform or lagenoid, rostrate, sulcate
caryopsis, the pericarp coriaceous, of even thickness
below, thickened at the apex, the basal position of
the embryo clearly manifest.
ETYMoLoGY.-The name Elytrostachys, formed
from the Greek elytro (combining form of eZytron)
sheath, cover (elytra = wingcover of beetles) and
stachys, spike, alludes to a fancied resemblance of
the lemmas to the wing-cover of a beetle.
TYPE-sPECIES.-E~ytrostachys typica McClure.
RELATIONSHIPS.-The only known New World
genus to which Elytrostachys appears to be closely
related is Atractantha. These two genera share the
following features: inflorescences of a peculiar form
characterized by rachises with long terminal segments,
each of which serves as the pedicel of an
abscissile spikelet; the absence of the transitional
glumes commonly found at the base of each spikelet,
and the presence, in their stead, of a bract inserted
on each of the two short internodes that
precede the elongated terminal segment of each
rachis; each spikelet typically containing but a
single perfect floret followed by a rudiment of a
sterile anthecium borne on the tip of a bristle-like
prolongation of the rachilla. From known members
of Atractantha, known members of Elytrostachys
differ in the following features: midculm
FIGURE SJ.--Elytrostuchys clavigera McClure. A, Portion of a
seedling plant, x 0.6; B, culm sheath at midculm node of
young culm shoot, x 0.6; c, midculm node and basal part
of a typical branch complement, x 0.6; D, leafy twig with
terminal inflorescence, x 0.6; E, two nodes from upper
part of leafless flowering culm, x 0.6. All drawings based
on McClure 21478 (US),

J
D
0
B
~
NUMBER 9 83
FIGURE 34.-Elytrostachys clauigera McClure. A, Leafy twig
with pseudospikelet after the leaf blades and the first spikelet
have fallen away, x 1.5; B, diagrammatic array of the parts
of a flowering branch, showing pseudospikelets in various
stages of development; c, pseudospikelet with the terminal
spikelet mature and disarticulated, x 1.5; D, laminiferous
bract (blade lacking) that subtended a young pseudospikelet,
x 1.5; E, young pseudospikelet; F, transitional glumes, x 1.5;
G, pedicel of the one-flowered spikelet, x 1.5; H, diagram
of cross-section of a floret; I, lemma, lateral aspect, x 1.5;
J. palea, lateral aspect, x 1.5; K, lodicule complement, x 3;
L, stamen, x 3; M, gynoecium, x 3; N, mature fruit, hilum
side showing sulcus, x 1.5; 0, mature fruit, embryo side
showing embryotegium, x 1.5. All drawings based on
McClure 21478 (US). For detailed caption for B, see McClure.
1942:178, fig. 6.
branch complements of restricted or gremial insertion,
each arising from a solitary branch bud containing
but a single initial primordium; auricles
and oral setae of leaf sheaths not only much more
conspicuously developed, but of different morphological
configuration; in the presence of at least two
empty bracts preceding the elongated terminal segment
of each rachis; in the loosely convolute, more
or less inflated form of the anthecia; and in the
presence of twice as many stamens in each flower.
As far as known to date, the respective geographical
ranges of the two genera are widely disjunct.
DISTRIBUTION.-of the two recognized species, E.
typica has been reported only from the borders of
a forest at the type-locality, El Limbn, Venezuela.
Elytrostachys clauigera McClure has been collected
in Colombia and all countries northward to Honduras,
and from situations at reported elevations
from 200 to 1500 m, described as river banks, old
river terraces, wet ravines, crest of range, edge of
forest, opening in forest. Both species are represented
as flowering and fruiting freely.
Key to the Species of Elytrostachys
la. Leaf blades scabrous toward the base on the upper surface; laminiferous bracts silkypubescent
in part; bladesless empty bracts (originally called empty glumes) I to 14 mm,
11 to 18 mm long; terminal segment of rachis to 17 mm long and pubescent in part;
lemma to 17 mm long; ovary hispidulous toward apex ........................................ 1. E. clavigera
lb. Leaf blades glabrous on both surfaces; laminiferous bracts glabrous; bladeless empty
bracts (originally called empty glumes) I to 15 mm, II to 30 mm long; terminal segment
of the rachis to 6.5 mm long and entirely glabrous; lemma to 23 mm long; ovary entirely
glabrous ......................................................
Glaziophyton Franchet
FIGURE 36
G l ~ z i o p h y t ~ n Franchet, 1889:277.-McClure, 1957203.
Arundinaria sect. Glaziophyton (Franchet) Hackel, in Engler
and Prantl, 1897:46.-Lamson-Scribner, 1899:Q.
Plants multicespitose, unarmed. Rhizomes pachymorph,
the neck of an occasional one much elongated,
thus creating an opening between tufts of
culms arising either by tillering, or from a cluster
of rhizomes with short necks. Culms strictly erect,
the. internodes (in leafless, juncoid culms) cylindrical,
the hollow interior partitioned by numerous
thin but firm septa of pith without vascular tissue,
the first above-ground internode disproportionately
elongate, the upper ones very short to obsolete;
juncoid culms in the vegetative state normally
branchless and leafless-only after burning (teste
Glaziou) do small secondary culms arise by tillering
from the stumps and emit leafy branches. Branch
buds at the nodes of these secondary culms solitary.
....... ........................................................ 2. E. t~pica
Vegetative branch complements of restricted insertion,
unrestricted or facultatively restricted monoclade,
with the median (primary) one dominant
(those present in the only extant leafy specimens,
Glaziou 8999 and 17914, not fully developed). Leaf
sheaths with auricles and oral setae weakly developed
or obsolete. Leaf blades with transverse veinlets
manifest to superficial view, more prominently
so on the abaxial surface. Culms in the flowering
state leafless; the very slender, elongate, subequal,
highly ramified flowering barnches borne in basally
appressed pleioclade complements each arising from
a solitary bud and inserted at one of the close-set
distal nodes of the culm and producing gigantic,
nodding, apically pendulous inflorescences (up to
1 m long, teste Franchet).
Inflorescences semelauctant, effusely branched
panicles with a deliquescent rachis, the branches of
all orders solitary, very slender and wiry, each subtended
by a bract, and each bearing a 2-keeled
84 SMITHSONIAN CONTRIBUTIONS TO BOTANY
NUMBER 9 85
FIGURE 35.-Elytrostachys clauigera McClure. A, Midculm
branch bud in early stage of germination, x 0.6; Ba, basal
portion of young midculm branch complement, x 0.6;
eb, basal portion of mature branch complement at upper
node of culm, x 0.6; c, branch complements at upper nodes
of culm in early stage of development, x 0.6; D, middle
portion of sterile leafy first-order axis of midculm branch
complement, x 0.6. All drawings based on McCEure 21478
(US) *
prophyllum at its basal node. Transitional glumes
at the base of the spikelet 0-3; I and 11 “empty
glumes”; III (when present) a “sterile lemma” that
is either empty or subtends a rudimentary flower.
Spikelets petiolate, each containing 2 (1-3) perfect
flowers and terminating in a small, loosely convolute
sterile anthecium (an occasional spikelet
contains several progressively depauperate florets
above the perfect ones); the rachilla segments somewhat
elongate, potentially disarticulating immediately
below the locus of insertion of each lemma
including the terminal, reduced one (s). Palea gaping
antically, dorsally sulcate and 2-keeled. Lodicules
3, the anterior two asymmetrical and paired,
the posterior one symmetrical and smaller. Stamens
3, the filaments filiform, free. Stigmas 2. Fruit
unknown.
ETYMoLoGY.-The name of this bizarre, monotypic
genus honors its discoverer, Auguste Fransois
Marie Glaziou (1833-1906), a French botanist and
plant collector, who made excellent specimens and
critical observations of this plant in the field during
a period of 15 years. Glaziou passed on his observations
to Franchet, apparently by oral communication.
TYPE-sPEcIEs.-Glaziophyton mirabile Franchet,
the sole known representative of the genus.
World bamboos endemic on mountaintops in New
Caledonia) resembles Glaziophyton superficially in
the branching habit of the inflorescence (with each
branch of the inflorescence subtended by a bract
and bearing a prophyllum at its basal node) and in
the manifest tessellate venation of the leaf blades,
but differs profoundly in having the flowering
c u h s leafy, with efistular internodes, and in the
presence of 6 stamens and 3 stigmas in each flower.
The disproportionate elongation of a single lower
internode of every culm-a feature of the typespecies
of Glaziophyton-is a vegetative character
shared by most species (including the type) of
RELATIONSHIPS.-GreSlania Balansa (genus of Old
Myriocladus and several species of Arthrostylidium
(e.g., A. schomburgkii). The peculiar, juncoid nature
of the septate lumina of the culm internodes
in the type-species of Glaziophyton apparently has
no close counterpart in any known species of bamboo.
Glaziophyton shows spikelet characters similar
to those of Myriocladus, and features within the
florets that link it in a general way with Merostachys,
Myriocladus, and Arthrostylidium. The
genus Glaziophyton, however, stands without any
obviously close affinities among the bamboos.
Glaziou assumed the existence of juncoid affinities
in this plant, and on the field label of his first
collection (Glaziou 8999, 7 June 1877) he wrote
a tentative (unpublished) binomial, placing it in
the genus Juncus. Franchet, however, in his discussion
of its relationships, compared Glaziophyton
mirabile to cyperaceous plants, such as Scirpus lacustris
and Cyperus articulatus. It is of interest to
note that both Juncus eflusus Linnaeus and Glaziophyton
mirabile Franchet have been recorded from
the same geographical area and at approximately
the same elevation, in Brazil.
DISTRIBUTION.-EXtant specimens indicate an endemism
restricting this species to open, dry mountain
tops in the general vicinity of Rio de Janeiro.
Glaziophyton mirabile poses interesting problems
for the serious student of the genetic, as well as
the physiological, bases of morphogenesis. During
a period of 15 years, Glaziou observed repeatedly
that the sterile culms were all typically juncoid,
without branches and without foliage leaves. When
the sterile culms were destroyed by burning, however,
secondary culms or tillers arising from the
stumps produced leafy branches. Flowering was observed
on leafless juncoid culms only.
We have here what appears to be an extremely
exaggerated expression of an ontogenetic feature
found in some other bamboos (e.g., Bambusa arundinacea)
where culms in the flowering state
regularly lack foliage leaves from the first. In
Glaziophyton mirabile, however, the phenetic gap
between the flowering state and the vegetative state
of the plant is very much more extreme morphologically,
more persistent, and apparently deeprooted
genetically. The only ecological condition
that is known to trigger the production of shoots
bearing leafy ramifications is a bizarre one, namely,
burning of the aerial part of plants previously re86
SMITHSONIAN CONTRIBUTIONS TO BOTANY
NUMBER 9 87
FIGURE 36.--Glaziophyton rnirabile Franchet. A, Habit of
underground part of plant: rhizomes with elongated necks
(shown in broken outline), double circles represent the
stumps of severed culms, greatly reduced; B, sterile leafless
culm (partly in longitudinal section, to show pith diaphragms)
and two young shoots, diameter x 0.6, length
x ca. 0.9; c, tlp of leafless sterile culm, showing close-set
budless nodes, x ca. 3; D, E, culm sheaths (apical portion
only) from nodes 8 and 13 of sterile culm, abaxial and
adaxial aspects, respectively, x 4.8; F, stumps of burned
leafless plant, with leafy culms that developed by tillering,
after burning, x 0.6; c, foliage leaf and part of its sheath,
abaxial aspect, x 1.8; H, junction of foliage leaf and its
sheath, adaxial aspect, x 1.8; I, portions of two mature
flowering branch complements, the latter with an intact
subtending sheath, x 1.2; J, bract subtending a middle-order
branch of the inflorescence, x 6; K, prophyllum of a
middle-order branch of the inflorescence, x 9; L, diagram of
the structure, in longitudinal section, of the ultimate-order
branching of the inflorescence. Each branch terminates in
a spikelet. The axis of the flower proper is represented by
a narrow cone, bearing the prophyllum-like palea; hl, spikelet,
showing extreme development in length and in number
of functional florets. Spikelets with 2 or 3 functional florecs
are typical, x ca. 2.4; N, transitional glumes, x ca. 12; 0,
floret, x 9; I>, diagram of floret Structure in cross-section; Q,
lemma, x 9; R, palea, x 9; s, lodicule complement, x 9:
T, stamen, x 9; IJ, gynoecium, x 15; v, branch of stigma,
x ca. 120. Drawings A, B based on Glaziou 8999 (A based on
Glaziou’s notes incorporated in Franchet’s description of the
plant), C-E on Glaziou 11383 (P) , and F-v on Glaziou I7914
(P).
maining in the unbranched, perennially leafless
state. How long will the “satiable curiosity” of
technically equipped persons ignore the challenge
to investigate this enigma-in depth!
Merostachys Sprengel
FIGURE 37
Merostachys Sprengel, 1825, 11: 132.-Endlicher, 1836-1840:
102.-Ruprecht, 1839:36.--Munro, 1868:46.-Doell, in Martius,
1880:207.-Bentham, in Bentham and Hooker, 1883:
1209.-Hackel, in Engler and Prantl, 1887:93.-McClure,
19573205.
Plants unicespitose, unarmed. Rhizomes pachymorph.
Culms self-supporting in the lower part,
the slender, sometimes long-attenuate upper part
nodding, pendulous or clambering; internodes
cylindrical, typically hollow, exceptionally, as in
some as yet unidentified species, efistulose or filled
with pith. Primary branch buds solitary at midculm
nodes, each bud giving rise to but a single
initial (primary) axis. Branch complement at mid-
Bradocalumus Nakai, 1933: 10.-McClure, 19573201.
culm nodes monoclade, with its primary element
solitary, flat, unsegmented, roughly ovate or cordate
to triangular in contour, thin, adnate to the
surface of the culm; secondary branches slender,
subsequal, of apsidate insertion, arising from primordia
formed on the margin of the primary element,
and displayed in fan-shaped array. Leaves
(blades of leaf sheaths) with transverse veinlets usually
not manifest externally.
Inflorescences semelauctant, spicate racemes terminating
usually leafy twigs, the rachis excurrent,
bearing spikelets of solitary (in a few species paired
or ternate) insertion and secund orientation, the
distal 1 or more spikelets typically depauperate or
rudimentary; bracts subtending spikelets (excepting
the Jowrrmost one) generally reduced, rudimentary,
or obsolete. Transitional glumes usually 2 and
very unequal, both usually empty, but one or the
other sometimes subtending what appears to be a
rudimentary bud which when present, is the lemma
of a sterile floret. Spikelets sessile or subsessile,
usually containing but a single perfect flower (in
some species now and then 2-as in an isotype of
Merostachys speciosa, teste Munro (1868:48)-in
others regularly 2; rarely-as in M. pluriflora-2-5
and exceptionally, as in M. polyantlza, even up to
10) and terminating in a rudimentary or more or
less depauperate sterile anthecium. Rachilla segments
potentially disarticulating immediately below
the locus of insertion of a fertile lemma.
(When the spikelet contains but a single perfect
floret, the prolongation of the rachilla is bristle-like
and bears a minute rudiment; when the spikelet
contains more than one perfect floret the prolongation
of the rachilla behind the palea of the terminal
perfect floret may be more robust than bristle-like,
and then may bear a sterile anthecium more substantial
than rudimentary.) Florets more or less
strongly inflated, the lemma fully embracing the
palea only basally at maturity, the palea gaping
antically, narrowly to broadly sulcate and &keeled
dorsally. Lodicules 3, in some species subequal and
subsimilar, elsewhere the anterior 2 asymmetrical
and paired, the posterior one smaller and sym.
metrical. Stamens 3, the filaments filiform, free.
Stigmas 2. Mature fruit a mucronate or weakly
rostrate, oblong or ovoid caryopsis with a thin,
coriaceous or crustaceous pericarp of even thickness
or more or less strongly thickened toward the base
and toward the apex; in the absence of both a sul88
SMITHSONIAN CONTRIBUTIONS TO BOTANY
cus and an embryotegium, the respective positions
of the hilum and the embryo are not manifest
externally.
ETYMoLoGY.-The name, Merostachys, formed
from the Greek, meros, part (partial, or incomplete)
and stachys, spike, alludes to the secund (onesided)
profile of the inflorescence.
TYPE-SPECIES.-Merostachys speciosa Sprengel.
RELATIoNsHIPs.-RhipidOcladum is the only bamboo
genus toward-whose members bamboos of the
genus Merostachys show any strong or striking resemblance.
This resemblance is shown most spectacularly
by the manner of development and form
of the branch complement at midculm nodes (Figures
37c, 42~). However, the form of the sheaths
(Figures 3 7 ~ , 4 2 ~ ) at midculm nodes of the young
culms in active growth (insofar as these are known
for plants of either genus) provides a convenient
means of establishing the generic affinity of a given
plant. Otherwise a deceptive technical resemblance
between members of Merostachys and members of
Rhipidocladum is manifested only where certain
features of the reproddctive structures appear in
exceptional combinations. The fact that in Merostachys
the thickening of the pericarp of the fruit
proceeds simultaneously from both apex and base
may provide an additional item of difference between
Merostachys and Rhipidocladum. Thus occasions
for confusion or uncertainty as to generic
affinity or generic discontinuity may involve those
species of Rhipidocladum having pectinately secund,
spicate inflorescences (e.g., R. maxonii) or
those species of Merostachys (e.g., M. pluriflora and
M. polyantha) having spikelets that contain several
perfect florets and terminate in a robust (not bristle-
like) rachilla segment bearing a sterile anthecium
more substantial than rudimentary. Specimens
of such mutually imitative members of Merostachys
and Rhipidocladum, however, may be sorted to
genus if, in addition to the midculm branch complement,
they both contain intact or complete examples
(in good condition) of sheaths from
midculm nodes.
I do not find any support for the opinion expressed
by Hackel (1909a:327) that Merostachys is
related to Chusquea.
DISTRIBuTIoN.-The 25 species of Merostachys
tentatively recognized here give the genus an aggregate
but discontinuous distribution extending
from Argentina to Guatemala and British Honduras,
at moderate elevations and up to 1500 meters.
As far as known today, the genus reaches its fullest
development and greatest diversity in Brazil, where
20 described species are found. Outlying stands of
single species have been reported from Argentina,
Uruguay, Paraguay, Peru, Guatemala, and Honduras;
two species have been recognized in Venezuela.
Annotated Checklist of
Species of the Genus Merostachys
1. illerostachys anomala Dutra, 1938:151, fig. 3.
[See Merostachys multiramea.]
2. Merostachys argyronema Lindman, 1900:22, pl.
15.
The Guatemalan plant referred to this species by
McClure (in Standley and Steyermark, ed., 1955:
207) is known only in the vegetative state. It probably
represents a species distinct from M. argyronema,
whose vegetative features are, as yet, insufficiently
known.
3. Merostachys bradei Pilger, 1927: 114. [See
4. Merostachys brevispica Munro, 1868:49.
5. Merostachys burchellii Munro, 1868:51.
M erosta ch ys p lurifloru.]
This species was initially described by Munro on
FIGURE 3’i.-Merostachys sp. (A,o) and Merostachys speciosa
Sprengel (B-N) . A, Seedling plant, basal part, showing
caespitose clump habit and pachymorph rhizomes (tip
of young culm shoot at left), x 1.2; B, culm sheath (apical
portion only) from midculm node of mature culm, x 0.6;
c, branch complement (base only of the component axes) at
midculm node, x ca. 0.9; D, leaf sheath (apex only) showing
ligule, base of oral setae, petiole, and base of leaf blade,
x 4.8; E, flowering branch, terminating in a secund spicate
inflorescence, x 0.6; F, single branch (a sessile spikelet)
of the inflorescence, x 2.4; G, diagram in longitudinal
section of a spikelet, showing two transitional glumes followed
by a segment of the rachilla, a bristle-like prolongation
of the rachilla (terminated by a rudiment), the lemma,
palea, and the floral axis - the last shown as a narrow
cone; H, diagram in cross-section of the floret; I, lemma in
two aspects, x ca. 3.1; J, palea in three aspects, and the
bristle-like prolongation of the rachilla in two aspects, x
ca. 3.1; K, lodicule complement, x 6; L, stamen, x 4.8; M,
gynoecium, x 6; N, branch of a stigma, greatly enlarged;
0, mature fruit (in two aspects, x 3.6) after it has produced
a seedling plant (shown at A ) . Drawings A, o based
on Swallen 8134 (US), B on Chase 9466 (US), c on Lofgren
1653 (US), D-F on SelEow 1286 (US ex B) and DN on Reitz
2268 (US).
NUMBER 9 89
90 SMITHSONIAN CONTRIBUTIONS TO BOTANY
the basis of a single collection consisting of sterile
specimens under Burchell 3243 (K). Dutra (1938:
151) described, under the same name, a flowering
specimen purporting to supplement Munro’s image
of that species without, however, citing any specimen
to document his description. But, in his introductory
statements, Dutra made reference to the
flowering, in 1907, of a plant identified by him as
Merostachys burchellii Munro. A flowering specimen
collected in that same year by Dutra under
no. 513 bears (at US) a tentative identification as
Merostachys burchellii. It may represent the same
collection as the uncited specimen on which Dutra
based the (emended) description of M. burchellii
referred to above. Judged by its vegetative features,
however, Dutra 513 is very different from
Burchell 3243, the type of Merostachys burchellii
hlunro. It is closely related to (but not identical
with) Dutra 518 (US), type-collection of Merostachys
anomala Dutra (9. v.), collected in 1906 in
the same general locality.
6. Merostachys ciliata McClure and Smith, in
7. Merostachys claussenii Munro, 1868:48.
Reitz, 1967:71, pl. ~ZD-H.
? Merostachys claussenii p mollior Doell, in Mar-
Ekman (1913:64) expressed the opinion that
Doell’s type of 6 mollior is identical with Munro’s
type of M . claussenii. Ekman supports this opinion
by the assertion that Claussen 399 “is indeed a
duplicate of the original plant,” i.e., is actually a
duplicate from the Claussen collection cited by
Munro. This assertion is neither verified nor supported
by my observation that Munro’s specimen of
“Claussen, Minas Geraes” (frag. ex K at US) is
clearl:! distinguishable from Doell’s type (Regnell
III n. 1425-frag. et photo ex S at US) on the basis
of features of the spikelets alone. The discrepancy
between Ekman’s interpretation and the way things
look to me suggests the need for a re-evaluation of
the evidence as to whether more than one recognizable
taxon is involved here.
8. Mcrostachys exserta Munro ex E. -G. Camus,
This is a species of which I have not seen a specimen.
The type (P) cited by Camus as “Auguste de
Saint-Hilaire, 1816 B 1821; Cat. D, no. 713,” is annotated
by Munro (teste Camus) with the words
tius, 1880:214.
1913, 1~74, pl. 44A.
“Spica exserta ab affinibus distat; spicula brevi M.
Fischerianae approximat.”
9. Merostachys fischeriana Ruprecht ex Doell, in
iMartius, 1880: 2 15.
10. Merostachys fistulosa Doell, in Martius, 1880:
209, pl. 55.
11. Merostachys glauca McClure and Smith, in
Reitz, 1967:74, pl. 12N.
12. Meyostachys kunthii Ruprecht, 1839:37, pl. 10:
fig. 30.
Aderostachys speciosa sensu Kunth, 1830:333, pl. 79.
[Non Sprengel, teste Ruprecht.]
13. Meyostachys maguireorum McClure, in Maguire,
Wurdack, et al., 1964:5.
14. Merostachys multiramea Hackel, 1909a:326.
Merostachys anomala Dutra, 1938: 151, fig. 3.
15. Merostachys neesii Ruprecht, 1839:37, pl. 10:
fig. 31.
Merostachys speciosa sensu Nees, in Martius, 1829:
527. “on Sprengel, teste Ruprecht.]
16. Merostachys pauciflora [as pauciflorus] Swallen,
1943:469, fig. 1.
17. Merostachys petiolata Doell, in Martius, 1880:
216.
18. Merostachys pluriflora Munro ex E. -G.
Camus, 1913, I:77.
Bambusa pubescens Doell, in Martius, 1880: 189, pl.
51: E. -G. Camus, 1913, I:124. (Non Loddiges
ex Lindley, 1835, III:357.)
Biasilocalarnus pubescens (Doell) Nakai, 1933: 10.
Merostachys hradei Pilger, 1927: 114.
A catalog of the hitherto unassembled series of
amazing and amusing taxonomic inadvertencies by
which the history of this species has become complicated
may prove instructive to the prospective
student of the bamboos. Having projected “Bam-
Dztsa” as the proper repository for a specimen from
Gaudichaud’s collection no. 97, Doell (in Martius,
1880) proceeded (certainly by inference rather than
on the basis of direct evidence) to include in the
illustration of his Bambusa pubescens a floral diagram
showing six stamens. When Doell’s otherwise
authentic plate, and his description (including the
statement “stamina 6”) caught the eye of Dr. Nakai,
the latter saw what he took to be a new genus, and
proceeded (Nakai 1933) to christen it Brasilocalamus.
Not having examined a specimen of the typeNUMBER
9 91
species, however, Nakai failed to detect Doell’s error
in giving it six stamens. At an earlier (unrecorded)
date, Munro had already annotated (at P) another
specimen from the same Gaudichaud collection
with the following words: “Merostachys sp. indescr.
M. plurifloya nov. sp. M. Kunthii Rupr. affinis differt
spiculae flosculis hermaphrod [itis] 3-5.” E. -G.
Camus (1913, I:77) published Munro’s manuscript
name, with a description consisting solely of the
Latin annotation associated with it. Camus failed
to take account of the fact that this name is based
upon a specimen from the same collection that
yielded the type of Bnmbusa pubescens Doell. Consequently,
the latter name also appears in the Camus
work (1913, I:124) along with a full
reproduction of Doell’s description of it translated
into French. Camus apparently overlooked the fact
that Bambusa pubescens Doell (in Martius, 1880)
is a later homonym of Bambusa pubescens Loddiges
in Lindley (1835) which name appears (Camus,
1913, I:152) in a list of synonyms under Dendrocalamus
strictus. In 1927 Pilger described Merostachys
bradei as new to science, on the basis of a
specimen from a different collection of M. pluriflora.
An English translation of Pilger’s notes in
German reads as follows:
The new species is distinguished by the short, thick, spikes,
as well as by the 2-3 fully developed flowers with normal
lemma and palea [in] short, thick, downy spikelets. E. -G.
Camus (1913, I:77) mentions a M[erostachys] pluriflora
Munro without a full description (‘les kpillets pluriflores’) ;
but that species, which cannot be considered as properly
published, can hardly be identical with my new species because
that one is considered to be nearly related to M.
Kunthii. The rudiments of glumes below the two empty
glumes cited by Eichler in Flora Brasiliensis were not observed
in our species.
A formal description of the following species,
apparently new to science, is included here because
by adding a member in which the number of anthecia
in its apically indeterminate spikelets sometimes
reaches the unprecedented figure of ten, it
dramatically extends the current circumscription of
a genus whose type-species has one-flowered spikelets
terminating in a bristle-like prolongation of the
rachilla bearing a rudiment.
19. Merostachys polyantha McClure, new species
Culmi 6-8 m alti (teste A. Lima); internodia
teretia, cava, parietibus tenuibus, extus infra unumquidque
nodum in zona angusta pilis albidis adpressis
confertim vestita, alibi papillis retrorse
aculeiferis praedita, demum luteo-viridia et punctis
vinosis minutis elongatisque omnino variegata.
Culmorum vaginae deciduae, tum ex omnibus cingulo
pilis albidis retrorse adpressis vestito remanenti;
auriculis et setis oralibus valde evolutis;
ligula perbrevis, margine irregulatim denticulata
ciliolataque; lamina decidua (non adhuc visa).
Foliorum vagina striato-sulcata, secus marginem
externam setis lustratis ornata, vaginis infimis dorso
et ut videtur omnino glabris, supremis papillis retrorse
uncinatis punctatis; auriculis et setis oralibus
circa ut in vaginis culmorum evolutis; ligula interiore
haud exserta, exteriore anguste lineari, margine
vel subglabra vel dense ciliolata; petiolo usque
8 mm longo, compresso, vel omnino glabro vel
utrinque dense scabro; lamina usque 150 x 35 mm,
oblongo-lanceolata, basi asymmetrice vel rotundata
vel cordata, apice asymmetrice et abrupte acuminata;
laminis infimis et eis supremis in quoque
axi florifero ovatis acuminatissimisque, omnibus
utrinque concoloratis et fere glabris; nervis longituclinalibus
omnibus in laminarum facie abaxiali
tantum salientibus; costa tantum basin versus laminae
distinguenda; nervibus secundariis utroque 5-8;
tertiariis in quoque spatio costae contiguo usque 8,
versus marginem laminae gradatim paucioribus ita
in quoque spatio marginali saepius ad 2 redactis;
venulis transversis in facie abaxiali tantum distinguendis,
vicinis 1.5-4.0 mm sese distantibus.
Inflorescentiae e racemis spicatis secundisque sistentes,
solitariae, axes saepius laminiferos terminantes;
pedunculo vel haud vel parum exserto,
apice dense canescenti; rhachidi 25-65 mm longa,
curvula, dense antrorse-pubescenti; bractea spiculam
basilarem subtendenti ovata, apice acuta apiculataque,
multinervia et valde carinata, dorso
marginibusque omnino glabra, usque 10 mm longa,
succedentibus gradatim redactis, distalibus vel rudimentariis
vel obsoletis. Spiculae compressulae,
anguste triangulae, apice indeterminatae, insertione
solitariae, confertae, alternae, distichaeque,
pleraeque 1.5-2.0 mm (infimae usque 4.0 mm) sese
distantes; spiculis in quoque racemo usque 26, infima
55 mm longa, sequentibus gradatim brevioribus,
tum distalibus 7-8 vel depauperatis vel
rudimentariis; pedicel10 sub-nil, latere externo prominenter
pulvinato; glumis transitionalibus saepius
92 SMITHSONIAN CONTRIBUTIONS TO BOTANY
2, alia vel alia aliquando rudimentum (velut gemmam
dormientem) foventi; I ca 4 mm longa, anguste
triangula, acuminatissima, valde carinata,
cetera enervosa, dorso invalide appresso-hirtula,
seas margines valde ciliata; II 6-9 mm longa, ovata,
acuminata, 5-nervia et valde carinata, extus valde
hirtula, marginibus versus apicem tantum incalide
ciliolata. Flosculi in spiculis proximalibus usque
10, eis in spiculis succedentibus gradatim paucioribus;
flosculo terminali vel plus minusve depauperato
vel etiam in statu rudimentario dirempto.
Anthoeciorum elementae utraeque crustaceae, extus
plus minusve dense strigosae, sews margines apicem
versus antrorse-ciliatae. Rhachillae segmenta
usque 4mm longa, semiclaviformia, omnino molliter
antrorse pubescentia. Lemma flosculi infimi
usque 20 mm longum, flosculorum sequentium gradatim
redactum, anguste lanceolatum, apicem verus
5-nervium naviculareque, alibi enervium. Palea
vulgo vel lemmam aequans vel parum brevior, raro
parum longior; oblongo-lanceolata, late sulcata,
omnino enervia, apice et profunde et anguste bilobata.
Lodiculae dimensionibus formaque variabilissimae,
textura tenuissimae, deorsum interdum
opacae, sursum semper translucidae, vel enerviae
vel invalide paucinerviae, omnino glabrae vel margine
(apice saltem) obscure ciliolatae. Antherae
atropurpureae, versus basin apicemque parum
attenuatae. Ovarium, stylus et rami styli infra stigmata
omnino glabri. Fructus non adhuc inventus.
Culms 6-8 m tall (teste A. Lima) the internodes
terete, hollow, thin-walled, more or less heavily
coated at first with white powder and densely
clothed in a narrow zone immediately below each
node with retrorsely appressed lustrous white hairs,
elsewhere sparsely strewn with papillae each tipped
with a minute retrorse hook; ultimately seen as
yellowing green in color, variegated with longitudinally
elongated vinaceous stipples. Culm
sheaths deciduous, upon abscission leaving a rather
prominent persistent girdle densely clothed outside
and fringed basally with retrorsely appressed lustrous
white hairs; the sheath proper ciliate on the
outer margin, apparently entirely glabrous abaxially;
auricles and oral setae prominently developed,
the auricles narrow, tortuous, dull, and glabrous
abaxially, densely hirtellous adaxially, the oral
setae 10-12 mm long, spreading from crowded loci
of insertion on the margin and on the adaxial surface
of the auricles, lustrous, very sparsely and
obscurely antrorse-scabrous throughout, crisped
distally; ligule short, truncate or slightly arcuate
apically, minutely and irregularly denticulate and
ciliolate on the margin; blade deciduous (not seen).
Leaf sheaths proper (represented on flowering twigs
only in the available specimens) minutely corrugated
(quasi-plicate) longitudinally, densely fringed
with lustrous hairs on and near the outer margin;
the proximal sheaths in a given series glabrous, the
distal ones strewn (on the exposed surface at least)
with retrorsely uncinate papillae; auricles and oral
setae approximately as seen on the culm sheaths;
inner ligule not exserted, the outer ligule narrowly
linear, subglabrous to densely ciliolate on the margin;
petiole up to 8 mm long, dorsiventrally
compressed, entirely glabrous to densely antrorsehispidulous
throughout on both surfaces; blade up
to 150 x 35 mm, oblong-lanceolate, asymmetrical
and broadly rounded or cordate at the base, asymmetrically
and abruptly acuminate at the apex,
abruptly becoming ovate-acuminate and greatly reduced
in size at the proximal and distal extremes
of every leafy flowering twig, concolorous and glabrous
or nearly so on both surfaces; longitudinal
nerves all salient on the abaxial surface only, the
midrib distinguishable only toward the base of the
blade; secondary nerves 5-8 on each side of the
midrib; tertiary nerves up to 8 in the first space on
each side of the midrib, the number in each successive
internerve space reduced progressively,
commonly 2 next to each margin of the blade;
transverse veinlets visible only on the abaxial surface,
adjacent ones in a given space 1.5 to 4.0 mm
distant from each other.
Inflorescences congested secund spicate racemes
terminal to usually leafy axes; peduncle scarcely to
slightly exserted, densely canescent apically; rachis
25-65 mm long, lightly curved, densely strigose;
bract subtending the lowermost spikelet ovate,
acute and apiculate apically, dorsally many-nerved
and prominently keeled, glabrous throughout on
the back and on the margins, up to 10 mm long,
those subtending successive spikelets progressively
reduced, the distal ones either rudimentary or obsolete.
Spikelets somewhat compressed, narrowly
triangular, of indeterminate growth apically, of
solitary but crowded, alternate and distichous insertion
at intervals of 1.5 to 4.0 mm from the base
NUMBER 9 93
of the raceme upward in each row; spikelets in
each raceme up to 26, the basal one up to 55 mm
long, the successive ones progressively reduced in
length to about 20 mm, followed by sometimes
seven or eight abruptly and progressively depauperate
to rudimentary ones; pedicels almost obsolete,
each prominently pulvinate on the external
side only; transitional glumes normally two, one or
the other sometimes subtending a bud-like rudiment;
I about 4 mm long, narrowly triangular,
acuminate, prominently keeled, otherwise without
visible nerves, weakly appressed hirtulous dorsally
and strongly ciliate on the margins; 11 6-9 mm long,
ovate, acuminate, &nerved and strongly keeled,
hirtulous dorsally and weakly ciliolate on the margin
toward the apex only. Florets in the lowermost
spikelets up to 10, progressively fewer and shorter
in succeeding spikelets, the terminal floret either
more or less strongly depauperate or stalled at a
rudimentary stage in its development. Elements of
the anthecia both crustaceous, more or less densely
strigose on the outer surface, antrorse-ciliolate on
the margins apically. Rachilla segments up to 4
mm long, semiclaviform, softly pubescent throughout
with antrorse hairs. Lemma in the lowermost
floret up to 20 mm long, progressively reduced in
the succeeding florets, narrowly lanceolate, 5-
nerved and strongly boat-shaped toward the apex,
elsewhere without manifest nerves. Palea commonly
as long as the lemma or a little shorter,
rarely a little longer; oblong-lanceolate, broadly
sulcate, lacking manifest nerves, deeply and narrowly
bilobed at the apex. Lodicules of variable
form and dimensions, of delicate texture, sometimes
opaque at the base, always translucent above, either
weakly few-nerved or nerveless, glabrous throughout
or obscurely ciliolate on the margin toward
the apex only. Anthers dark purple, lightly attenuate
toward the base and at the apex. Ovary,
style and branches of the style (below the stigmas
at least) entirely glabrous. Fruit not yet found.
COTYPES (= syntypes, sensu Frizzell 1933:647).-
Herbarium Bradeanum no. 22857, leg. Flavia Torgo
(s.n.) 19 viii 1926; and Herb. Kew, leg. A. Lima
(= Dartlano de Andrade-Lima) no. 62-4122, 24
viii 1962 (ex Instituto de Pesquisas Agronomicas
Secao de Botanica, Sao Paulo, no. 13192), both
collections taken simultaneously from the same
flowering plant (teste A. Lima): “Serra de Mongagui,
pr. Praia Grande, Municipio Mongagua, Sao
Paulo, Brasil, a sombra da mata no alto da serra
(100-120 m). 6-8 m. F1. roxoescuro. Caule c/peq.
aculeos cuticulares (?) volt. p/base.” Plants shaded
by forest at the summit of a hill 100-120 m above
sea level-culms 6-8 m tall, with minute cuticular
hooks on the lower internodes; flowers dark red.
[Vernacular name:] ‘taquara miju.’ ” Notes ex A.
Lima. Both cited specimens seen. Photographs of
both cotype specimens, and a duplicate of the specimen
collected by A. Lima are on deposit at the
U.S. National Herbarium.
RELATIONSHIPS.-Merostachys polyantha differs
from the closely related M. pluriflora in the more
robust development of the inflorescences, spikelets,
anthecia and the auricles, and oral setae; as well as
in the greater maximum number (up to 26) of
spikelets in each raceme; the greater maximum
number (up to 10) of florets in each spikelet; the
shape and vesture of the parts of the anthecia; and
the much broader, oblong-lanceolate shape of the
midrange leaf blades.
DIsTR1suTIoN.-Merostuchys polyantha is known
as yet by only two collections taken, simultaneously
(teste A. Lima), from a single source: Serra de Mongag&,
prope Praia Grande, Municipio SPo Paulo,
Brasil. Its nearest known relative, Merostachys
PlzirifZora, has been collected near Iguape, State of
S5o Paulo, and on the Island of Santa Catarina,
State of Santa Catarina.
20. Merostachys retrorsa McClure, in Maguire,
Wurdack, et al., 1964:6.
21. Merostachys riedeliana Ruprecht ex Doell, in
Martius, 1880: 2 13.
22. Merostachys sellovii Munro, 1868:51.
23. Merostachys sparsiflora Ruprecht, 1839:37, pl.
10: fig. 32.
24. Merostachys speciosa Sprengel, 1825:249 (Fig
ure 3 7 ~ - ~ ) .
Under this name Munro (1868:48) writes ‘‘I find
in one of Sellow’s specimens the second hermaphrodite
flower which I had first observed in M. Clausseni.”
25. Merostachys speciosa sensu Kunth, 1830: 333,
334, pl. 79. [See Merostachys kunthii.]
26. Merostachys speciosa sensu Nees, in Martius,
1829:527-531. [See Merostachys neesii.]
94 SMITHSONIAN CONTRIBUTIONS TO BOTANY
27. Merostachys ternata Nees, in Martius, 1829:
28. Merostachys vestita McClure and Smith, in
The following authors offer additional information
on some of the species of Merostachys listed
above: Doell, in Martius, 1880:207-218; McClure,
in Swallen, 1955:207-209; McClure and Smith, in
Reitz, ed., 1967:63-75; Metcalfe, 1960:564, 586-588;
Munro, 1868:47-52; Nees, in Martius, 1829:527-
531; Parodi, 1936:242-244; Ruprecht, 1839:36-38.
529.
Reitz, 1967:72, pl. 12 i-j.
Annotated Checklist of Excluded Species
Erroneously Allocated to the Genus Merostachys
1. Merostachys capitata Hooker, 1840:273. [See
A throostachys capitata.]
Merostachys capitata latifolia Doell, in Martius,
1880:2 17.
Merostachys capitata @ angustifolia Doell, in Martius,
1880:217.
2. Merostachys race.mipora (Steudel) Fournier,
1881: 131. [See Rhipidocladum racemiflorum.]
3. Merostachys sikokianus (Makino) Nakai, in
Hara, 1935:74. [A Japanese plant of the genus
Sedum-Crassulaceae.]
Myriocladiis Swallen
FIGURES 38-40
Myriocladus Swallen, in Steyermark, et al., 1951:34, fig. 4;
in Maguire, Wurdack, et. al., 1957:237-239.-McClure,
195$:205.
Plants, unicespitose, unarmed. Rhizomes pachymorph.
Culms self-supporting, the internodes terete,
in some species hollow, in others efistular, the
lowermost above-ground one usually disproportionately
elongated and followed by one to several more
or less completely suppressed (obsolete) ones with
the intervening nodes crowded closely together.
Branch buds at culm nodes solitary. Branch complements
restricted monoclade, the successive ones
sometimes brought deceptively closely together by
the suppression of culm internodes. The primary
element of branch complements apparently never
proliferating from any bud at its proximal nodes,
this perhaps due to suppression of the buds by
powerful physical pressure exerted by the very
thick, rigid sheaths that subtend them. [A proximal
internode of the single component branch is also
(as in the culm) commonly elongate and followed
by one to several suppressed (obsolete) ones with
the intervening nodes crowded closely together,
making the corresponding sheaths strongly imbricate.]
Leaves (blades of leaf sheaths) petiolate or
subsessile, the transverse veinlets usually obscure or
invisible externally or (as in M. paludicolus) more
or less conspiculously manifest on both surfaces.
Inflorescences semelauctant, terminating leafy or
leafless branches and (in some species) the culms as
well; generally long and narrow, of diverse branching
habit, racemose to paniculate, usually with part
or all of the branches of secund orientation and
lacking both prophylla and subtending bracts.
Spikelets either pedicellate, subpedicellate or sessile,
containing 2 (1-5) perfect florets, and terminating
in a depauperate sterile anthecium. Rachilla
segments short, potentially disarticulating immediately
below the locus of insertion of the first only,
or (in some species) each of the fertile lemmas.
Transitional glumes typically 3: I and 11 “empty
glumes,” 111 a “sterile lemma” that is either empty
or subtends a rudimentary flower. Fertile lemma
fully embracing its palea only basally at maturity,
the palea broadly sulcate dorsally, the margins gaping
basally and lightly overlapped above. Lodicules
3, either subsimilar in shape or the anterior 2 more
or less noticeably asymmetrical and paired, the
posterior one symmetrical and slightly smaller. Stamens
3, the filaments filiform, free. Style very short,
terminating in two stigmas. Mature fruit a mucronulate
or weakly rostrate, fusiform caryopsis with a
thin membranaceous pericarp of even thickness, the
sulcus and the basal position of the embryo clearly
manifest.
ETYMoLoGY.-The name Myriocladus, formed
from the Greek, myrios, numberless, and cludos,
branch, ostensibly alludes to the profuse ramification
of the inflorescences.
TYPE-sPF.cIEs.-Myriocladus virgatus Swallen.
RELATIONSHIPS.-Certain of the striking generic
features of Myriocladus are shared by individual
FICURF 38.-MyriocZadus confertus Swallen. Aa, Ab, Leafy
flowering plant, x 0.6; B, seedling, x 1.2; c, first-order
branch of inflorescence, x 1.2; D, spikelet, x ca. 9.6; E,
apex of leaf sheath and base of blade, x 1.8; F, two nodes of
culm with shortened internode and an intact branch bud,
x ca. 6. All drawings based on Steyermark and Wurdack
1220 (US).
u3 w
96 SMITHSONIAN CONTRIBUTIONS TO BOTANY
NUMBER 9 97
FIGURE 39.-Myriocladus virgatus Swallen (Aa-AC) and hlyriocladus
confertzts Swallen (B). Aa, Leafy tip of sterile culm,
x 0.6; Ab, apex of leaf sheath and base of blade (abaxial
aspect), x 1.2; AC, apex of leaf sheath and base of blade
(adaxial aspect), x 1.2; B, leafy tip of sterile CUh, X 0.6;
Drawing Aa based on Cowan and Wurdack 31376 (US), Ab,
AC on Steyermark 58293 (F), and B on Steyermark and
IVurdack 1220 (US).
members of other genera. A disproportionately
elongate proximal above-ground culm internode
followed by two or more obsolete internodes is a
feature shared by Glaziophyton mirabile, by Aulonemia
queko, and by a number of species of
Arthrostylidzum, including A. schomburgkii. The
strongly thickened and indurate culm sheaths,
branch sheaths and leaf sheaths with leathery
blades, and the (sometimes irregular) incidence of
solitary and appressed branches at culm nodesfeatures
characteristic of all known species of
Myiiocladus-are found also in Aulonemia de-'
pexa and A. eflusa, and in all known species of
Greslania. Probably because some species of both
genera have sessile leaf blades Myriocladus has been
said to resemble (be related to?) Neurolepis (Swallen
in Steyermark, et al., 1951:34). However, in
its over-all combination of vegetative and reproductive
features, Myriocladus is clearly disjunct from
all other bamboo genera.
DISTRIBUTIoN.-The twenty known species of Myriocladus
are known only from the sandstone tablelands
region of Venezuela. They occupy generally
mesophytic, rarely swampy habitats, at recorded altitudes
ranging from 1025 m to 2500 m.
Checklist of New World Bamboos
Described Under the Genus Myriocladus
1. Myriocladus afinis Swallen, in Maguire, Wurdack,
et al., 1957:244, fig. 3f.
2. Myriocladus cardonae Swallen, in Steyermark,
et al., 1951:35.
3. Myriocladus confertus Swallen, in Maguire,
Wurdack, et al., 1957:248, fig. 4c; in Maguire,
Steyermark, Wurdack, et al., 1957:397. (Figures
38, 39~).
4. Myriocladus churunensis Swallen, in Steyermark,
1967: 132.
5. Myriocladus distantiflorus Swallen, in Maguire,
Wurdack, et al., 1957:248, fig. 4d.
6. Myriocladus exsertus Swallen, in Maguire,
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Wurdack, et al., 1957:242, fig. 3c.
Myriocladus gracilis Swallen, in Maguire, Steyermark,
Wurdack, et al., 1957:393, fig. 74.
Myriocladus grandifolius Swallen, in Maguire,
Wurdack, et al., 1957:245, fig. 3h.
Myriocladus longiramosus Swallen, in Maguire,
Wurdack, et al., 1957:243, fig. 3e.
Myriocladus maguirei Swallen, in Maguire,
Wurdack, et al., 1957:239, fig. 3b.
Myriocladus neblinaensis Swallen, in Maguire,
Wurdack, et al., 1957:240, fig. 1.
Myriocladus paludicolus Swallen, in Maguire,
Wurdack, et al., 1957:246, 248, fig. 4b.
Myriocladus paraquensis Swallen, in Maguire,
Wurdack, et al., 1957:246, figs. 3i, j.
Myriocladus paruensis Swallen, in Maguire,
Wurdack, et al., 1957:244, fig. 3g.
Myriocladus purpureus Swallen, in Maguire,
Wurdack, et al., 1957:248, fig. 4e; in Maguire,
Steyermark, Wurdack, et al., 1957:397.
Myriocladus simplex Swallen, in Maguire,
Wurdack, et al., 1957:242, fig. 2.
Myriocladus steyermarkii Swallen, in Maguire,
Wurdack, et al., 1957:247, fig. 3a; 395.
Myriocladus uariabilis Swallen, in Maguire,
Wurdack, et al., 1957:248, fig. 4a; in Maguire,
Steyermark, Wurdack, et al., 1957:396.
Myriocladus virgatus Swallen, in Steyermark,
et al., 1951:34, 35, fig. 4. (Figures 3 9 ~ a - ~ c , 40)
Myriocladus wurdackii Swallen, in Maguire,
Wurdack, et al., 1957:248, fig. 4f; in Maguire,
Steyermark, Wurdack, et al., 1957:398.
Neurolepis Meisner
FIGURE 41
Neurolepis Meisner, 1843, I:426, II:325.-Pilger, in Engler
Platonia Kunth, 1829: 139 [non Rafinesque, 1808, nec Martius,
Planotia Munro, 1868:70.-Bentham, in Bentham and Hooker,
and Prantl, 1906:21.-McClure, 1957:206.
18291,- Nees, 1834:486.
1883: 1209.-Hackel, in Engler and Prantl, 1887:93.
Plants typically unicespitose, exceptionally-as in
N. aristata (Munro) Hitchcock-forming dense,
more or less extensive thickets; unarmed. Rhizomes
pachymorph, relatively slender. Culms self-supporting,
typically unbranched in the vegetative state,
the internodes fistular and cylindrical or terete.
Leaf blades typically coarse and grass-like, petiolate
or sessile, persistent or articulated with the leaf
98 SMITHSONIAN CONTRIBUTIONS TO BOTANY
NUMBER 9 99
FIGURE 40.--Myriocladus virgatus Swallen. A, Tip of leafy
flowering culm, x 0.6; 6, spikelet, x 12; c, transitional
glumes, x 12; D, floret I, showing dorsal aspect of palea,
x 12; E, floret 11, showing terminal depauperate floret lying
in the sulcus of t&e palea, x 12; F, lemma I, dorsal aspect,
x 12; c, palea, ventral aspect, x 12; H, lodicule complement,
x 12; I, stamen, x 12; J, gynoecium, x 12. Drawing A
based on Cowan and Wurdack 31376 (US), and B-J on
Steyermark 58293 (F) .
sheath, the transverse veinlets not at all visible to
clearly manifest, externally.
Inflorescences semelauctant, each terminal to an
unbranched culm, broadly to narrowly paniculate
with a strong excurrent rachis, the branches appressed
or spreading, commonly only one order of
the branches (the first or the last or, in some species,
none) showing secund orientation; prophylla lacking;
subtending bracts commonly lacking or
obsolete. Transitional glumes at the base of the
spikelet 4 (-5, as in N. angusta Swallen): I and 11
“empty glumes,” 111 and IV (-v) “sterile (empty)
lemmas.” Spikelets pedicellate, typically containing
but one perfect floret (the terminal one), the
rachilla very short, articulated immediately below
the locus of insertion of either the first sterile
lemma only, or both the fertile and the sterile
lemmas, and not prolonged behind the palea of the
terminal perfect floret. Fertile lemma sometimes
embracing its palea fully, but as a rule only basally
so at maturity. Palea gaping antically, dorsally
convex, in some species with an entire apex and not
at all sulcate, elsewhere showing a very short,
narrow sulcus terminating in an emarginate or
bimucronulate apex. Lodicules 3, subequal or
the anterior two appreciably asymmetrical and
paired, the posterior one symmetrical and smaller
(at least either shorter or narrower). Stamens
typically 3-exceptionally -4, -5, -6 as in Cuatrecams
14813 (US) (AT. aristata (Munro) Hitchcockthe
filaments filiform, free. Stigmas 2. Mature
fruit an oblong, straight caryopsis with a uniformly
thin, glabrous, dry pericarp, the sulcus abbreviated
distally and proximally, the basal position of the
embryo clearly to weakly manifest. [Examples of
mature fruits known to me only from a specimen
(US) identified by Soderstrom as Neurolepis elata
(Kunth) Pilger, collected in Colombia by Cuatrecasas
(no. 11522). These are extremely minute
2.0-2.5 mm long) but well formed.]
ETmfoLom.-The name Neurolepis, coined from
the Greek words, neuron, nerve, and lepis, scale,
apparently alludes to the strong, keel-like median
nerve of the empty glumes in the type-species of this
initially monotypic genus.
(Kun th) Pilger
(in Engler and Prantl, 1906:Zl); Platonia elata
Kunth (1830:327, pl. 76).
RELATIONSHIPS.-Plants comprised by Neurolepis
share the following features with plants comprised
by the genus Swallenochloa hollow clum internodes,
a similar spikelet structure, and inflorescenses
with branches in secund orientation. Plants
of the genus h’eurolepis differ from plants of the
genus Swallenochloa most conspicuously in their
coarse, grass-like appearance and generally restricted
radial development, besides bearing very much
larger inflorescences that are confined to the terminal
position on culms devoid of lateral branches.
Ostensibly allied to Chusquea by a similar spikelet
structure and (in some species) by certain
features of the inflorescence, members of the genus
A’eurolepis differ from members of the genus
Chusquea most conspicuously in bearing an inflorescence
only at the tip of an unbranched culm,
and in the incorporation of the following vegetative
features: the generally coarse, grass-like (not bamboo-
like) appearance of the plant (due to the shape
and dimensions of the leaf blades, plus the lack of
a strong differentiation between culm sheaths and
leaf sheaths); the relatively slender form of the
pachymorph rhizomes; and culms with hollow internodes
but no lateral branches.
When Kunth (1829: 139) published the genus
Platonia, he stated that it differs from Chusquea
only in a peculiar habit. However, on page 327 of
the same work (1830), he said of Platonia elata
Kunth-basionym of Neurolepis elata (Kunth)
Pilger-“Cette plante prksente tous les caracteres
du genre Chusquea; mais elle en diffkre tellement
par son port, que je n’ai pas hCsitC B en former
un genre particulier.” Nees (1834:486) reduced
hTeurolepis (as Platonia Kunth) to the rank of
subgenus under Chusquea. Ruprecht (1839:30),
who did not include Neurolepis (Platonia elata
Kunth) in his monograph, expressed the opinion
that Platonia in its true nature is very far from
Clausquea: “Platonia vero natura sua valde a Chusquea
distat.“
T Y PE-SPEC~E~. -New0 lep is e la ta

NUMBER 9 101
FIGURE 41.--NeuroZepis aristata (Munro) Hitchcock (A) and
Neurolepis pittieri McClure (B-N). A, Flowering plant, x 0.6;
B, branch of an inflorescence, x 2.4; c, spikelet, x 4.8; D,
E, transitional glunies, x 4.8; F and G, sterile lemmas, x 4.8;
H, fertile lemma, x 4.8; I, palea, x 4.8; J, lodicule complement,
x 4.8; K , stamen, x 4.8; L, gynoecium, x 4.8; M, apex
of leaf sheath, x 1.2; N, diagrammatic cross-section of floret.
Drawing A based on Cuatrecasas 14813 (US) and B-N on
Pittier I0067 (US).
Freier’s study (1945: 103) of Neurolepis pittieri
and N. aristata reveals a bambusoid type of leaf
anatomy in both species. It seems to me, however,
not unlikely that intensive comparative studies of
the known members of the genus Neurolepis, undertaken
on a broader interdisciplinary basis, may
result in their being given a revised disposition
within the subfamily Bambusoideae, possibly as a
tribe distinct from the “true bamboos.”
The genus was treated briefly by Soderstrom (in
Maguire, et al., 1969) and a key given to the nine
species which he recognized. Gould and Soderstrom
(1970) reported a chromosome count of 2n = 48
for a specimen of N. aperta from Colombia.
DIsTRIBUTIoN.-Species of Neurolepis have been
reported from an area that embraces sites in Peru,
Ecuador, Colombia, Venezuela, and the island of
Trinidad. All are reported from relatively high
elevations (2900-4500 m); some of them inhabit
ecological formations characterized as “pAramo.”
According to Munro (1868:73) Chusquea aperta
Munro [= Neurolepis aperta (Munro) Pilger] was
collected in Colombia by Goudot, who recorded
for it the local name “Chusquea de PAramo.”
Notes by Jameson, collector of the type-specimen
of Chusquea aristata Munro [= Neurolepis aristata
(Munro) Hitchcock] are quoted by Munro (1868:
61) as follows: “It is a tall reedy grass, found only
on the Eastern chain of the Andes. At 13,000 feet
[ca. 4000 m] it makes its appearance, in irregular
patches; at 15,000 feet [5000 m] it completely covers
the whole surface, forming what the natives call
a ‘carizal,’ impenetrable to man or beast. It continues
upwards nearly to the limits of perpetual
snow.”
Annotated Checklist of Recognized Species of
the Genus Neurolepis
1. Neurolepis angusta Swallen, in Maguire, Wurdack,
et al., 1957:249.
Neurolepis densiflora Swallen, in Maguire, Steyermark,
Wurdack, et al., 1957:399.
2. Neurolepis aperta (Munro) Pilger, in Engler
and Prantl, 1906:21.
Planotia aperta Munro, 1868:73.
Planotia ingens Pilger, 1898:721.
Neurolepis ingens (Pilger) Pilger, in Engler and
Prantl, 1906:21.
3. Neurolepis aristata (Munro) Hitchcock, 1927b:
313 (Figure 4 1 ~ ) .
Chusquea aristata Munro, 1868:61.
Planotia acuminatissima Munro, 1868:72.
Planotia stiibelii Pilger, 1898:720.
Planotia tesselata Pilger, 1898:720.
Neurolepis acuminatissima (Munro) Pilger, in
Neurolepis stiibelii (Pilger) Pilger, in Engler and
Neurolepis tesselata (Pilger) Pilger, in Engler and
Neurolepis weberbaueri Pilger, 1921:446.
Engler and Prantl, 1906:21.
Prantl, 1906: 21.
Prantl, 1906:Zl.
4. Neurolepis diuersiglumis Soderstrom, in Maguire,
et al., 1969:16-18, 20.
5. Neurolepis elata (Kunth) Pilger, in Engler
and Prantl, 1906:21.
Platonia elata Kunth, 1830:327, pl. 76.
Planotia elata (Kunth) Munro, 1868:71.
Planotia nobilis Munro, 1868:72.
Nezi?‘olepis nobilis (Munro) Pilger, in Engler and
6. Neurolepis glomerata Swallen, in Maguire,
Prantl, 1906:21.
Steyermark, Wurdack, et al., 1957:399.
7. Neurolepis mollis Swallen, 1931: 14.
8. Neurolepis pittieri McClure, 1942:181, fig. 8
(Figure 41~-N).
9. Neurolepis virgata (Grisebach) Pilger, in
Engler and Prantl, 1906:Zl.
Platonia virgata Grisebach, 1864:530.
Planotia uirgata (Grisebach) Munro, 1868:71.
Neurolepis nigra Swallen, 1957b: 400.
Rhipidocladum McClure, new genus
FIGURE 42
Plantae unicaespitosae, inermes. Rhizomata
pachymorpha. Culmi infra saltem medium sese
sustinentes erecti, sursum vulgo vel scandentes vel
102 SMITHSONIAN CONTRIBUTIONS TO BOTANY
apice penduli; internodiis cylindratis, intus fistulosis.
Ramorum complementum ad nodos medianos
culmorum insertum; eiusdem axi primario applanato,
subtriangulato, ad culmi superficiem appresso
adnatoque, neque cylindrato neque segmentato;
axibus secundariis numerosis, gracilibus, subaequalibusque,
margine axis primarii insertis, ita ordinatione
rhipidiforme sese ostendentibus. Foliorum
lamina venulis transversis in speciebus plerisque
extus vulgo haud manifestis, alibi plus minusve
Clare visibilibus.
Inflorescentiae semelauctantes, ramificatione
typice racemosa, rachidi vel percurrenti vel deliquescenti;
spiculae insertione et orientatione diversae;
prophyllis bracteisque nullis. Glumae
transitionales (glumae vecuae et lemmata sterilia)
numero diversae atque nonnumquam inconstantes.
Spiculae vel sessiles vel subsessiles vel pedicellatae,
flosculos hermaphroditos vel paucos vel plures continentes,
apice in anthecio depauperato sterilique
terminantes. Lemma fertile maturitate saltem paleam
suam basi tantum circumplectens. Palea dorso
2-carinata et late sulcata, antice hians. Lodiculae
(nonnumquam -2, -1, -0) typice 3, duabus anterioribus
plus minusve asymmetricis atque geminatis,
posteriore symmetrica sat minoreque. Stamina 3,
filiamentis filiformibus liberisque. Stigmata 2.
Fructus maturus caryopsidens; caryopsis oblonga
sulcataque, leviter compressa, apice cuspidata;
pericarpio vel pergamineo vel crustaceo, glabro,
vel opaco vel nitido, crassitudine uniformi; sulco
et embryotegio aut valde aut debiliter manifestis.
Plants unicespitose, unarmed. Rhizomes pachymorph.
Culms self-supporting below, commonly
either scandent or pendulous above, the internodes
cylindrical, hollow. Primary branch buds (when
present) at midculm nodes, solitary, each containing
a single initial primordium. Branch complement at
midculm nodes monoclade, of restricted insertion,
the primary element neither cylindrical nor segmented,
but thin, roughly triangular, appressed and
more or less strongly adnate to the surface of the
culm, the secondary axes numerous, slender, subequal,
arising from primordia formed on the margin
of the flat, primary element, of close-set apsidate
insertion, with a fan-like pattern of orientation.
Leaves (blades of leaf sheaths) with transverse
veinlets in most species not at all (or scarcely)
manifest, but in an occasional one more or less
clearly visible externally.
Inflorescences semelauctant, typically of racemose
branching, the rachis either deliquescent, or percurrent
and straight or zigzag, the spikelets inserted
and oriented in diverse patterns. Excepting the
lowermost one, bracts subtending primary branches
usually obsolete or lacking; typical prophylls lacking.
Transitional glumes (empty glumes and sterile
lemmas) 2 or 3 (rarely 4) the uppermost one often
subtending a depauperate flower, the others empty.
Spikelets comprising few to several perfect florets,
and terminating in one or more progressively depauperate
sterile anthecia. Rachilla segments possessing
at maturity the potential for disarticulating
just below the locus of insertion of each fertile
lemma. Lemma (when subtending a functional
flower) fully embracing the palea only basally at
maturity. Palea broadly sulcate and 2-keeled dorsally,
the margins not at all or only slightly and
partially overlapping. Lodicules typically 3, the
anterior two asymmetrical and paired, the posterior
one smaller (at least shorter or narrower) and
symmetrical (all three symmetrical and subequal in
R. parviflorum; exceptionally -2, -1, -0 as recorded
for R. ampliflorum). Stamens 3, the filaments filiform,
free. Stigmas 2. Mature fruit (as far as
known; available examples are rare) an oblong,
subfusiform, sulcate, apically mucronate caryopsis
lightly compressed and lightly curved or asymmetri-
FICURI: 42.-Rhipidoclrrdurn harmonicurn (Parodi) McClure
(A-Q) and Rhipidoclnduna vet?ici/laturn (Nees) McClure
(R-T) , A, Rhizome and base of its culm, x 0.6; B, culm internode
between nodes v and 1’1, showing first-order element of
the branch complement with second-order elements developing
from buds on its margin, x 0.6; c, first-order element of
branch complement from 8, removed and shown from the
adaxial aspect, x 1.8; D, fully developed midculm branch
complement, basal portion only, in situ, x 0.48; E, midculm
sheath, adaxial aspect, x 0.6; F, sterile leafy twig, x 0.6;
c, leafless flowering twig: x 0.6: H, apex of leaf sheath and
base of blade, x ca. 3; I, spikelet, sessile with portion of
rachis attached, x ca. 1.8; J, transitional glumes from I,
x ca. 6; K, floret, x ca. 3.6; L, lemma, x ca. 3.6; M, palea,
x ca. 3.6; N, lodicule complement, x ca. 7.2; 0, stamen,
x ca. 6; P, gynoecidm, x ca. 14.4; Q, stigmatic processes,
greatly enlarged; R, fruit, embryo side, x 7.2; s, fruit, hilum
side, x 7.2; T, fruit, longitudinal section, x 7.2. Drawings
A, B based on McClure 21398 (US), c, E, F, H on McClure
21416 (US), D on Camp E-1613 (US), G, I-M on Killup and
Smith 25622, N-Q on Sodiro s. n., Oct. 1900 (US), and
R-T on Glaziou 22424 (P) ,
NUMBER 9 10s
104 SMITHSONIAN CONTRIBUTIONS TO BOTANY
cal in the dorsiventral plane; pericarp pergamineous
(crustaceous in R. verticillatum), glabrous, of
uniform thickness; the sulcus and embryotegium
weakly to strongly manifest.
ETYMoLoGY.-The name Ripidocladum, derived
from the Greek, rhipis, fan, and klados, branch,
alludes to the fan-like pattern formed by members
of a mature midculm branch complement in all of
the bamboos herein allocated to the genus.
TYPE-sPEcIEs.-RhiPidocladum harmonicum (Parodi)
McClure, new combination. Arthrostylidium
harmonicum Parodi (1944:478, fig. 1).
RELATIONSHIPS.-The combinations of morphological
features used herein to define Rhipidocladum
places this genus between Merostachys and Arthrostylidium,
with some of its features indicating
affinities toward Merostachys and other features
indicating affinities toward Arthrostylidium. The
taxa I have allocated to Rhipidocladum are perfectly
uniform in terms of the ontogeny and morphology
of the midculm branch complements and are,
in this aspect, indistinguishable from the taxa I
have retained in Merostachys. However, a comparison
of the shape and orientation of the blade
borne on culm sheaths representing their midculm
range, insofar as these features have been documented,
suffices to distinguish plants of taxa allocated
to the genus Rhipidocladum (Figure 4 2 ~ )
from plants of taxa retained in the genus Merostachys
(Figure 37~).
As Merostachys and Rhipidocladum are here
circumscribed, the component species of both
genera show a considerable range of variation in
morphological features of their inflorescences. The
range of these diversities is the more marked in
Rhipidocladum, where at one extreme it embraces
deceptive simulations of Merostachys and at the
other, Arthrostylidium. Fournier (1881: 131) was
misled by this latter case into transferring Steudel’s
Arthrostylidium racemiflorum to Merostachys.
As shown in the key below, the species currently
allocated to Rhipidocladum embrace, in
their inflorescences, six distinct patterns of insertion
and orientation of the spikelets. An intimation
of ostensible relationship between Rhipidocladum
and Merostachys appears in the noticeable to
marked secund orientation of the spikelets in the
spicate racemes of seven species of Rhipidocladum.
Rhipidocladum and Arthrostylidium coincide
here and there (compare R. harmonicum with
Arthrostylidium venezuelae, and R. urbanii with
A. cubense) with respect to certain forms assumed
by the inflorescences. As mentioned above, however,
the respective members of these two genera are
readily distinguishable by the characteristics of
their midculm branch complements.
DISTRIBUTIoN.-The eleven named species comprehended
by the present circumscription of the
genus Rhipidocladum represent taxa that occupy
open or forested mesophytic situations at low to
moderate altitudes in frost-free areas, from Mexico
to Brazil and Bolivia.
Key to Recognized Species of Rhipidocladum
la. Rachis deliquescent; spikelets pedicella te . , , , . , , , , , . , 11. R. verticillatum, new combination
1b. Rachis excurrent; spikelets sessile or subsessile ................................................................. .2
2a. Spikelets inserted oh the rachis in predominantly binate clusters ......................................
......................................................................................... 4. R. geminatum, new combination
2b. Spikelets of monate (solitary) insertion ............................................................................ ..3
3a. Rachis ,geniculate (at least in its distal portion) ; spikelets in distichous orientation,,,,. ..
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. R. harmonicurn, new combination
3b. Rachis not genticulate; spikelets in secund orientation,, ................................
4a. Spikelets oriented within a single plane giving a banneret-like form to each inflorescence
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. R. maxonii, new combination
4b. Spikelets not oriented within a single plane ........................................................................ 5
5a. Spikelets 4-5 cm long; the rachilla not disarticulating promptly at maturity ......................
5b. Spikelets not exceeding 2.5 cm in length; the rachilla disarticulating promptly at maturity
3. R. bartlettii, new combination; 7 . R. pawiflorum, new combination; 8. R. pittkd, new
combination; 9. R. prestoei, new combination; 10. R. rocemiflorum, new combination.
.................................................................................... 1. R. ampliflorum, new combination
(the following six species) : . . . . . . . . . . . . . . . . . . . . . . . . . 2. R. angustiflorum, new combination;
NUMBER 9
The six names (in couplet 5b) represent Hackel’s
(1903a:69) image of the genus Arthrostylidium as
tentatively adumbrated but not typified by him
(see p. 19). The corresponding binomials label
several closely related taxa that are poorly represented
by their type-specimens, and but weakly
differentiated from each other by their protologs.
Numerous currently available specimens represent
these and similar taxa too incompletely and too
incongruently to support confident decisions with
respect to the existence of either suspected synonymies
or possible specific or subspecific novelties.
Annotated Checklist of Recognized Species
of Rhipidocladum
1. Rhipidocladum ampliporurn (McClure) Mc-
Arthrostylidium ampliflorum McClure, 1942: 167,
HoLoTYPE.--Karsten s. n. (US). Specimens examined
include the holotype and elements of the
holotype collection (Us ex LE, US ex W).
Clure, new combination.
fig. 1.
DISTRIBUTION .-Venezuela.
2. Rhipidocladum angustiflorum (Stapf) Mc-
Clure, new combination.
Arthrostylilium angustiflorum Stapf, 1913:268.
H0LoTYPE.-“Communicated by Messrs. Sander
8e Sons, Bruges, 3rd May, 1912” (K). Specimens
examined include the type-collection (US ex K).
DrsTRIBLJTIoN.-“Tropical America.”
3. Rhipidocladum bartlettii (McClure) McClure,
Arthrostylidium bartlettii McClure, 1954:81.-Mc-
HOLOTYPE.--Bartlett 12154 (us). Specimen exnew
combination.
Clure in Swallen 1955:38.
amined, the holotype.
DIsTRIBuTI0N.-Guatemala.
4. Rhipidocladum geminatum (McClure) Mc-
Arthrostylidium geminatum McClure, 1942: 169,
HoLoTYPE.-“Jahn 125” (=Jahn 11, teste Dra.
Zoraida Luces de Febres) (VEN). Specimen examined,
the holotype.
Clure, new combination.
fig. 2.
DIsTRIBLJT1oN.-Venezuela.
5. Rhipidocladum harmonicum (Parodi) Mc-
Clure, new combination (Figure 4 2 ~ - ~ ) .
Arthrostylidium harmonicum Parodi, 1944:479, fig.
1.-McClure, in Sohns and Swallen, 1955: 133.
HoL0TYPE.-Vargas 3260 (Parodi herbarium).
Specimens examined include McClure 21416 (US);
Sodiro “25-Pilger” (US ex Sodiro herbarium);
Vargas 3260 (US ex Parodi).
6. Rhipidocladum maxonii (Hitchcock) McClure,
D1sTRIBuTIoN.-Ecuador; Peru.
new combination.
Arthrostylidium maxonii Hitchcock, 1927a:SO.
HoLoTYpE.-Maxon 8154 (us). Specimens examined
include the holotype and Standley 33643,
33812, 39096, 39105, 39179, 39411 (US); Standley
and Torres 50981 (US); Standley and Valerio
49781 (US).
DISTRIBUTION .-Costa Rica.
To the midculm branching habit that gives to
the culms of all species of Rhipidocladum a strong
merostachyoid appearance, R. maxonii adds another
merostachyoid feature - the conspicuously
secund orientation of the branches of its banneretlike
inflorescence. Other less conspicuous but clearly
divergent morphological features, however, testify
(in the key) to the authenticity of a generic disjunction
between Rhipidoclalum and Merostachys.
7. Rhipidocladum parviflorum (Trinius) Mc-
Clure, new combination.
Armdinaria parviflora Trinius, 1835: 6 19.
1891 (LE).
Arthrostylidium trinii Ruprecht, 183929.
HOL0TYPE.-“in sylvis Brasiliae pr. Ypenema
Bras. Riedl!” [I891 (LE). Specimens examined include
Macedo 4555 (US); Regnell ser. IIZ no. 1420
(US ex Regnell herbarium); Riedl189 (US ex LE);
Soderstrom 979 (US); Tamayo 2718 (US ex VEN).
8. Rhipidocladum pittieri (Hackel) McClure,
HOL0TYPE.-“in sylvis pr. Ypenkma Bras.” [Riedl
DIsTRIBLJTIoN.-Brazi~; Venezuela.
new combination.
Arthrostylidium pittieri Hackel, 1903a:75.
HoLoTYPE.--Pittier 7193 (W). McClure, in Swallen,
1955:40.
Arundinaria pittieri (Hackel) E. -G. Camus, 1913,
I:40.
Specimens examined include Grant 1036 (US ex
A); Heyde and Lux 4302 (US); Pittier 7193 (US);
Standley 78356 (US).
106 SMITHSONIAN CONTRIBUTIONS TO BOTANY
DISTRIBUTION.-costa Rica, Guatemala, Nicaragua.
In his original description of Arthrostylidium
pittieri, Hackel (1903a:76) gives this species only
2 lodicules but 3 stigmas; however, a specimen
from the type-collection-Pittier 7193 (US)-shows
lodicules typically 3 and stigmas typically 2. Some
lodicules in this specimen show an abnormality
in the form of a style-like apical prolongation
which is terminated by several (often 3) bristles.
It seems unlikely that this structure was misconstrued
by Hackel as the pistil. The occasional appearance,
however, of morphological anomalies in
the pistil itself (e.g., a supernumerary stigmatic
branch) might be responsible for the erroneous
impression recorded by him for this species.
9. Rhipidocladum prestoei (Munro) McClure,
new combination.
Arthrostylidium prestoei Munro, 1895: 186.
HoLoTYPE.--Prestoe [Trinidad Botanical Garden
Herbarium no. 16751 (K).
Arundinaria prestoei (Munro) Hackel, 1903d:516.
Specimens examined include Broadway 4922 (US
ex TRIN); Prestoe [Trinidad Botanical Garden
Herbarium no. 16751 (US ex B [compared with
type at Kew by Agnes Chase]).
10. Rhipidocladum racemiflorum (Steudel) Mc-
Arthrostylidium ? racemiflorum Steudel, 1854:336.
Merostachys racemiflora (Steudel) Fournier, 188 1 :
Specimens examined include Brother Angel 721
(US); Camp E-3814 (US); Curran 198 (US); Davidson
1297 (F); Ghiesbreght 234 (US ex LE, US ex
P); Liebmann 144 (US ex C); McClure 21221 (US);
Molina 8072 (US ex EAP); Skutch 3806 (US).
DIsmIBuTIoN.-~ana~ Zone, Colombia, Costa
Rica, Ecuador, Honduras, Mexico, Panama.
11. Rhipidocladum verticillatum (Nees) McClure,
new combination.
Arundinaria verticillata Nees, in Martius, 1829:
523.-Kunth, 1834:483, pl. 155, 156; 1835:348.-
Ruprecht, 1839:25, pl. 3: fig. 7.-Munro, 1868:
23.-Doell, in Martius, 1880: 166.
Ludolphia verticillata (Nees) Willdenow, 1833:25.
HOLOTYPE.-Sellow s.n. (€3). Specimens examined
DIsmIBuTIoN.-Trinidad.
Clure, new combination.
HoLoTYPE.-Gh iesbregh t 234 (?LE).
131.
include Glaziou 20248 (US ex P; US ex C),
Sellow s.n. (US ex B; US ex LE).
(F);
DIsTRIBUTIoN.-Brazil.
Swallenochloa McClure, new genus
FIGURES 43-45
Plantae unicespitosae, inermes. Rhizomata pachymorpha.
Culmi sese sustinentes, in speciebus plerisque
adhuc saltem cognitis rigide erecti; internodiis
fistulosis, supra locum insertionis vel ramorum vel
gemmarum complementi tantum vel breviter vel
prorsus sulcatis, alibi teretiblis. Ramorum complementum
ad nodos culmorum medianos insidens
raro absens, nonnumquam monocladum, typice
solutim pleiocladum, axibus primariis vulgo 3
(usque 5), ad libram confertissime insertis et ex
gemmis discretis, unoquoque mox vel demum ex
gemmis ad nodos proximos suos proliferanti. Foliorum
lamina venulis transversis in speciebus nonnullis
haud visibilibus alibi plus minusve Clare
manifestis.
Inflorescentiae semelauctantes, typice confertim
paniculatae, lineares et secundae; rhachidi percurrenti;
ramis omnibus brevibus, plus minusve valde
appressis; prophyllis bracteisque nullis. Glumae
transitionales imo spiculae insidentes 4: I et 11
“glumae vacuae,” 111 et IV “lemmata (vacua) sterilia.”
Spiculae pleraeque florem perfectam unicam
terminalemque continentes; rhachilla post paleam
haud producta. Anthoeciorum fertilium maturorum
lemma paleam suam imo tantum circumnectans.
Palea antice hians, saltem apicem versus vulgo et
sulcata et nonnumquam plus minusve manifeste
bicarinata, dorso alibi convexa. Lodiculae 3, duabus
anterioribus aequalibus, plus minusve valde asymmetricis
et congruenter geminatis, posteriore symmetrica
et aliquanto minore. Stamina 3, filamentis
filiformibus liberisque. Stigmata 2, quorum unum
nonnumquam sed rarissime supra basin suam in
duas partes divisum. Fructus maturus non inventus.
FIGURE 43.--Swallenochloa subtesseliata (Hitchcock) McClure.
A, Leafy flowering branch complement, x 0.6; B, apex of
leaf sheath and base of leaf blade, x 3; c, base of branch
complement high on culm, x 3; D, tuft of ultimate branches
of inflorescence with transitional glumes attached, x 7.2; E,
spikelet, x 7.2; F, sterile lemma I, x 7.2; G, sterile lemma 11,
x 7.2; H, fertile lemma, x 7.2; I, palea, x 7.2; J, lodicule
complement, x 25; K, stamen, x 15; L, gynoecium, x 30.
All based on Tonduz 3367, type-species (US).
NUMBER 9 107
B
108 SMITHSONIAN CONTRIBUTIONS TO BOTANY
Plants unicespitose, unarmed. Rhizomes pachymorph.
Culms self-supporting, the internodes typically
fistular, either terete or lightly to strongly
sulcate above the locus of insertion of a complement
of buds or branches. Branch complement at
midculm nodes rarely lacking, occasionally monoclade,
typically unrestricted pleioclade, the initial
components most commonly 3 in number (rarely
to 5), unequal (each of independent origin, i.e.,
developing from a separate primary bud), of level
insertion (with their bases close together in a
horizontal row) the middle one dominant, each
usually proliferating from buds at its proximal
nodes. Leaves (leaf sheath blades) with transverse
veinlets not at all evident or more or less clearly
manifest externally.
Inflorescences semelauctant, commonly terminal
to the culm as well as to leafy or leafless lateral
axes, typically linear in profile, compact-paniculate,
the rachis percurrent, the primary branches of
secund orientation, those of all orders usually
short and more or less strongly appressed; prophylls
and bracts lacking throughout. Transitional glumes
at the base of the spikelet 4: I and 11 “empty glumes”
in some species sub-obsolete; 111 and IV “sterile
(empty) lemmas.” Spikelets typically containing
but one perfect floret, the rachilla very short, potentially
disarticulating immediately below the
locus of insertion of the first sterile lemma and of
any fertile lemma, and not prolonged behind the
palea of the terminal floret. Fertile lemma fully
embracing its palea only basally at maturity. Palea
gaping antically, sulcate and weakly to strongly
bicarinate for a short distance at the usually emarginate
or bidentate or bimucronate apex, dorsally
convex elsewhere. Lodicules 3, the anterior two
typically asymmetrical and congruently paired, the
posterior one smaller (at least either shorter or
narrower) and symmetrical. Stamens 3, the filaments
filiform free. Stigmas 2, one of them sometimes
(but rarely) divided above the point of its
origin as observed in one flower of Weberbauer
4415, (type-collection of S. weberbaueri). Mature
fruit unknown.
ETYMoLocY.-The name Swallenochloa commemorates
the initial recognition of this genus by Dr.
Jason R. Swallen, well known for his extensive
field studies of New World grasses, and for his numerous
contributions to the taxonomy of the
Gramineae.
TYPE-SPECIES.-SWa~lenoch~oa subtessellata (Hitchcock)
McClure.
RELATIoNsH1Ps.-Members of Swallenochloa share
with members of Chusquea and members of Neurolepis
a basically similar spikelet structure. Beyond
this, Swallenochloa touches each of the other
two at a few points where one or two technical
features exceptional in one genus are sliared by
one of its members with a member of one of the
other genera. Examples follow. A specimen under
Pittier 3069 (US 677457)-cited by Hitchcock
(1927a181) as Chusquea subtessellata Hitchcock, now
Swallenochloa subtessellata (Hitchcock) McClure,
from which it differs in several technical detailsappears
to conform to the other members of Swallenochloa
in all generic aspects except the profile
and dimensions of the strongly chusqueoid inflorescence.
Specimens of an undescribed species of
Chusquea, under McClure 212?4 (US) have the
satellite buds and branches at midculm nodes reduced
to two, as is usual in Swallenochloa, but in
all other technical respects the material clearly falls
within the natural boundaries of Chusquea. A
whole plant under Fosberg 20856 (US 273832)
shows unbranched culms with hollow internodes
and a terminal inflorescence as in Neurolepis, but
the nature of its foliage places it clearly in Swallenochloa.
Bamboos of the genus Swallenochloa differ most
conspicuously from bamboos of the genus Chusquea
by combining inflorescences of linear profile
with distinctive features of the vegetative organs,
including hollow culm internodes, primary complement
of buds or branches at midculm nodes comprising
fewer members, presenting several markedly
distinctive morphological aspects, and foliage leaves
FIGURE 44.-?wallenochloa subtessellata (Hitchcock) McClure.
A, Habit of leafy flowering branch showing tessellation of
leaf blades, x 0.6; R, sheath subtending flowering twigs
(left) and detached blade (right), x 1.2; C, apex of leaf
sheath and base of leaf blade, x 7.2; D, tuft of ultimate
branches of inflorescence with transitional glumes attached, x
7.2; E, spikelet (without pedicel and transitional glumes) , x
7.2; F, sterile lemma I, x 7.2; G, sterile lemma 11, x 7.2; H,
fertile lemma, x 7.2; I, palea, oblique aspect; J, palea, dorsal
aspect, x 7.2; K, lodicule complement, x 15; L, stamen,
x 15; M, gynoecium, x 30. All drawings based on Pittier
3069, a variant form of the species, (US).
m
110 SMITHSONIAN CONTRIBUTIONS TO BOTANY
generally showing definitely xeromorphic structure,
texture, and habit.
Swallenochloa subtessellata (as Chusquea subtessellata)
is by Niilo Virkki (1963) given a chromosome
number 2n = 36. Janaki Ammal (1959) gives
three species of Chusquea the chromosome number
of 2n = 48.
The genus Swallenochloa shares the following
features with the genus Neurolepis: hollow culm
internodes, a similar spikelet structure, and a strong
tendency to secund orientation in the branches of
the inflorescences. However, members of Swallenochloa
differ from members of Neurolepis most
conspicuously in the dense and extensive development
of individual plants, the freely branched
culms; the smaller size, bamboo-like shape, xeromorphic
texture, and upright habit of the foliage
leaves; and the very much smaller and generally
more compact inflorescences, which are not confined
to the terminal position on the culms.
Among the many available specimens that clearly
belong to the genus Sz~lallenochloa but remain unidentified
as to species, certain features of the reproductive
structures show a high incidence of
intergradation. For this reason, the differentiation
and definition of species or other subgeneric taxa
on the basis of typical conventional herbarium
specimens presents difficulties at many points. The
high incidence of abortive or distorted structures
in the spikelets of many of the variants, and the
scarcity of normal, mature fruits suggest (1) the
possibility of a hybrid origin for the genus or the
continued but only partially successful introgression
between its members, and (2) the existence
of insects or fungi whose parasitic activities may
be responsible for the widespread incidence of distortion
shown by the spikelets. As matters now
stand, so great is the incidence of variation and
distortion in the spikelets, and so obscure are the
bases for these diversities, that it is difficult to determine
with confidence a suitable taxonomic disposition
for many of the available collections.
The genera Chusquea, Neurolepis, and Swallenochloa
represent groups of species that are
closely related in the essential features of the reproductive
apparatus, members of Chusquea leading
in the diversity of form expressed in the
branching of the rhizomes, culms, and inflorescences.
From the other bamboo genera of the New
World flora, these three are set off, as a group, by
the regular appearance in their members of two
usually empty “sterile lemmas” following the second
“empty glume” at the base of the spikelet, and
by the termination of the spikelet in a perfect
floret, without a prolongation of the rachilla behind
it. Chusquea is unique among the three in
the efistulose structure of all segmented vegetative
axes; in the universal presence of more than one
branch bud at midculm nodes; and in the constellate
insertion of the branch buds and members of
the branch complement at midculm nodes. Species
of Chusquea show, in aggregate, the most extreme
range of ecological tolerance, or preference. Neurolepis
recedes from the other two members of the
group in the coarse, grass-like character of the
plant; the usually large, persistent leaf blades; the
universal lack of branching in the aerial range of
the vegetative part of the culm; and the usually
rather weak development of the rhizome system. At
least some species of Neurolepis are found at relatively
high altitudes and in ecological formations
characterized as pkramo. Species of Swallenochloa
form dense thickets, consisting of erect culms with
hollow internodes, branch buds and branches at
midculm nodes in level (not constellate) insertion.
Their natural distribution is confined to high altitudes,
commonly in ecological formations some of
which are characterized as pPramo.
The sharing by Swallenochloa, Chusquea, and
Neurolepis of features not found elsewhere among
New World bamboos raises questions as to the phylogenetic
relationship of these three genera. Could
Swallenochloa be the product of introgression between
members of Chusquea and Nezirolepis that
FIGURE 45.-Szoallenochloa tessellata (Munro) McClure. A,
Small leafy flowering culm, x 0.6; B, leafy flowering culm
with its pachymorph rhizome, x 0.6; c, midculm complement
of three branch buds, x 3; D, base of midculm complement
of three independent primary branches, x 3; E,
apex of leaf sheath and base of leaf blade, x 6; F, secondary
branch of inflorescence, bearing three pedicellate spikelets,
x 3; G, spikelet, x 6; H, two transitional glumes, X 6; I, two
“sterile (empty) lemmas,” x 6; J, “fertile lemma,” x 6;
K, palea, x 6; La-c, lodicule complement, x 12; M, stamen,
x 6; N, gynoecium, x 24; 0, rhizome and base of culms,
x 0.6; P, cross-section of rhizome, x 5.4. Drawhg A based
on Triana 306, type-species (US) ; B, F-N on Schuttes 18771
(US), c, P on Garcia-Barriga 11643-A; D, E on Garcla-
Barriga 11684 (US), and o on Soderstrom 1339 (US),

112 SMITHSONIAN CONTRIBUTIONS TO BOTANY
are, or have been, sympatric in the high altitudes
where the three genera meet today?
DIsTRIBUTION.-~rigina~~y allocated to the genus
Chusquea, the five bamboos currently recognized as
species of Swallenochloa, together with numerous
collections representing as yet undescribed congeneric
diversities, are all endemic to the New
World. Their aggregate geographic range of distribution
extends from Bolivia and Brazil to Costa
Rica, and embraces areas distinguished by characteristic
complexes of ecological factors locally
given distinctive names such as “piramo,” “subpiramo,”
etc. Recorded altitudinal limits fall between
2700 m and 4000 m above sea level. The
highest density of incidence appears in Colombia.
Fosberg (1944) describes in extensive and vivid
detail the plant life and the ecological features of
the Piramo de Sumapaz of Colombia where he collected
a species of Swallenochloa referred to as follows
(1944:230).
Another grass (Chusyuea weberbaueri) is locally abundant
and conspicuous. It, too, is very stiff, but it usually
forms colonies rather than bunches, and its blades are flat
and lanceolate. It is not climbing like other species of its
genus. In exposed situations it is low but erect, 6 or 8
decimeters tall when not in fruit, and in very unfavorable
[situations] it even fruits at 1 or 2 decimeters. Usually the
spikes make it somewhat taller. In wetter places it becomes
a meter or more tall, and in sheltered spots, it forms small
cane-brakes up to 3 or 4 meters deep and very hard to
push one’s w‘ay through.
The field notes that accompany the specimen
(Fosberg 20856, US 2181972) read as follows: Colombia:
“Cordillera Oriental, Dept. Cundinamarca,
Cordillera de las Cruces, Piramo de Sumapaz. Common
everywhere on piramo slopes; rough piramo
plateau with wet forest running up the sides to
about 3500 m, brush, up to 3650 m. [Plants] varying
greatly in size. Aug. 19, 1934.” The specimen
represents an apparently undescribed species of
Swallenochloa.
Weberbauer in his monumental work entitled
El Mundo Vegetal de 10s Andes Peruanos (1911
[1945: 1461) says of three species of Swallenochloa
(under the names Chusquea spicata, C. depauperata,
and C. weberbaueri) that “these dwarf forms
occupy peat bogs and open grasslands at elevations
between 3200 and 3500 meters” and that “with
them are associated members of the curious genus
Neurolepis, whose ensiform leaves emerge from the
base of the stem and reach, in N. weberbaueri, a
length of 2.5 meters.”
The present published content of the genus
Swallenochloa is limited to five recognized species
that were originally allocated to the genus Chusquea.
la.
lb.
2a.
2b.
3a.
3b.
4a.
4b.
Key to Currently Recognized Species of Swallenochloa
Transitional glumes I and 11 (“empty glumes”) subobsolete; branches of inflorescence entirely
glabrous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. S. depauperata, new combination
Transitional glumes I and 11 (“empty glumes”) about 1 mm or more in length; branches
of inflorescence more or less noticeably to densely strigose ...........................
Branches of inflorescence heavily waxy; spikelets tinted with wine , , , , . , ,. , , , , , ,,
Branches of inflorescence not noticeably waxy; spikelets of stramineous hue
Transitional glume IV (upper sterile lemma) almost as long as the lemma, a
well-developed veins on each side of the midrib . . . 3. S. subtessellata, new combination
Transitional glume IV (upper sterile lemma) less than 3/4 as long as the lemma, and show-
4. S. tessellata, new combination
Anthecium dull; pale stramineous; awn of lemma glabrous . . . 2. S. spicata, new combination
Anthecium lustrous; dark stramineous: awn of lemma antrorse-scabrous ...............
................................................................................... 5. S. weberbaueri, new combination
ing only one well-developed vein on each side of the midrib ....................
.........................................................................
Annotated Checklist of Recognized Species of the
1. Swallenochloa depauperata (Pilger) McClure, 2. Swallenochloa spicata (Munro) McClure, new
Chusquea depauperata Pilger, 1905: 149.
HoLoTYPE.-Weberbauer 3709 (B).
Genus Swallenochloa DISTRIBUTION.-Peru.
new combination. combination.
Chusquea spicata Munro, 1868:60.
NUMBER 9 113
HOLOTYPE.-Lechler 2154 (K) the principal element.
Chusquea humilis Lechler ex Munro, 1868:60.
ment.
Chusquea simplicissima Pilger, 1905: 145.
HoLoTYPE.-kchkr 2694 (K) the principal ele-
HoLoTYPE.-Weberbauer 2217 (B).
Hitchcock (1927b:310) made Chusquea simplicissima
Pilger a synonym of Chusquea spicata
Munro. I cannot with confidence verify this disposition
of the matter on the basis of the documenting
material available to me. Dr. Hitchcock’s
decision, however, was based on a study of the intact
type-specimens of which I have seen only fragments
in which the spikelets do not exemplify a
clearly representative norm.
3. Szuallenochloa subtessellata (Hitchcock) Mc-
Clure, new combination (Figures 43, 44).
Chzisq uea subtessellata Hitchcock, 1927a: 8 1.
DISTRIBUTION.-PerU.
HoLOTYPE.-TOnduz 3?67 (us).
DIsTRIBUTIoN.-cOsta Rica.
Pittier 3069 (US) cited by Hitchcock (1927a:81)
as belonging here deviates noticeably from the holotype
in both vegetative and reproductive features
(cf. Figure 44).
4. Swallenochloa tessellata (Munro) McClure, new
combination (Figure 45).
Chusquea tessellata Munro, 1868:60.
97 (K).
TYPE-iVATERIAL,-Tria?’Ia [306] (K) and Holton
DIsTRIBuT1oN.-Colombia.
Munro gives the altitude at which Triana’s collection
was made at 3000 feet. In Triana’s field
notes the same figure is given in meters.
5. Swallenochloa weberbaueri (Pilger) McClure,
Chusquea weberbaueri Pilger, 1905: 146.-Fosberg,
new combination.
1944: 230.
HoL0TuPE.-Weberbauer 4415 (B).
~IsTRIBuTIoN.-Co~ombia to Peru.
Yushania K. H. Keng
FIGURES 46, 47
Yushania Keng f., 1957:355.
Plants unarmed, compactly unicespitose at first;
mature clumps of pluricespitose habit in Y. niitakayamensis
and of diffuse (open) habit in Y. aztecol-
um. Rhizomes pachymorph, the rhizome neck
in Y. niitakayamensis and in Y. aztecorum becoming,
in plants of mature stature, up to ca 50 cm
long. Culms of small or medium stature, self-supporting;
each midculm node bearing a single initial
branch bud; margins of the prophyllum not fusea;
the initial primordium giving rise (simultaneously
or nearly so and apparently before the germination
of the bud) to typically three growing points, these
becoming three subequal primary axes with a common
base. The germination of each primary branch
bud is accompanied by the apparently simultaneous
elongation of its prophyllum. Branch complement
arising at midculm nodes with the first three component
axes typically dominant over any axes that
proliferate from buds at their proximal nodes.
Sheath at midculm nodes typically lacking a wellmarked
basal girdle. Leaf blades with transverse
veinlets not at all to weakly to sometimes strongly
manifest externally.
Inflorescences semelauctant, forming contracted
to more or less prolifically branched panicles, their
branches epulvinate and eprophyllate, with subtending
bracts (excepting the somewhat well-developed
basal one) small to rudimentary to obsolete.
Spikelets few-flowered to several-flowered, each
terminating in a depauperate sterile anthecium.
Transitional glumes typically 2, empty. Lemma of
fertile florets embracing its palea only basally at
maturity. Palea broadly sulcate and 2-keeled dorsally,
the margins not at all or barely imbricate.
Lodicules 3, the anterior 2 more or less asymmetrical
and paired, the posterior one smaller and symmetrical.
Stamens 3, the filaments filiform, free.
Stigmas 2. Mature fruit, as far as known, a subfusiform,
sulcate apically mucronate caryopsis, the
pericarp pergamineous, glabrous, of uniform thickness,
the sulcus and the embryotegium clearly manifest.
ETuhtoLoGY.-The name Yushania, yu, jade, and
shan, mountain, is a Latinized form of the two
components of the Chinese name (in the National
dialect) of the type-locality of Y. niitakayamensis,
type-species of the genus. A renowned geographical
feature of Formosa (the Island of Taiwan), Yushan
is known in Japanese as Niitakayama, and in English
as Mount Morrison.
TYpE-spEcIEs.-I’ushania niitakayamensis (Hayata)
Keng f. (see McClure, 1959:209). Based on
114 SMITHSONIAN CONTRIBUTIONS TO BOTANY
Amndinaria niitakayamensis Hayata (1907:79).
RELATIONSHIPS.-The simultaneous or subsimultaneous
emergence, at midculm nodes, of more than
one primary branch in Yushania recalls a similar
feature in Apoclada. In Apoclada, however, the
several primary axes of a midculm branch complement
apparently do not have a common base, and
the respective primordia from which they arise are
apparently not initially enclosed in a common
prophyllum (bud scale) as they are in Yushania.
Members of Yushania as the genus is herein circumscribed,
resemble members of the genus Au-
Zonernia in the paniculate form of the inflorescence.
Plants allocated to Yushania, however, differ most
basically from plants allocated to Aulonemia in
terms of ontogeny (and certain details of the form)
of the midculm branch complement. In members
of the genus Yushania, the primordium of the
initial bud at midculm nodes, while still enclosed
in the prophyllum, gives rise to typically 3 initial
growing points in quick succession. This situation
suggests the presence of the phenomenon conventionally
referred to as the “telescoping” of an axisactually
the shortening of the interval between the
loci of emergence of the successive growing points
produced by a body of primordial tissue. A satisfactory
description of the distinctive sequence of
events that produces the midculm branch complement
that is characteristic of members of the genus
Yushania awaits the intervention of anatomical
studies designed to supplement the results achievable
by way of gross morphology.
DISTRIBuTIoN.-In the Chinese text that follows
his Latin diagnosis of the genus, K. H. Keng (1957:
356) says of the Old World component of Yushania:
“Of this genus only the type species is known at
present. It is native of Taiwan, and is found also
on Luzon, P.I., and in the mainland Chinese provinces
of Szechuan and Yunnan.” The known distribution
of the New World component of the
genus (subgenus Otatea) extends from Mexico to
Honduras. Both the Old World component and
Key to Subgenera of Yushania
la. Culms proliferating from buds at their subterranean nodes; inflorescences with few spikelets:
lemma weakly or not at all awned; palea acutely bifid apically; bamboos endemic
to areas of the Old World Subgenus Yushania
lb. Culms not proliferating from buds at their subterranean nodes; inflorescences with numerous
spikelets; lemma conspicuously awned; palea obtuse apically; bamboos endemic to
areas of the New World Otatea, new subgenus
FIGURE 46.-Yushania aztecorum McClure and E. W. Smith.
A, Diagram of a portion of the rhizome system (greatly
reduced) of a plant of mature stature, with culms and roots
omitted from the rhizomes proper: the rhizome necks are
naturally budless and rootless. In young plants of the species
illustrated, the necks are shorter than the rhizomes proper,
but as a plant approaches its mature stature the necks of
successive orders (generations) of rhizomes become progressively
more elongated. [This feature has not yet, however,
been established as generic in Yushania.] B, Basal portion
of a culm, showing the characteristically short internodes and
wire-like fibrovascular remains of the weathered, nonabscissile
culm sheaths, x 0.6; c, midculm internodes (greatly
reduced) showing the shallow groove (sulcus) that extends
upward from the locus of insertion of each primary bud;
D, midculm node, showing the solitary primary branch bud
with its prophyllum intact, x 1.5; E, primary branch bud at
a midculm node, with the prophyllum removed, revealing
three growing points, each a prophyllate but unsubtended
bud, x 1.5: F, diagram of the pattern of the loci of insertion
of the prophylla of the three branch buds portrayed in E,
in relation to the prophyllum of the initial primordium
(the body of special meristematic tissue from which the
three buds emerge), x 4.8; G, base of a midculm branch
complement, showing the initial components in an early
stage of development, with sheaths removed, x 0.9; H, midculm
branch complement with the initial components fully
developed, x 1.2; I, base of midculm branch complement
in a later stage of development, after the production of
axes of higher orders following tht germination of buds
at the proximal nodes of the initial component axes, x 1.2;
J, nonflowering leafy branches (distal portion only) from
culms of relatively advanced age, showing solitary nature
of ramification at this level, leaf blades in profile, x 0.6; K,
leaf sheath apex with petiole and base of a narrow leaf
blade, x 7.2; L, culm sheath from midculm range (apex
only), adaxial aspect (left) and abaxial aspect (right) ,
x 0.9. All drawings based on McChre 21204 (US).
NUMBER 9 115
C
116 SMITHSONIAN CONTRIBUTIONS TO BOTANY
the New World component of the genus Yushania
embrace taxa still requiring intensive study in field
and laboratory to determine their infrageneric
status. The much-needed monographic treatment
of the genus should, in my opinion, be deferred
pending the completion of such studies,
Subgenus Otatea McClure and E. W. Smith,
new subgenus
Culmi e gemmis suis subterraneis haud proliferentes;
inflorescentiae spiculas multas gaudentes;
lemma aristatum dorso dense strigosum; palea apice
obtusa; plantae in terris occidentalibus indigenae.
ETYMoLoGY.-The name Otatea is a Latinized
form of the vernacular name otate (corrupted form
of a word from the Nahuatl, language of the Aztecs)
by which some native bamboos (more particularly
those pertaining to this subgenus) are
currently known and referred to in Mexico.
TYPE-SPECIES.-YUS~U~~U aztecorum McClure and
E. W. Smith,
1. Yushunia (Otatea) aztecorum McClure and
E. W. Smith, new species
FIGURES 46, 47 A-M
Plantae initio conferte caespitose demum sensim
aperientes. Rhizomatum collum initio (in plantis
staturae juventae) rhizomate proprio suo brevius,
demum (in plantis staturae adultae) rhizomate
proprio multo longius. Culmi usque ad 6 m alti
et 2.5 cm diametro; internodia glabra, in plantis
juvenibus intus farcta, in plantis adultis fistulosa,
ea supra nodos medianos ramiferos sulcata. Ramorum
complementum ad nodos medianos oriens
initio axes primarios 3 subequales subcoaetaneosque
vulgo capiens; axes ordinum superiorum a
primarii valde superati. Vagina internodios culmi
fovens dorso initio setis castaneis fragilis vel sparse
vel dense vestita, demum calvescens sed papillis
persistentibus conspersa; auriculae et setae orales
vel nullae vel invalide evolutae; ligula arcuata,
margine pro parte ciliata et pro parte crasse fimbriata;
lamina appressa et late vel anguste triangula;
vaginae ad nodos inferiores culmorum
orientes haud deciduae, in situ demum fatiscentes.
Vagina foliorum prominenter nervoso-striata, inter
nervos plus minusve dense usque haud papillata,
vel omnino glabra vel hic illic vel omnino
puberula; auriculae et setae orales vel vix ullae vel
nullae; ligula interior et ligula exterior dimensionibus,
forma, margine vesturaque maxime variabiles;
lamina foliorum vel lanceolata cum nervis secundariis
usque 3-4, vel lineari-lanceolata usque linearis
et nervos secondarios vulgo 1-2 praebens;
venulae transversae extus haud usque Clare manifestae;
pubescentia laminae charactere, crebretate,
et distributione maxime variabilis.
Inflorescentiae ramos et ramulos vel foliiferos vel
efoliatos terminantes; pedunculus, rhachis et radii
scaberuli; bracteae sat minimae nonnunquam obsoletae.
Spiculae omnino pallide strigosae, laxae,
usque ad 4 cm longae, flosculos 3-7 fungentos capientes,
ubi maturae sat friabiles. Glumae transitionales
typice 2, lanceolatae, valde carinatae, apice
in aristam scabram ca 2.5 mm longam terminantes;
I (dempta arista): 2.5-3.5 mm longa, nervos secundarios
1-2 praebens; 11 (dempta arista): 3.5-4.5 mm
longa, nervos secundarios 2-3 praebens. Rachillae
segmentum unumquidque dimidium lemmatis sui
superantia, omnino scabrosa, gracilia, apice subito
dilatato conspicue ciliata. Lemma floscularum
FIGURE 47.-Yushania artecoruna McClure and E. W. Smith
(A-M) and Yushania sp. (N) . A, Examples of leaf blades
showing the range in size and shape encountered within a
single population. The largest lanceolate one was taken from
a young flowering culm. The long, narrow one represents
an associated form borne by older culms in both the vegetative
state and in a weakly flowering state. The several
progressively smaller ones were taken from flowering
branches borne on culms of progressively more advanced
age, x 0.6. B, Leaf sheath apex with petiole and base of a
broad leaf blade, x 7.2; c, flowering branch-one of three
initial component axes of a branch complement high on a
young culm-with several contracted and leafless flowering
branches arising from buds at the proximal nodes of the
initial axes, x 0.6 D, tip of a flowering branch from a culm
of relatively advanced age, characterized by a pronounced
reduction in the size of the foliage leaf and the inflorescence
proper, x 1.2; E, spikelet, x 1.8; F, transitional (empty)
glumes, x 6; c, floret, x 6 H, lemma, x 6; I, palea, x 6;
J, lodicule complement, x 12; K, stamen, x 6; L, gynoecium,
x 12; M, seedling, x 0.6; N, fruit, embryo side, showing
embryotegium (left), and hilum side (right), showing the
sulcus (conventionally referred to as a hilum) , x 7.2. Drawings
A-M based on McClure 21204 (IJS) and N on Ortega
4341 (US).
NUMBER 9 117
118 SMITHSONIAN CONTRIBUTIONS TO BOTANY
fungentium lanceolatum usque (dempta arista scabra
2.5 mm longa) 9 mm longum, prominenter carinatum,
alibi nervos secundarios 2+ 3 praebens.
Palea usque 9 mm longa, oblonga, apice integra
truncata, inter et secus carinas strigosa, nervis secundariis
extus haud manifestis. Lodiculae subopacae,
nervosae, utrinque glabrae, apice ciliis albidis
ca 0.5 mm longis fimbriatae; duae anticae (demptis
ciliis) usque ad 1.5 mm longae, leniter asymmetricae
et gemmatae, quasi pulvinatae; postica (demptis
ciliis) usque ca 1.0 mm longa, elliptica. Antherae
maturitate stramineae, 5-6 mm longae. Ovarium
glabrum. Fructus non adhuc inventus.
Plants at first compactly cespitose, becoming diffuse
in habit as they mature. Neck of rhizomes at
first shorter than the rhizomes proper, becoming
longer in successive annual innovations, and eventually
much longer than the rhizomes proper as the
plant approaches its mature stature. Culms up to
6 m tall and 2.5 cm in diameter, with glabrous
internodes pith-filled in young plants and hollow in
plants of mature stature; internodes above ramiferous
midculm nodes sulcate. Branch complement
at midculm nodes at first typically comprising 3
subequal primary axes that emerge simultaneousiy
or nearly so and dominate axes of higher orders
that develop subsequently from buds at their own
proximal nodes. Culm sheaths at lower (unbranched)
nodes of the culms (Figure 46 B, L)
not at all deciduous, slowly disintegrating in
place. Leaf sheaths prominently nerved-striate,
more or less densely to not at all papillate between
the nerves; glabrous throughout to puberulous here
and there or puberulous throughout; auricles and
oral setae weakly developed to lacking entirely;
inner and outer ligules highly variable in size,
shape, margin, and vesture. Blade of leaf sheath
varying in shape from lanceolate with up to 3-4
secondary nerves, to linear with 1-2 secondary
nerves, the transverse veinlets not at all to clearly
manifest externally; pubescence of the leaf blade
highly variable in character, density and distribution.
Inflorescences terminating leafy or leafless
branches and twigs; peduncle, rachis, and ramifications
scaberulous; bracts rather small and
sometimes obsolete. Spikelets up to 4 cm long,
comprising 3-7 functional florets, obscurely and
pallidly strigose throughout, open, more or less noticeably
zig-zag at maturity. Transitional glumes
typically 2, lanceolate, strongly keeled, terminating
in a scaberulous awn about 2.5 mm long; I (without
the awn): 2.5-3.5 mm long, with 1-2 secondary
nerves; 11 (without the awn): 3.5-4.5 mm long,
with 2-3 secondary nerves. Rachilla segments
slender, each more than half as long as its lemma,
conspicuously ciliate at the abruptly dilated apex,
elsewhere slender and scaberulous; readily disarticulating
at maturity. Lemma of functional florets
lanceolate, up to 9 mm long (without the ca
2.5 mm long scaberulous awn), prominently keeled,
with 2-3 secondary nerves. Palea up to 9 mm
long, oblong, entire and truncate apically, strigose
along the keels and between them; secondary
nerves not manifest externally. Lodicules subopaque,
nervose, glabrous on both faces, fimbriate
apically with cilia about 0.5 mm long, the two
anterior ones about 1.5 mm long without the cilia,
weakly asymmetrical and paired, apparently pulvinate
at anthesis, the posterior one elliptical, up to
about 1.0 mm long without the cilia. Anthers stramineous
at maturity, 5-6 mm long, the connective
not exserted. Ovary glabrous. Fruit not yet found.
HOLOTYPE.--F. A. McClure 21204 (US), collected
at about 450 m on La Dispensa, an elevated area
inland from Rosario, State of Sinaloa, Mexico, 5
August 1943. Vernacular name Otatk.
RELATIONsHIPS.-The foregoing description is
based principally upon the most complete available
specimen, McCZure 21204 (US). It is supported in
this and that detail by specimens from thirty-five
other collections, the bulk of which are too fragmentary
for confident infrageneric taxonomic disposition.
DIsTRrnuTIoN.-Field notes accompanying available
specimens of thirty-six collections of members
of the subgenus Otatea examined are generally
meager. Most commonly encountered in herbaria
are specimens from Mexico where, as observed by
the senior author of this subgenus, plants of this
subgenus often occupy relatively (seasonally?) dry
sites, sometimes in association with cacti. Specimens
collected in El Salvador and Honduras are
more rarely seen. Plants we recognize as belonging
NUMBER 9 119
to the subgenus Otatea have been collected at reported
elevations from 200 m to 2700 m in Mexico
and Central America.
2. Yushunia (Otatea) acuminata (Munro)
McClure, new combination
Arundinaria acuminata Munro, 1868:25.
120 SMITHSONIAN CONTRIBUTIONS TO BOTANY
FIGURE 48.-Chart of Generic Characters. Fruit: view from embryo side (left), view from hilurn
side (right). Stigma: number of stigmas. Number in parentheses indicates that the number is
of rare occurrence. Stamen: number of stamens. Number in parentheses indicates that the
number is of rare occurrence. Spikelet Apex: Diagrammatic representation of the apical element
of a spikelet. Male-female sign indicates a perfect floret. All others indicate empty florets
reduced to varying degrees, in some cases quite depauperate. Br Subt: branches of the inflorescence
(at least the lowermost ones) subtended by a bract. Br Proph: Branch prophyllum; plus
indicates each branch of the inflorescence is subtended by a prophyllum at its basal node.
Znfl Br: type of inflorescence branching. Pan, panicle or paniculate; Rac, raceme or racemose;
Con to Ope, congested to open; Sec, secund; Sol, solitary; Spi, spicate; Div, diverse: Cap, capitate;
Dif, diffuse; Dig, digitate. Infl Dew inflorescence development. S, semelauctant; I, iterauctant.
Rhizome: L, leptomorph; P, pachymorph; LP, leptomorph and pachymorph condition existing
within the same plant. Dotted lines indicate that the rhizome system has not been seen but
assumed to be pachymorph. Midculm Prophyllum and Primordium: Primary buds at midculm
node consisting of the initial primordium (solid black triangle) covered by the prophyllum
(cone). The single initial primordinm represented by zigzag lines indicates that it gives rise to
a specialized primary element with buds of apsidate insertion. Dotted lines indicate what the
situation is assumed to be, although not seen. Initial (Pritnary) Branch: Either lacking, a single
dominant one, three or more subequal ones, or many subequal ones arising from a primary
element in apsidate fashion. Open triangles represent buds capable of germination while the
black dots represent buds which do not germinate. L/S Znt: Long/short internodes. In plants
of those genera marked with both a plus and a circle, there is a tendency for culms to have
elongated internodes followed by shortened internodes. A question mark within the circle
indicates that this feature possibly may not be present in some culms.
NUMBER 9 121
Midculm Initial
Prophyllum and (Primary)
Primordium Branch
Fruit Genus
QO
QO
Arundinuriu
Chuspea
Not yet
known
sec
Con
Pan Swatlemchloa
Pan
Muriocladzu J
Notyet
known
Pan 08
Aubmmia A Pan
Spi
Rae
Div
see
Spi
Sol
RE c
_I
(See
descr, -
&&, -
-
Arthrosty lidhm
Not yet
known
Not yet
known [
f
D if
Pan
Not. yet
known
Dig
&p
The following terms are defined with special
reference to the usages adopted by the author for
describing various characteristics of the bamboo
plant. Most of these first appeared in the glossary
of McClure (1966b:295-318), reprinted in Taxon
(McClure 1966a: 220-235). In that glossary newly
coined terms were indicated by an asterisk (*);
others, indicated by a dagger (f), were given new
or extcnded meanings to describe conditions not
satisfactorily covered by conventional usage, or by
botanical glossaries.
Some of the following definitions incorporate
more 01 less drastic modifications that, in the present
dissertation, are given precedence over the
definition of the same terms that appear in the
above publications. For example, the content given
the terms monoclade and pleioclade in the earlier
glossary has been drastically changed. When unqualified
(standing alone) these terms now refer
exclusively and to the first order element (s) of a
branch complement. The qualifying adjectives “restricted”
and “unrestricted” refer to their potential
for proliferation from buds (if any exist) at the
proximal nodes of their first order elements. Any
modified, or newly added, terms are indicated by a
double asterisk (”*) in the present glossary.
ABAXIAL (L. ab, away from; axis). Away from
an axis; designates that surface of an appendage
(such as a culm sheath or a sheath blade) which
faces, in a structural sense, away from the axis on
which it is inserted or relies for support. The terms
abaxial and adaxial (q.v.) should be employed
instead of upper and lower, or dorsal and ventral,
whose use-with reference to the two faces of a
foliage leaf or a reflexed sheath blade, for exampleis
apt to perpetrate ambiguity, and so cause confusion.
This is because the orientation of such a
structure may vary-in time, and from one botanical
entity to another-especially among the bamboos.
Therefore, neither upper and lower, nor dorsal and
ventral, will always be correlated reliably with
Glossary
122
anatomical dorsiventrality, to which important
structural basis all pertinent observations must be
oriented unmistakably in order to satisfy the requirements
of scientific discipline, dependability,
and clear thinking. Cf. ADAXIAL.
ABSCISSILE (L. abscissilis, from abscissus, cut off).
Susceptible of being cut off or disarticulated by the
formation of an abscission layer; applies to the
petiole of a deciduous leaf or, as in many bamboos,
to the segments of a rachilla. See DECIDUOUS.
ACROPETAL (Gr. ak~on., the highest point; L. peto,
I go toward). Progressing toward an apex, as when
the development or emergence of appendages on
any axis takes place serially in the direction of the
apex of the axis. Cf. BASIPETAL.
ACROPETALLY (adv.) Toward an apex.
ADAXIAL. Toward an axis; designates that surface
of an appendage which faces, in a structural
sense, toward the axis on which it is inserted, or
relies for support. Cf. ABAXIAL.
“ADNATE (L. adnascor, I grow to). “Attached
the whole length” (Jackson, 1949). “Congenitally
grown together; said especially of unlike parts”
(Webster, 1959). Used herein as applying to any
degree or extent of union between the tissues of
unlike parts. See CONNATE.
* ‘ADVESTITIOUS (L. adventitius). Occurring in
a location other than the usual one. In the bamboos,
and in other grasses as well, the principal
complement of roots is adventitious, arising at the
nodes of culms and rhizomes, and not from the
primordial root. In the genus Chusquea; for example,
adventitious branch buds arise independently,
both at the right and at the left of the one at the
usual site, which is median to the base of a subtending
culm sheath. See CONSTELLATE.
““AMPHIMORPH (Gr. amphi, both; morphe,
forms). Embracing both of two distinct forms.
Example: The same plant of Chusquea fendleyi
gives rise to culms from both pachymorph and leptomorph
rhizomes.
NUMBER 9 123
* *ANANTHOUS (L. anathus). Lacking a flower;
“transitional glumes,” and “sterile lemmas” are
ananthous.
(Gr. anastomGsis, a new outlet).
“Union of one vein with another, the connection
forming a reticulation” (Jackson, 1949).
ANDROECIUM (Gr. andros, male; oikos, house). The
stamen complement of a flower. I have not followed
Arber (1934:152 et passim) in adopting the spelling
(andraeceum) advocated by Kraus (1908). R.
W. Brown (1954), Jackson (1949), and Nybakken
(1959) all give androecium. Cf. GYNOECIUM.
* “ANTHECIUM (Gr. anthos, flower; oikos, house).
The protective structure in the Gramineae, which
is formed by the lemma and palea and which, typically,
encloses a single flower.
ANTICALLY, anticously (L., forward, in front).
Toward the front; introrsely or adaxially. Jackson
(1949) records diverse and even antonymous definitions
oi antical, mentioning the synonymous use of
introrse as “occasional.” The bamboo palea often
gapes antically.
* *APOCLADA (Gr. apo, separate; klados, branch).
A bamboo genus; the name alludes to the independent
origin of the primordial elements of midculm
branch complements, a feature that appears
in distinctive expression in this taxon.
In convex arcuate
array; characterizes the pattern of insertion of the
secondary component (second-order axes) of a monoclade
branch complement, as exemplified by
members of the genera Merostachys and Rhipidocladum,
qq. vv. See CONSTELLATE; GREMIAL; LEVEL;
RESTRICTED.
ASSIMILATION TISSUE. Mesophyll; the interior
ground tissue of a leaf blade.
AWNED. Bearing a bristlelike apical appendage
called an awn (OE). The empty glumes of the
spikelet in the type and some other species of
Neurolepis are awned. In Streptochaeta (a genus
of grasses related to bamboos), the lemma terminates
above in a very long, apically coiled structure
called an awn (L. arista) by Nees (1834), the author
of the genus; Trinius called it a tail (L. cauda).
The distinctions between the ways in which the
different types of sheathing structures or their foliar
appendages terminate apically are not sharp. Technical
terms (such as apiculate, awned, cuspidate,
and mucronate) and their conventional definitions
do not clearly distinguish the diverse expressions
ANASTOMOSIS
““APSIDATE (L. apsis, arch).
that are a part of the morphological intergradation
that baffles efforts at hard-and-fast categorization
here, as in many other facets of plant structure.
BAMBOO (bumbu, a vernacular word of undetermined
Oriental origin). The wood of bamboo
culms; a plant so classified.
BAMBOOS. A taxonomic group of plants comprising
the tribe Bambuseae of the Bambusoideae,
a subfamily of the Gramineae; living plants, or
culms (stems) severed from plants of this group.
BASIPETAL (L. basis, base; peto, I go toward).
Progressing toward a base, as when the development
or emergence of appendages on any axis takes
place serially in the direction of the base of the
axis. Cf. ACROPETAL.
BASIPETALLY (adv.). Toward a base.
* “BINATE (as applied to pleioclade midculm
branch complements). Typically consisting of only
two axes or branches, e.g., typical midculm branch
complement in the Old-world genus Phytlostachys.
BLADE (AS. blued, leaf). “The limb or expanded
portion of a leaf” (Jackson, 1949). In the bamboos,
a thin, expanded, chlorophyll-bearing, sessile or
petiolate, apical appendage (lamina) of a sheath
proper. See LEAF.
BOTTLEBRUSH (bottle brush, a brush of cylindrical
shape, with the bristles uniformly distributed on
the axis and oriented approximately at right angles
to it). A borrowed term, given a slightly modified
form, and used here (in the adjectival sense apparently
fiot recorded in dictionaries) to describe the
approximate pattern of distribution and orientation
of the proliferations of the stigmatic surface in certain
bamboos, a pattern loosely called plumose or
aspergilliform by some authors. Cf. DENDROID.
**BRACT (L. bructea, a thin plate of metal). A
foliar organ of the rachis of bamboo inflorescences.
Circumcingent bracts subtend branch buds and
branches of the rachis in all bamboo genera with
iterauctant inflorescences, but at one or more of the
basal nodes immediately above the prophyllum they
are usually empty. Bracts are usually rudimentary
or obsolete in most bamboo genera with semelauctant
inflorescences; in Glaziophyton, however, they
are well developed and circumcingent.
““BRANCH COMPLEMENT. The monoclade or
pleioclade array of primary branches (first-order
axes) that develops at a single midculm node. A
complete definition of each recognized type of
branch complement includes (1) the mode or pat124
SMITHSONIAN CONTRIBUTIONS TO BOTANY
tern of insertion of its primary (or secondary) components
or axes (see INSERTION); and (2) its
potential (restricted or unrestricted) for proliferation
from buds at the proximal nodes of its primary
(first-order) component (s).
BRANCH SHEATHS. The sheathing organs borne
singly at each node of an aerial vegetative branch
of any order (except the culm itself), excluding the
neck sheaths and the leaf sheaths.
“The nascent state of a flower
or branch” (Jackson, 1949). In the usage here
adopted, the term bud is applied only to those
primordial vegetative or reproductive branches that
(1) are enclosed in a prophyllum, and (2) have a
resting stage. Those (such as branch primordia in
determinate inflorescences generally; and root primordia,
for example) that lack either one or both
of these features are referred to simply as primordia.
Branch primordia in the determinate inflorescences
of Glaziophyton, Greslania, and some other genera
are prophyllate, but they apparently have no resting
stage.
**BUD, PRIMARY. A primary bud is one whose
initial primordium produces the primary (firstorder)
element (axis) in any series of branches of
ascending order. It appears that, in the members
of some bamboo genera a solitary primary bud may
incorporate more than one initial primordium, and
that each of these initial primordia normally produces
an independent first-order element of a pleioclade
branch complement. See PRIMORDIUM, INITIAL.
**CALLUS. A growth consisting internally of
iindifferentiated pith-like parenchyma that in some
bamboos forms an external ridge at the junction of
two structures (as, for example, at the locus of
insertion of a sheath upon its axis). Holttum (1958:
15, fig. 3) applies the term to what Hackel (1899:
716) calls the “exterior (outer) ligule” of the leaf
sheath, and what Pilger (in Diels 1937:57) calls a
collar. In that structure the tissue is usually of a
firmer nature, and is brought into prominence by
the abscission of the petiole, See CUPULE; GIRDLE;
and “outer ligule” under LICULE.
**CARYOPSIS (Gr. karyon, a nut; opsis, likeness).
“A one-celled, one-seeded, superior fruit, with a
pericarp united to the seed” (Jackson, 1949, as
cariopsis). A caryopsis is indehiscent, i.e., the pericarp
does not open to liberate the seed. In some
bamboos (Melocanna baccifera, for example) the
seed is free from the pericarp at maturity. The
BUD (ME. budde).
separability of the pericarp from the seed has been
much used by agrostologists as a taxonomic feature,
but mature caryopses of bamboos are rarely available
for comparison.
**CATAPHYLL (Gr. kata, down; phyllon, leaf).
“The early leaf forms of a plant or a shoot, as cotyledons,
bud-scales, rhizome scales, etc.” (Jackson,
1949). “Any rudimentary scalelike leaf which precedes
the foliage leaves, as a bud scale, etc.” (Webster,
1959). As herein used, the term designates the
sheaths that are incident at the lowermost nodes of
any vegetative axis of a bamboo plant. In cataphylls
the sheath proper is smaller in all dimensions than
it is in the sheaths borne at the more distal nodes
of the same axis, and its appendages are all rudimentary
or obsolete.
CENTRIFUGALLY (L. centrum, center; fugo, I flee
from). Outward from a center.
CENTRIPETALLY (L. centrum, center; peto, I go
toward). Toward a center.
**CESPITOSE (L. caespes, a sod; caespitosus, forming
a more or less compact clump or tuft). Describes
the normal clump habit of bamboos with
pachymorph rhizomes, except where the rhizome
neck Is very much elongated and the primary culms
proliferate by tillering or by way of short-necked
rhizomes as in the Old World bamboo currently
known as Yushania niitakayamensis. See DIFFUSE;
PLURICESPITOSE; UNICESPITOSE.
**CHARACTER (L.). “An instrument for marking,
character” (Webster, 1959). “The technical
difference whereby allied forms are distinguished, as
ordinal, generic, specific, and so on” (Jackson,
1949). “A member of mutually exclusive dates”
(Kendrick and Proctor, 1964:66). Any distinctive
form 01 behavior, or characteristic feature of demonstrrible
reliability (constancy) for purposes of
recognition or for the composition of a formal description
or for the construction of a key for the
differentiation of similar or related taxa. Constant
features serve in some cases alone as recognition
characters, in other cases as members of a set of
regularly associated constant features. Webster’s
definition of the term character, as used by biologists
(“Bot. & Zool. A feature or peculiarity (as of
structure, color, form or size) possessed by the individuals
of a variety, species or other category,
and by means of which they may be recognized or
differentiated; as a specific character, a generic character,
etc. An inherited character is regarded as the
NUMBER 9
observable effect of a gene or genes”) makes feature
a synonym of character. I differentiate these two
terms with the objective of promoting clear thinking
in relation to taxonomic procedures. It is my
position that a feature doesmot become a character
in the taxonomic sense until it can be shown to be
dependable for use in the differentiation of two similar
or related taxa. See FEATURE.
CHARTACEOUS (L. charta, paper; -aceus, made of,
or belonging to). Paperlike; papery.
CIRCUMAXIAL (L. circum, around; axis). Completely
around an axis; describes the typical reach
of the locus of insertion of a sheathing appendage,
or an assemblage of adventitious roots. Jackson’s
(1949) definition, under Circumaxile, is limited to
the special sense, “surrounding a central axis which
separates when the fruit splits open.”
“*CIRCUMCINGENT (L. circum, around; cingens,
girding). Characterizes a sheathing appendage (e.
g., a bract or a prophyllum) with respect to the
reach of its locus of insertion when this completely
encircles an axis.
CIRCUMNUTATION (L. circum, around; nutans,
wavering). The circular movement that commonly
takes place in the distal portion of a growing axis,
causing the apex to follow a spiral path as elongation
proceeds.
**COLLARET (Arber 1934:309). The girdle or
callus at the base of a leaf sheath. See CALLUS;
GIRDLE.
**CONGRUENT (L. congruens, conformable). In
agreement; consistent; harmonious. Applied to
herbarium specimens, of different but allied bamboo
species, which show a range of corresponding
characters that can be compared. Incongruent specimens
show different ranges of characters and thus
are sometimes useless as evidence of distinctions between
species.
CONNATE (L. connatus, born at the same time).
“United, congenitally or subsequently” (Jackson,
1949), in reference to the union of members of a
set of homologous parts (such as the filaments in
a stamen complement) that takes place in some
bamboos. See ADNATE; DIADELPHOUS; MONADELPHOUS;
TRIADELPHOUS.
CONNECTIVE (L. connectivum). A very narrow
strip of tissue upon the approximately opposite
sides of which the two parts of an anther (that is,
the two pairs of locules) are inserted longitudinally.
Jackson (1949) describes the connective as “distinct
from the filament,” of which it appears to be an
apical extension.
* “CONSTELLATE (L. constellatio, a cluster of fixed
stars; a constellation). Arranged in a constellationlike
cluster; characterizes (1) the pattern of insertion
of the primary components (first-order axes)
of a pleioclade branch complement as exemplified
by members of the genus Chusquea (where the
central element is several times as large as the
auxiliary ones that flank it), as well as where component
first-order elements are subequal, as in
Athrocstachys; and (2) the pattern of insertion of
the bud complement from which such a branch
complement arises. See APSIDATE; GREMIAL; LEVEL;
RESTRICTED.
CORNIFOKM (L. cornu, horn). Shaped like a horn.
CULM (L. culmus, stalk, stem). A segmented
aerial axis that emerges from a rhizome, and forms
a part of a gramineous plant; the term is used most
commonly with special reference to bamboos. Syn.:
halm, haulm haum.
“The ‘knot’ of a
grass stem” (Jackson, 1949). The term node is generally
applied in a loose, comprehensive sense to
that complex locus-the junction of adjacent internodes
in a segmented axis of a gramineous plant.
Hackel (in Lamson-Scribner and Southworth, 1890:
2, 3) stressed the importance of the “constantly overlooked”
difference between the culm nodes and the
sheath nodes in non-bambusoid grasses. However,
neither Hackel’s description nor his illustration
does justice to the distinctive features of these structures
as they appear in the bamboos. Culm node is
here defined with special reference to that level
within the node (sensu lato) where secondary elongation
(intercalary growth) takes place, and where
branch buds and adventitious roots are inserted.
Each culm node (sensu stricto) is located just above
a sheath node (q.v.), from which it is usually distinguishable
by the transverse thickening or ridge
(the “supranodal ridge” of some authors) that appears
at the level of insertion of a bud or a branch
comp!ement. Branch node and rhizome node should
be similarly differentiated, respectively, from the
corresponding sheath nodes. See SHEATH NODE.
One of the sheathing organs
borne singly at each node of a bamboo culm, below
the level at which the sheaths of foliage leaves take
their place. Kurz (1876:268) anticipates this usage
with “halm-sheath.” The transition between culm
CULM NODE (L. nodus, knot).
**CULM SHEATH.
126 SMITHSONIAN CONTRIBUTIONS TO BOTANY
sheaths and foliage leaves may be rather abrupt or
very gradual.
CULM SHOOT. Bamboo shoot; a young culm in
any stage of its development short of maturity in
height. See SHOOT.
* “CUPULE. The cup-shaped depression that
marks the locus of insertion of the petiole upon
the leaf sheath (McClure, 1941:33-34). The cupule
is apparent, of course, only after the leaf has fallen
away. See CALLUS; LIGULE.
* *CUSPIDATE (L. cuspidatus, from cuspis, a point,
especially the point or head of a spear [Marchant
and Charles, 19581). “Tipped with a cusp” (Jackson,
1949). Terminating in a small, hard point;
used herein in this sense to characterize the apex
in bamboo caryopses where the indurated conical
base of a slender style persists at maturity.
**CYMBIFORM (L. cymbiformis; cyrnba, boat).
Boat-shaped, like the lemma in many bamboos, especially
when it is keeled.
“DECIDUOUS (L. deciduus). Abscissile; falling
away; “falling in season” (Jackson, 1949). The term
applied especially to organs (the culm sheaths in
some bamboos) or parts that become cut off at or
near the locus of their insertion by development of
a special layer of cells (an absciss layer), or by the
formation of a cleavage plane along a well-defined
line of separation. In many bamboos several of the
distal internodes of leafy twigs (as well as the individual
leaves) are deciduous. They fall away after
the leaf blades have fallen, upon the formation of
a transverse rift at a node-usually the most distal
node bearing a bud-this bud having already produced
a new leafy twig. See PERSISTENT.
DECLINED (L. declinatus, turned aside). “Directed
obliquely” (Jackson, 1949); describes the habit
or orientation of the culms in some species of
bamboos.
DELIQUESCENT (L. deliquescens, melting away,
disappearing). Becoming dispersed or expended,
as when the main axis of any structure (the rachis
of an inflorescence of Aulonernia quexo, for example)
loses its identity among its own branches, or
“loses itself by repeated branchings” (Jackson,
1949). See EXCURRENT.
DENDROID (Gr. dendron, tree, -oid, having the
form of). Treelike; bearing branches of more than
one order; describes the pattern of proliferation of
the stigmatic surface in certain bamboos.
*#DEPAUPERATE (L. depauperatus, impoverished).
Reduced in size or functional efficiency, or
both; the term characterizes whole plants, or individual
parts or organs thereof such as florets, pseudospikelets,
whose development is impeded or restricted
by adverse conditions, external or internal.
DERMOGRAMME (Gr. derrna, skin; gramrna, something
drawn). A French term used by Prat (1936:
178) ta designate any example of his diagrammatic
illustrations of the cellular details of the structure
of the epidermis of any grass or bamboo. See
SPODOGRAM.
DETERMINATE (L.determinatus, having fixed limits).
Of limited growth; characterizes an axis, or a
system of immediately related axes (a determinate
inflorescence, for example), whose development or
potential for development is confined to a single,
definitely limited “grand period of growth,” after
which no meristem usually persists. The occasional
persistence of meristem at the tip of individual
rachillas sometimes prolongs the grand period of
growth in a determinate inflorescence atypically. A
rachilla may then continue to produce new florets
until the spikelet is greatly elongated-to as much
as a foot, while its normal length would be 3 to 4
inches. Again, this meristem may revert to the
vegetative state, prolificating to produce a leafy
axis. Both of these teratic phenomena have been
observed in Arundinaria dolichantha. See PROLIFERATION;
SEMELAUCTANT. Cf. INDETERMINATE.
DIADELPHOUS (Gr. dis, twice; adelphos, brother).
Having “two groups of stamens”; cognate with Diadelphia,
the name of “a Linnean class having the
stameiis in two bundles of brotherhoods” (Jackson,
1949). Diadelphous describes the condition of a
stamen complement of a flower when its members
have their filaments connate to form two distinct
groups. In the bamboos, the relative numbers of
stamens in the respective groups may vary from
one genus to another. Cf. MONADELPHOUS; TRIADIAGEOTROPIC.
Adjectival form derived from
diageotropism, which is defined under GEOTROPISM.
DIAPHRAGM (Gr. diaphragma, a partition wall).
The transverse internal layer of parenchyma found
at the level of every sheath node. It is reinforced by
the crossing over and anastomosis of vascular bundles,
and forms a rigid structural element that lends
strength (mechanical resilience) to the segmented
vegetative axes of bamboos. “A dividing membrane
or partition” (Jackson, 1949).
DELPHOUS.
NUMBER 9 127
““DIFFUSE (L. diflusus, spread out). Growing in
open array; characterizes the normal mature clump
habit of most bamboos with leptomorph rhizomes,
and also those whose pachymorph rhizomes have a
greatly elongated neck and solitary culms, e. g.,
Bambzisa sp. (McClure 21438-A) of Peru. (See
McClure, 196Gb:27, fig. 10:5).
DISTAL (L. disto, I stand apart). Remote; far
out; designates loci of insertion, or structures, situated
at or near the tip of an axis. Cf. PROXIMAL.
DISTICHOUS (L. distichus, consisting of two rows).
“Disposed in two vertical ranks, as the florets in
many grasses” (Jackson, 1949). Distichous also describes
the pattern of insertion of buds and
branches, and the sheathing appendages on all segmented
axes, both vegetative and reproductive, of
the bamboo plant. As tacitly recognized by agrostologists,
but generally not mentioned in published
definitions, the two ranks are normally inserted on
opposite sides of an axis but are sometimes secund.
DISTICHY (substantive of L. distichus). Arber’s
(1934) word for the two-ranked arrangement of
buds, branches, and sheathing appendages generally
characteristic of segmented axes in the Gramineae.
See DISTICHOUS.
““DOMINANT (L. dominans). Superior in size
among several associated objects of a class or kind,
as the primary axis among members of a pleioclade
midculm branch complement in Bambusa vulgaris,
or the trees that overtop all others in a forest.
* “EMBRYOTEGIUM (Gr. embryon, a foetus, tega,
a covering). The embryotega of Stearn (1966).
Literally, embryo cover; the “little shield” (Sp.
escudete) of Parodi (1961); to be distinguished from
the scutellum, which is a part of the embryo itself.
“A callosity in the seed coat of some seeds near
the hilum, and detached by the protrusion of the
radicle on germination” (Jackson, 1949). Since
Jackson’s definition is based on the nature of the
embryotegium in the seed proper, it is not strictly
applicable to its nature in a caryopsis. Hackel (in
Lamson-Scribner and Southworth, 1890: 20, fig. 6)
describes and illustrates the embryotegium in a
caryopsis (without, however, giving it a technical
name) as “a place where the embryo lies covered
only by the pericarp, and plainly visible on the outside.’’
The position of the embryotegium is clearly
visible in caryopses with a thin pericarp as in bamboos
of certain genera such as Bambusa and Arundinaria.
In some genera, including Meiocanna (Stapf,
1904a, pl. 45: fig. 3, pl. 47: fig. 40), and Ochlandra,
where the pericarp is thick, the embryotegium is
ordinarily not externally recognizable (cf. Figure
““EMPTY GLUMES. Revision of McClure’s
(1966a:ZZj) definition. See TRANSITIONAL GLUMES.
* “ENDOSPERM (Gr. endo, within; sperma, seed).
“The nutritive tissue formed within the embryo sac”
[of a seed] (Webster, 1959)-referred to by earlier
writers as “albumen” (Jackson, 1949). In some
bamboos the embryo consumes the endosperm as
fast as it is formed (Stapf, 1904a). The stored endosperm
in some bamboos (Arundinaria gigantea) is
granular and starchy, in others (Apoclada simplex)
it appears to be waxy or glutinous, not granular.
EXCURRENT (L. excurrens, running out). Extending
through, as when the main axis of any
structure (the rachis of the paniculate inflorescence
in Indocalumus sinicus, for example) maintains its
identity among its own branches; “where the stern
remains central, the other parts being regularly
disposed round it” (Jackson, 1949). Cf. DELIQUES-
““FACULTATIVE (Fr. facultatif, from L. facultas,
capability). “Of such a character to admit of existing
under various forms or conditions; of happening
or not happening, etc.” (Webster, 1959).
“Occasional, incidental, as opposed to obligate”
(Jackson, 1949). Describes the freedom, or potential
faculty, to give expression to one or the other of
two or more states, forms, products or features;
possessed by (1) the proximal buds of the primary
branch in certain bamboos with respect to germination
or continued dormancy (Aulonemia effuso)
(see Key, p. 8); or (2) the apical meristem of
the developing axes in certain bamboos (Arundinaria
gigantea ssp. tecta; Semiarundinaria fastuosa)
to produce either a culm or a rhizome.
““FEATURE (ME. feture; L. factura). “The . . .
form or outward appearance of anything” (selected
from Webster, 1959). In conventional discourse
(and in technical papers as well) the term “feature”
is often used in the sense here given “structure”
(e.g., eyes, nose, mouth, chin-individual component
structures of a countenance). In the present
work the technical meaning of feature is restricted
to characteristic aspects (such as form, size, number,
color, behavior) of various structures or parts thereof.
It is my position that a feature does not become
a cliaracter in the taxonomic sense until it can be
37 0).
CENT.
128 SMITHSONIAN CONTRIBUTIONS TO BOTANY
shown to be invariable (dependable) and either
unique to a given taxon or a member of a set of
contrasting features that distinguish two similar or
related taxa. Cf. CHARACTER.
Numerous,
slender, and not conspicuously tapered; the meaning
of the term as conventionally used to describe
the gross aspect of the roots of bamboos and other
grasses, a definition apparently not recorded in dictionaries
or glossaries. Jackson’s (1949) definition
of the term is not applicable here.
FISTULOSE (L. fistula, a pipe). Hollow; having
a lumen; characterizes the internodes of culms and
branches in most bamboos.
FLORET (dim. OFr. flor, flower). One of the units
into which a spikelet breaks up when the rachilla
segments disarticulate. Regardless of whether the
rachilla segments are abscissile or not, a floret consists
of (1) a segment of the rachilla, (2) the lemma
that is inserted upon it, (3) a branch (the axis of
a flower) subtended by the lemma, (4) a prophyllum
(the palea) of the axis of the flower, and (5)
the parts of the flower that are inserted on its axis.
Attention is directed especially to the fact that the
gramineous floret includes structures from axes of
two oi-ders while the flower is confined to a single
axis. The distinction between floret and flower is
not always observed in published works on agrostology.
The loose use of the two terms interchangeably
is to be avoided. Cf. FLOWER.
FLOWER (OFr. fEor). That portion of a branch of
the rachilla that is distal to its own prophyllum
(the palea), together with the reproductive organs
(androecium or gynoecium or both) borne by it.
The lodicules, when present, are included in this
concept of the flower, but the palea is not. Cf.
FLORET.
A leafy or leafless segmented
axis that bears one or more inflorescenses. A flowering
branch is distinguishable from an inflorescence
proper in that it retains all of the morphological
characteristics of the vegetative state, with the
sole exception that in some cases it does not produce
foliage leaves. The inflorescence proper, on
the other hand, always has morphological features
peculiar to it that are not found on the flowering
branch proper. In describing and illustrating bamboos,
some authors fail to distinguish the two. Arber
(1934: 108) characterizes a leafless flowering bamboo
FIBROUS (L. fibra, thread, filament).
FLOWERING BRANCH.
as a “truly gargantuan inflorescence.” Cf. INFLORESCENCE.
FOLIAR ORGAN. See SHEATHING ORGAN.
““FREE (ME. fre). “Not adhering, the reverse
of adnate” (Jackson, 1949). Here given the sense of
non-connate, in reference to the filiform filaments
of the stamens in certain bamboos.
FUGACEOUS (L. fugax, fleeting). Promptly deciduous;
“soon perishing” (Jackson, 1949).
FUSIFORM (L. fusus, spindle). Spindle-shaped;
circular in cross-section, thickest in the middle and
tapering toward each end. See SUBFUSIFORM.
GENICULATE (L. geniculatus, with bended knee).
“Bent abruptly at an angle, like the bent knee”
(Webster, 1959). When a growing culm adjusts the
direction of its orientation (as in situations where
negative geotropism operates to effect the restoration
of a fallen or deflected culm to an upright
postuie), the adjustment takes place through differential
elongation in the upper and lower portions,
respectively, of the zone of intercalary growth
immediately above a node, and the node involved
becomes geniculate.
GEOTROPISM (Gr. ge, the earth; plus tropism,
defined by Webster (1959) as “the innate tendency
of an organism to react in a definite manner to
external stimuli”). A physiological potential, and
the consequent act, involved in the assumption and
maintenance of a particular direction of growth by
an axis or organ with reference to the force of
gravity, or to the centrifugal force generated by
rapid circular motion. As commonly used, geotropism
ambiguously embraces (and confuses) two
component phenomena: the visible physical response,
and the more fundamental, invisible
response, which is physiological. Jackson’s (1949)
definition: “the force of gravity as shown by curvature
in nascent organs of plants” is incomplete.
Three distinct classes of geotropic responses are
recognized: positive, negative, and transverse. The
last named is called diageotropism.
**GERMINATION (L. germen, an offshoot, sprout,
bud). A term usually restricted to the beginning of
growth of the embryo of a seed; here extended to
the beginning of growth of a temporarily dormant
axillary bud on a bamboo culm.
*“GIRDLE (AS. gyrdel, belt: Latin equivalent,
cingulum). A conspicuous horizontal band of special
tissue occupying a position basal to the culm
sheath proper in some bamboos (e.g., Melocalamus
NUMBER 9 129
compactiflorus and some scandent species of Chusquea).
It is expansible in all directions as long as
the corresponding intercalary zone of the embraced
internode is in active growth, and it takes on an
asymmetrical form at geniculate nodes. Upon the
abscission of the sheath, this band remains attached
at the culm nodes. The distinction between a girdle
and a sheath callus (q.v.) is not always clear, being
based principally upon their relative vertical dimension
(area). Cf. CALLUS; SHEATH CALLUS.
**GLUME (L. gluma, a hull or husk). “The
chaffy, two-ranked members of the inflorescence of
grasses and similar plants” (Jackson, 1949). Excepting
prophylla, this very general definition ostensibly
embraces all of the sheathing appendages of
the inflorescence, including the bracts that, in some
cases, subtend its branches. As shown elsewhere, the
application (meaning and usage) of the terms
lemma and palea and bract has become rather
generally standardized. However, in those cases
where it is difficult to distinguish empty glumes and
sterile lemmas from each other, I have propose‘d to
cover both categories with the comprehensive term
transitional glumes, q.v.
GREGARIOUS (L. gregarius, of or pertaining to a
flock or herd). “Growing in company” (Jackson,
1949). Gamble (1896:viii) makes gregarious mean
simultaneous, to describe the flowering behavior
manifested when all of the members of a given
generation of bamboo plants (from seeds of a common
origin) enter the reproductive state at approximatelv
the same time. Cf. SPORADIC.
**GREMIAL (L. gremialis, of a lap or bosom).
“Growing in a pollard-like cluster” (R. W. Brown,
1954); characterizes the mode of insertion of a tuft
of subequal second-order axes that develop from
buds OR a decurrent extension of the primary (firstorder)
axis of an unrestricted monoclade branch
complement (as in Bambusa swalleniana [Guadua
spinosa] and in numerous species of the genus
Aithrostylidium), Cf. PROMONTORY.
GYNOECIUM (Gr. gynb, female; oikos, house).
The complete pistil, consisting of the ovary, the
ovule (s), style (s), and the stigma (s), of a single
flower. I have not followed Arber (1934:120 et
passim) in adopting the spelling (gynaeceum) advocated
by Kraus (1908). Jackson (1949) and Nybakken
(1959) both give gynoecium. Cf. ANDROECIUM.
HALM, haulm, haum (terms used in British agrostology).
See CULM.
**HERMAPHRODITE. Same as PERFECT, q.v.
**HILUM (L. hilum, a trifle). “The scar left on
a seed where formerly attached to the funicle or
placenta” (Jackson, 1949). This definition confines
the connotation of the term hilum to a scar on a
deciduous seed that has been freed from its attachment
on a dehiscent carpel. However, in the
literature of agrostology, the term hilum is quite
generally applied to the longitudinal furrow or
groove that marks, on the exterior of the indehiscent
gramineous caryopsis, the locus of attachment
of the indeciduous seed that remains confined within
an indehiscent pericarp. In the present treatise,
this external furrow is referred to as a sulcus. See
SULCATE.
IMBRICATE (L. imbricatus, covered with tiles).
“Overlapping like roofing tiles and shingles” (R.
W. Brown, 1954). Sheathing organs at successive
nodes of an axis are imbricate when each one exceeds,
in length, the internode it embraces.
INDETERMINATE (L. indeterminutus, unlimited).
“Not terminated absolutely” (Jackson, 1949). Continuing
apical growth characterizes an individual
axis as indeterminate. The term is here given another
connotation (namely, iterauctant) with particular
reference to the continuing ramification of
a system of immediately related axes, as exemplified
by the indeterminate inflorescences of certain bamboos.
In such infloresences, the buds at the proximal
nodes of each new order of branches bear fresh
bodies of meristem. The development of new
branches from these buds may be continuous (or
intermittent and reactivated on a seasonal basis)
for a period lasting in some cases for several successive
years. Each branch of an indeterminate inflorescence
(an axis with its appendages) has the
appearance of a spikelet. Because of this, it is
called a pseudospikelet. Each pseudospikelet has
its own independent grand period of growth. The
active period of the meristem at the apex of the
rachilla that terminates each pseudospikelet is
limited. Thus, while the branching of indeterminate
bamboo inflorescences is of a continuing
nature, each individual branch is apically determinate
in its growth. See ITERAUCTANT; PSEUDOSPIKELET;
cf. DETERMINATE.
**INFLORESCENCE. A discrete aggregation of spikelets
associated with a common primary rachis or a
common peduncle. Exceptionally, an inflorescence
may comprise but a single spikelet. I know of no
130 SMITHSOXIAN CONTRIBUTIONS TO BOTANY
exception to the rule that the inflorescence in a
given bamboo genus is uniformly either semelauctant
or iterauctant (q.v.) in its manner of development.
The occurrence of individual features of an
intermediate (incongruous) nature is rare. Cf.
FLOWERING BRANCH; see also INFLORESCENCE, ITERAUCTANT;
INFLORESCENCE, SEMELAUCTANT.
* *INFLORESCENCE, ITERAUCTANT. An inflorescence
ir, which typically both the initiation and the
termination of growth takes place in serial sequence
in the flowering axes of successively higher orders,
each branch possessing residual meristematic tissue
in the form of basal branch primordia (buds) that
continue the expansion of the inflorescence as long
as physiological conditions in the adjacent tissues
are favorable to their germination and growth.
**INFLORESCENCE OF MIXED BRANCHING. An inflorescence
with peduncles proliferating (duplicating
themselves indefinitely from nascent or dormant
prophyllate branch buds) the rachises of determinate
branching (without nascent or dormant
branch buds) and with the development of all
spikelets simultaneous or nearly so (e.g., the Old
World genus Chimonobambusa).
* “INFLORESCENCE, SEMELAUCTANT. An inflorescence
in which typically the initiation of growth is
simultaneous or nearly so in all axes, and the termination
of growth is likewise simultaneous or nearly
so; and no meristematic tissue remains afterwards
in the form of dormant buds.
#*IKSERTION (L. insertio, a putting in). The
mode, pattern, or locus of attachment of an organ
or a part. With a view to characterizing the several
patterns of insertion herein distinguished with respect
to the component first- (or second-) order axes
of a midculm branch complement, the following
terms have been selected and defined: APSIDATE,
CONSTELLATE, GREMIAL, LEVEL, and RESTRICTED.
* WITERAUCTANT (L. iterum, again; auctans, growing).
Embracing more than one grand period of
growth, as the mode of development of the successive
orders of the branches of an inflorescence composed
(as in bamboos of the genera Bambusa,
Elytrostachys, and Atractantha) of pseudospikelets
(q.v.). Cf. SEMELAUCTANT; INDETERMINATE.
**LEAF (AS). Foliage leaf; bamboo leaf; the
chlorophyll-bearing, usually petiolate, blade of a
leaf sheath proper. This concept of leaf, in the
sense bamboo leaf, includes the petiole when it is
present, but excludes the leaf sheath proper. Leaf
is here defined in this circumscribed sense in order
to avoid a commonly encountered ambiguity occasioned
by its indiscriminate use in the literature.
The word is commonly used by agrostologists, without
modification or qualification, to refer to any
one of the morphologically, anatomically, and functionally
divergent forms of sheathing organs borne
on the several kinds of vegetative axes of a bamboo
plant. This imprecise use of the term, in disregard
of the importance of the manifest differences thus
loosely covered by it, apparently is due to the
generalized and unqualified nature of the definition
of “leaf” given by Jackson (1949): “the principal
appendage or lateral organ borne by the stem or
axis.” The use of the term “leaf” for the petiolate
leaf blade as distinct from the leaf sheath in bamboos
h2s ample precedent in Munro’s “folia” which
is described separately from “vagina.” I think that
we may rationalize this usage by postulating that,
in the grasses, the sheath is a distinct organ or
structure inserted between the leaf proper and the
axis frcm which it emerges (or on which it is inserted).
We may continue with the postulate that
the girdle (where it appears) is a structure inserted
between the sheath proper and the sheath node.
Cf. SHEATH; LEAF SHEATH; see also SHEATHING OROne
of the leaf-bearing sheaths
inserted at the distal nodes of each aerial segmented
axis whether culm, branch, or twig. The conspicuous
part of a leaf sheath is the usually petiolate
blade, inserted at the apex of the sheath, just below
the ligule. Cf. LEAF; SHEATH.
**LEMMA (Gr., a husk). The glume that subtends
n flower. Lemmas that subtend normally constituted
and developed flowers are conventionally
called fertile lemmas, to distinguish them from
sterile lemmas, whose subtended flowers are depauperate
or obsolete. See TRANSITIONAL GLUMEs.
**LEPTOMORPH (Gr,, Zeptos, slender, and
morphe, form). Characterizes the rhizomes proper
(of a bamboo plant) when they are long and slender,
and have the following associated characteristics:
a cylindrical or subcylindrical form, with a
diameter typically less than that of the culms
originating from it; internodes longer than broad,
relatively uniform in length, symmetrical or nearly
so, rarely solid, typically fistulose, the usually narrow
central lumen interrupted at each node by a
diaphragm; nodes in some genera usually somewhat
GAN.
**LEAF SHEATH.
NUMBER 9 131
elevated or inflated, in others not; lateral buds in
the dormant state boat-shaped, with a distally
oriented apex. Cf. PACHYMORPH.
**LEVEL (L. libella, a water level or plumb level).
Even, horizontal. Characterizes the pattern of insertion
of the prima_ry components (first order
axes) of a pleioclade branch complement, as exemplified
by members of the genus Apoclada.
LIGULATE. Having a ligule.
LICULE (L. Zigula, a little tongue). A thin, apical
extension of a sheath proper, adaxial to the
locus of insertion of the sheath blade (or leaf
petiole). The outer rim of the little cup revealed
by the abscission of the petiole of a foliage leaf is
sometimes referred to as the “outer ligule,” especially
when it is conspicuously developed, as it is
in some variants of Arundinaria gigantea ssp. tecta.
The term “inner ligule” is then applied to the
adaxial structure conventionally known simply as
the ligule. The ligule is reduced to a mere line, or
is even lacking entirely, in the culm sheaths of
certain species of bamboos.
Locus (L., place). The locale in which some
event takes place, or where something is to be
found. Locus is here given the connotation, point,
or line, or area, principally in the phrase “locus o€
insertion.” See INSERTION.
One of
the small, usually thin, delicate and transparent
structures (by some authors referred to as “scales”)
inserted usually in a single whorl of 3, immediately
below the stamens in the bamboo flower. The
lodicules are relatively large, opaque, and parchment-
like in the flowers of known species of
Streptochaeta. Their number is variable in some
bomboos (up to 12 or more in OchZandra stridula);
in known species of Gigantochloa, typical lodicules
are lacking entirely.
LUMEN (L., opening). “The space which is
bounded by the walls of an organ, as the central
cavity of a cell” (Jackson, 1949); the central cavity
of a hollow internode of any segmented axis of a
bambco plant.
**MAVERICK. Nonconformist (see Webster,
1959). A term borrowed here (see page 15 et passim)
for the purpose of characterizing ontogenetic
and/or morphological features that appear to range
alone, as if moving in independent trajectories.
Maverick features appear at random as discrepancies
(phenotypic diversities) among the other fea-
LODICULE (L. lodicula, a small coverlet).
tures of one or more species of different genera
that are otherwise more or less distantly related,
or at least phenotypically disjunct. I was at first
inclined to credit their presence in such discrepant
settings as due to recent introgressions of unlinked
genes. More recently I have considered the possibility
that they come from a common ancestor by
way of recessive genes.
‘“MBG. McClure Bambo Garden, Bethesda,
Maryland, U.S.A. See also P.I.
MERISTEM (Gr. meristos, divisible). A body of
tissue in which cell division and differentiation are
active or potential.
METAMORPH (Gr. meta, implying change; morpht?,
form). Of intermediate form; a term proposed here
to designate certain underground portions of segmented
axes, whose transitional character cannot
be clearly indicated by the use of existing conventional
terms. See METAMORPH I; METAMORPH 11
A transitional axis that occupies
a position between the culm neck and the base of
the culm proper, where no clearly defined rhizome
intervenes; in some cases doubtfully distinct from
an elongated culm neck.
A transitional axis intermediate
in form and position between the apex of a rhizome
(either pachymorph or leptomorph) and the culm
into which the rhizome is transformed apically. It
appears where the transformation of the apex of
a rhizome into a culm takes place gradually and not
abruptly.
METAMORPHOLOGICAL (Gr. meta, beyond, plus
morphological). Pertaining to the techniques, the
areas of investigation, or the data of disciplines
other than morphology.
MONADELPHOUS (Gr. monos, one; adeZphos,
brother). United in a single brotherhood; an adjective
cognate with Monadelphia, the name of
“a Linnean class in which the anthers are united
by their filaments into a single brotherhood” (Jackson,
1949), and conventionally used to describe the
stamen complement of a flower when all of its
members are united by their connate filaments.
Cf. DIADELPHOUS; TRIADELPHOUS.
METAMORPH I.
METAMORPH 11.
MONOCARPIC. See MONOPERIODIC.
““MONOCLADE (Gr. monos, one; klados, branch).
Characterizes those branch complements that
contain only one primary axis. See BRANCH
COMPLEMENT; MONOCLADE, RESTRICTED; MONOCLADE,
UNRESTRICTED. Cf. PLEIOCLADE.
132 SMITHSONIAN CONTRIBUTIONS TO BOTANY
* *MONOCLADE, RESTRICTED. Characterizes those
monoclade branch complements whose primary
axis is incapable of proliferating basally, for lack
of buds at its proximal nodes. When buds are present
at the proximal nodes of the solitary axis of a
monoclade branch complement but remain dormant,
such a branch complement is referred to
herein as facultative restricted monoclade.
““MONOCLADE, UNRESTRICTED. Characterizes those
monoclade branch complements whose solitary primary
axis is capable of proliferating basally from
buds at its proximal nodes.
MONOPERIODIC (Gr. monos, one; periodos, a completed
course). Having but a single reproductive
cycle within the lifetime of a plant; characterizes
bamboos (such as Bambusa arundinacea) that
flower but once, then perish. Monocyclic suggests
itself ior use in the sense here given the newly
coined term, monoperiodic, but it already bears
the disqualifying connotation annual, with reference
to the life span of a plant-besides indicating sets
of parts (such as sepals and petals) that comprise
but a single whorl or cycle. In the interest of
clarity, the term monocarpic, commonly used to
indicate the monoperiodic character of certain
bambcos, should be reserved (with its orthographic
variants; see Jackson, 1949) for reference to a pistil
comprising but a single carpel. Cf. POLYPERIODIC.
MONOPODIAL (Gr. monos, one; pous, podos, foot).
Having the form of a monopodium, which is defined
by Jackson (1949) as “a stem of a single, and
continuous axis.” This term was used earlier (Mc-
Clure, 1925) to designate the type of rhizome
described as leptomorph (q.v.). Cf. SYMPODIAL.
MORPHOGENESIS (Gr. morphd, form; genesis, beginning
or origin). “The production of morphological
characters” (Jackson, 1949). As defined
here, morphogenesis connotes not only its external
manifestations but also, and primarily, the nature
and sequence of events in the program of genic
monitoring of physiological states associated with
the emergence of the structural features that determine
the characteristic form of a plant.
MUCRONATE (L. mucronatus, pointed). “Possessing
a short, straight point, as some leaves”
(Jackson, 1949). Cf. CUSPIDATE; MVTICOVS.
“Blunt,* awnless’’
(Jackson, 1949); used primarily (when applicable)
to describe the apex in small sheathing ap-
MurIcous (L. muticus, docked).
pendages on the various axes of the bamboo in-
Horescence.
The constricted basal part,
characteristic of all, or most, of the segmented
vegetative axes of a bamboo plant.
NECK SHEATH. One of the reduced, bladeless
sheaths that clothe the constricted proximal part,
the neck, of various vegetative axes in the bamboo
plant. Neck sheaths are sometimes referred to as
CATAPHYLLS.
” *NOTHOMORPH (Gr. nothos, hybrid; morphos,
form). “Applied to any hybrid form, whether F1,
segregate or back-cross“ (International Code of
Botanical Nomenclature, Article H.5). Cf. Figures
17 and 18.
OBSOLETE (L. obsoletus, worn out). “Wanting
or rudimentary; used of an organ which is scarcely
apparent or has vanished” (Jackson, 1949).
OBTERETE. Circular in cross-section, tapering
progressively from one end to the other, and
smallest at the proximal end (for example, a
rhizome neck). Cf. TERETE.
“Having the solid figure or lengthwise
outline of an egg; popularly elliptical or ellipsoidal.
Bot. broadly elliptical” (Webster, 1959). “A body
or figure in the shape of the long section of an egg,
or popularly an ellipse” (Webster, 1959).
“Shaped like an egg; oval” (Webster,
1959). “With an outline like that of a hen’s egg cut
in two lengthwise, the broader end downward”
(Bailey, 1914; Jackson, 1949).
**OVOID AND OVATE. The terms “ovoid” and
“ovate” are used in diverse senses. The original
botanical sense of ovate appears to be [that of]
Bailey (1925) and Jackson (1949) but it has (under
popular influence) drifted to mean elliptic or ellipsoid
(Stearn, 1966; Webster, 1959).
Abbreviation of “Plant Introduction,” prefixed
to a permanent identifying number assigned
by the U.S. Department of Agriculture to each lot
of living plant material (seeds, plants, or cuttings)
accessioned in its record of plant introductions.
These numbers are published, along with pertinent
names and documenting notes, in a continuing
series of “Plant Inventories.” See also MBG.
**PACHYMORPH (Gr. pachys, thick; morphd,
form). Term proposed by McClure to describe
the rhizome proper (of a bamboo plant) when it is
short and thick, and has the following associated
characteristics: a subfusiform (rarely subspherical),
NECK (AS. hnecca).
““OVAL.
““OVATE.
P.I.
NUMBER 9
usually more or less curved (rarely straight) shape,
with a maximum thickness typically somewhat
greater than that of the culm into which it is always
transformed, apically; internodes broader than long,
asymmetrical (longer on the side that bears a bud),
solid (apparently never fistulose); nodes not elevated
or inflated; lateral buds solitary, in the
dormant state asymmetrically dome-shaped, with a
subcircular margin and an intramarginal apex.
Cf. LEPTOMORPH.
PALEA (L., [a piece of] chaff). The prophyllum
of the axis of a gramineous flower. Jackson’s (1949)
definition: “the inner bract or glume in grasses,
called ‘Palet’ by North American writers” is too
vague to be useful. See PROPHYLLUM.
PANICLE (L. pun.icuZu, a tuft). A determinate inflorescence
with branches of more than one order.
Jackson’s definition of panicle is not applicable
here.
* ‘PANICULATE ( L . paniculutus, tufted). Inflorescences
with branches of more than one order, the
rachis excurrent or deliquescent. See PANICLE.
**PARAMO (Sp., from L. purumus, teste Larousse,
in AugP: and Gisbert, 1949:697). “Extensive treeless
regions that dominate the summits of the Andean
cordillera from the upper limit of the forestan
elevation of 3800 m (locally 3220 m)-up to the
level of permanent snow (4700 m);” teste Cuatrecasas
(1959:249) where distinctive ecological, vegetational,
and floristic features of the piramo are
described at length-in Spanish.
PARENCHYMA (Gr. parenchein, to pour in beside).
Fundamental tissue; ground tissue; a tissue, such as
pith, composed of thin-walled, undifferentiated,
isodiametric cells (adapted, from Jackson, 1949).
PEDICEL (L. pedicellus, dim. of pes, foot). The
stalk of a spikelet, that is, the distal segment or
internode of the axis immediately below the glumes
that marks the base of a spikelet. The ultimate
branches of a determinate inflorescence are all pedicels,
since each terminates in a spikelet. In spicate
racemes, and in pseudospikelets, the pedicels are
usually very short. Cf. PEDUNCLE; RACHIS.
PEDICELLATE. Having a pedicel.
PEDUNCLE (L. pedunculus, dim. of pes, foot).
The stalk of ah inflorescense, that is, the unbranched
segment of the inflorescence axis that is
immediately below the rachis. In determinate inflorescences,
the first sheathbearing node below the
first branch of the inflorescence is here taken arbitrarily
as marking the base of the peduncle, and the
locus of insertion of the first branch of the inflorescence
as marking the apex. In those rare cases
where a determinate inflorescence includes only
one spikelet, the peduncle is not distinguishable
from the pedicel. In indeterminately branching
inflorescences, the peduncle is usually very short.
An exception appears where the primary pseudospikelet
is terminal (instead of lateral) to a flowering
branch. Here the peduncle is more elongate.
Cf. PEDICEL; RACHIS.
PEDUNCULATE. Having a peduncle.
* “PERCURRENT (L. percurrens, passing through).
Used here in a sense identical with that given
EXCURRENT.
**PERFECT. Signifies the presence, within an
anthecium, of a flower containing a normally developed
complement of sexual elements, both
staminate and pistillate. Same as hermaphrodite.
PERCAMINEOUS (L. pergamenu, parchment).
“Like parchment in texture” (Jackson, 1949).
PERSISTENT (L. persistens, remaining in place).
Not deciduous; applies to organs that remain in
place after they have fulfilled their natural functions
(for example, the culm sheaths of Arundinaria
gigantea ssp. tecta).
PINNATE (L. pinnutus, feathered). “Featherlike”
(Webster, 1959). Having lateral appendages distributed
in two continuous series inserted, respectively,
on opposite sides of an axis and antrorsely
oriented; a borrowed term used here to describe the
pattern of insertion and orientation of the proliferations
of the stigmatic surface in certain bamboos.
Jackson (1949) and Webster (1959) both confine
the botanical application of the term pinnate to
the arrangement of the leaflets on the rachis of a
compound leaf.
PISTIL (L. pistillurn, a pestle). See GYNOECIUM.
* “PLEIOCLADE (Gr. pleion, more; Itlados, branch).
Characterizes those branch complements that contain
more than one independent primary (firstorder)
axis. See BRANCH COMPLEMENT; PLEIOCLADE,
RESTRICTED; PLEIOCLADE, UNRESTRICTED. Cf. MONOCLADE.
* *PLEIOCLADE, RESTRICTED. Characterizes those
pleioclade branch complements whose component
axes are incapable of proliferating basally for lack
of buds at their proximal nodes.
* “PLEIOCLADE, UNRESTRICTED. Characterizes those
pleioclade branch complements whose component
134 SMITHSONIAN CONTRIBUTIONS TO BOTANY
axes are capable of proliferating basally from buds
at their proximal nodes.
PLEIOGENY (Gr. pleion, more; genos, birth). “An
increase from the parental unit, as by branching or
interpolation of members” (Jackson, 1949); a general
term for the particular examples described
under PROLIFERATION and PLEIOCLADE, as pertaining
to vegetative axes, and under INDETERMINATE, as
pertaining to reproductive axes.
PLICATE (L. plicutus, folded). “Folded into
plaits, usually lengthwise” (Jackson, 1949); marked
by longitudinal ridges suggesting a prior condition
of being folded like a collapsible Chinese fan; an
appearance shown by some bamboo leaf blades.
* “PLURICESPITOSE (L. pluris, of more; caespitosus,
from caespes, a tuft). Characterizes the clump
habit of a bamboo plant that embraces more than
one tuft of culms, all tufts belonging to an individual
plant being connected by subterranean axes
as in Arundinaria gigantea ssp. tecta. Cf. UNICESPITOSE.
POLYCARPIC. See POLYPERIODIC.
POLTPERIODIC (Gr. polys, many; periodos, a complete
course). Having many reproductive periods,
alternating with vegetative periods, within the lifetime
of one plant; characterizes bamboos (such as
plants of the known species of Phyllostachys) that
flower repeatedly (and usually periodically) during
an indefinite life span. Polycyclic suggests itself for
use in the sense here given polyperiodic, but is
rejected because of its long-established use with
reference to the occurrence of an indefinite number
of whorls of parts, such as sepals or petals in a
flower. Pleiocyclic is already burdened with the
connotation perennial, which is incompatible as a
meaning alternate to that here given polyperiodic.
Moreover, monocyclic, which would pair with the
pleiocyclic, already means annual, in reference to
the life span of a plant. In the interest of clarity,
the term polycarpic, sometimes used to indicate the
polyperiodic character of certain bamboos, should
be reserved (with its orthographic variants) for
reference to pistils with numerous carpels (see
Jackson, 1949). Cf. MONOPERIODIC.
**PRECOCIOUS (as applied to the proliferation of
the branch primordium at midculm nodes). Elaborating
secondary branch primordia before the
rupture of the bud-scale (prophyllum) that accompanies
the “germination” of the branch bud.
““PRIMARY. The first (initial) element in a
series of related axes or structures of a given category,
as branch buds or branches.
**PRIMITIVE. “Ancestral” in the sense of L. L.
Forman (1964:390 footnote).
PRIMORDIAL (L. primordialis). “First in order
of appearance” (Jackson, 1949). See PRIMORDIUM.
PRIMORDIUM (L., the beginning). An axis or an
outgrowth of an axis in its earliest recognizable
condition, or in an early dormant state. In the
bamboos, a branch primordium borne on a segmented
vegetative axis is always enclosed in a
prophyllum; an adventitious root primordium is
never so enclosed. See PROPHYLLUM.
An initial primordium
is a body of undifferentiated meristem which, when
enclosed by a prophyllum and remaining dormant,
forms part of a primary bud; it then constitutes an
incipient axis of the first order. When inserted at
a culm node, its first potential is to give rise to a
primary element of a branch complement. See
BUD, PRIMARY.
PROLIFERATION (L. proles, offspring; fero, I bear).
“Bearing progeny as offshoots” (Jackson, 1949).
The term is brought into focus here with reference
to the rapid multiplication of members of a branch
complement by the prompt awakening of buds at
the proximal nodes of the component members.
The proliferation of the culm itself by the same
process (without the intercalation of a rhizome) is
called stooling or tillering. Proliferation should not
be confused with prolification, which Jackson
(1919) differentiates as “The production of terminal
or lateral leaf buds in a flower.” An example
of prolification is cited under DETERMINATE. See
also PLEIOGENY.
**PROMONTORY (a physiographic term described
by Webster, 1959). I have borrowed this term to
describe the tapered bulge that extends downward
from the locus of initiation of the primordium of
the primary bud at midculum nodes in all of the
known species I here include in the genus Arthrostyfidium.
See GREMIAL.
PROPHYLLATE. Bearing a prophyllum (q.v.).
PROPHYLLUM (L. prophyllum, first leaf). A sheathing
organ, usually 2-keeled and inserted circumaxially
at the first (proximal) node of a branch. In
vegetative axes, and in inflorescences of indeterminate
branching, the prophyllum at first surrounds
the branch primordium to form a bud. Jackson’s
* *PRIMORDIUM, INITIAL.
NUMBER 9 135
(1949) definition of prophyllum limits it to what
is now generally called a palea. Prophylla are by
some authors called bracts, or bracteoles. See
PALEA; BUD.
PROXIMAL (L. proximus, nearest). Basal; situated
at or near the base of an axis or an organ;
designates loci of insertion, or structures, so situated.
Cf. DISTAL.
PSEUDOSPIKELET (L. pseudo, false; by extension,
superficially resembling; spicula, spikelet). A spikeletlike
branch of an iterauctant (indeterminately
branching) inflorescence. See INDETERMINATE.
PULVINUS (L., cushion). A trBpical organ (L.
trGpicus, pertaining to a turning) associated with
movement of differential growth. Pulvini usually
manifest themselves as dome-shaped eminences,
commonly in pairs, strategically located in relation
to the organ with whose orientation they are concerned-
at or near the base of the leaf petiole, for
example, or a branch of an inflorescence. Pulvini
function through one or the other of two physiological
mechanisms: changes of turgidity and differential
growth.
*#QUINATE (as applied to pleioclade midculm
branch complements). Typically consisting of only
five axes or branches.
RACEME (L. racemus, a bunch of grapes). A
semelauctant gramineous inflorescence with a single
order of (usually solitary) branches. Jackson’s
(1949) definition is not applicable here. A raceme
in which some of the branches emerge in fascicles
of tw:, or three may be characterized (interpreted)
as a paniculate raceme, if the extra branches in
each fascicle are taken to be secondary ones arising
subcutaneously from the base of a primary one.
Athroostachys capitata (Figure 19c) bears paniculate
racemes in which all branches are relatively
short. This gives the inflorescence a capitate superficial
appearance. A spicate raceme is one of which
the pedicels are so short that the inflorescence resembles
a spike.
RACHILLA, rhachilla (Gr. dim. of rachis, q.v.).
The axis of a spikelet in any gramineous plant.
Jackson (1949) defines rhachilla vaguely as “a secondary
axis in the inflorescence of grasses.”
RACHIS; rhachis (Gr. rhachis, backbone). The
primary axis of an inflorescence; its position is
terminal to the peduncle.
REDUCED. Subnormal in size; connotes also (in
some applications) either a failure to fulfill a normal
function, or a diminution in the expected
number of parts in a set (of stamens, for example).
Cf. DEPAUPERATE.
*#RESTRICTED (L. restrictus). Characterizes (1)
a branch complement-either monoclade or pleioclade-
as unable to proliferate from buds at proximal
nodes of its first-order component member (s);
and (2) the locus and the mode of insertion of a
monoclade branch complement, regardless of
whether the primary component (first-order) axis
is dominant or not.
RHIZOME (Gr. rhizbma, a mass of roots). An individual
component branch of the subterranean system
of segmented axes that constitute the “chassis”
(popularly referred to as the “rootstock”) of a bamboo
plant. A rhizome consists of two parts: the rhizome
proper and the rhizome neck. Two distinct
types of rhizome are differentiated: leptomorph and
pachymorph (q.v.).
RHIZOME SHEATH. The husklike sheathing organ
inserted at each node of a rhizome proper (as distinct
from the rhizome neck).
Of higher order in a given system
of related axes. Example: the branches that
arise by proliferation from buds basal to the primary
branch at midculm nodes in Arthrostylidium.
Cf. SUBSIDIARY.
* #SECUND (L. secundus, following). “Arranged
on one side only; unilateral, as flowers in some
racemes, spikes, etc.” (Webster, 1959). Applied to
“parts or organs directed to one side only, usually
by torsion” (Jackson, 1949). The term comes into
use in the description of bamboo inflorescences
with an excurrent rachis, where the modification of
the rachis, either by torsion or by asymmetrical
development, causes all of the spikelets, or branches,
to be oriented more or less strongly toward each
other on one side of the rachis, and away from
each other on the other side of it. In actual practice,
the term may be applied either to the inflorescence
as a whole, or to the parts (spikelets or
branches) whose orientation is the focus of attention.
SEMELAUCTANT (L. semel, once; auctans, increasing,
growing). Embracing but a single grand
period of growth. A determinate bamboo inflorescence
may be said to be semelauctant. See DETERMINATE;
ITERAUCTANT.
SENSU LATO; SENSU STRICTO (L.). In a broad
sense; in a restricted sense; expressions used to in-
**SECONDARY.
136 SMITHSONIAN CONTRIBUTIONS TO BOTAKY
dicate the intended scope of a given application of
a scientific name or a scientific term.
SHEATH (ME. shethe). A sheathing organ, the
basal part of which, the sheath proper, completely
surrounds the vegetative axis on which it is borne,
its locus of insertion being circumaxial. In its
simplest form (as in typical neck sheaths and
rhizome sheaths) the sheath proper terminates
apically in a short, hard point. In its fully elaborrated
form-that characteristic of examples in the
middle of any series-the sheath proper usually
terminates above in a ligule, at the base of which
is inserted a more or less expanded laminar appendage
referred to in a comprehensive sense as the
sheath blade. In addition, the sheath proper commonly
bears, at or near each extremity of the locus
of insertion of the blade, a tuft of bristles (oral
setae). These may be borne either directly on the
sheath proper or on the margin of an auricle. When
there is a pair of auricles (one on each side) the
two may be similar and subequal, or they may be
unequal in size and dissimilar in shape. To avoid
the ambiguity occasioned by a commonly encountered
indiscriminate usage of the term leaf (q.v.)
to refer to any one of the diverse forms of sheaths
borne on the vegetative axes of the bamboo plant,
two expedients are followed: (1) the adoption of
the differential terms rhizome sheath, neck sheath,
culm sheath, branch sheath, leaf sheath, and prophyllum
(q.v.); and (2) the differentiation of the
sheath proper from its appendages. A clear precedent
for this effort to avoid ambiguity is found
in the now fairly common adoption of a precise
terminology to differentiate the several types of
sheathing structures borne on the reproductive axes
of the bamboo plants, as bracts, prophylla, glumes,
lemmas, and paleas. See SHEATHING ORGAN.
SHEATH BLADE. A distinct foliar part, the
lamina, that is appended apically on the laminiferous
culm sheaths proper and branch sheaths
proper in any series. A sheath blade is distinguishable
from a leaf, first of all by the relatively proximal
position of the sheath proper of the former on
any aerial vegetative axis, while leaf sheaths are
always inserted at the distal nodes of any culm or
branch. Sheath blades are always sessile, while the
leaf blades are petiolate (in all known bamboos
except some species of Neurolepis). Moreover, characteristic
differences in shape are usually quite
marked. The strong divergence in form, as between
leaf blades and sheath blades in the bamboos, is
one of the commonest and most useful of the gross
morphological distinctions between members of the
Bambusoideae and those of other grasses. See LEAF.
SHEATH CALLUS (L. callus, hard skin). A somewhat
prominent ring of parenchymatous (not hard)
tissue which remains at a sheath node after the
abscission of a sheath (of a culm, especially) in
some bamboos (Phyllostachys nidulnria, for example).
Cf. GIRDLE.
SHEATH NODE. The circumaxial locus of insertion
of a sheath on any vegetative axis of a gramineous
plant (elaborated after Hackel, in Lamson-
Scribner and Southworth, 1890: fig. 1). The sheath
node is marked externally by a more or less prominent
offset in the surface of its axis. At the level of
each sheath node the crossing over and anastomosis
of fibrovascular bundles take place through a diaphra,
m which marks the internal boundary between
adjacent internodes. Cf. CULM NODE; see DIASHEATH
SCAR (Gr. escharu, mark). Jackson
(1949) defines Scar as “a mark left on a stem by a
separation of a leaf, or a seed by its detachment; a
cicatrix.” In the bamboos, a sheath scar is a narrow,
transverse, circumaxial trace, the locus of abscission
of a sheath proper. A sheath scar marks the pmition
of a sheath node (q.v.).
Any sheathing structure inserted
at a node of any vegetative or reproductive
axis in a gramineous plant. Among the bamboos,
distinguishable types of sheathing organs are rhizome
sheaths, neck sheaths, culm sheaths, branch
sheaths, leaf sheaths, prophylla, bracts, empty
glumes, lemmas, and paleas. Unless used, and interpreted,
with discretion, the often-encountered
term foliar organ is apt to be ambiguous, since it
may be construed as referring either to a sheath
proper alone, to a sheath proper with all of its
appendages, to the blade only, or to a bladeless
sheathing structure, such as a prophyllum. See
LEAF.
SHOOT (noun, from AS. sczotan, to move rapidly).
“ (1) A young growing branch or twig; (2) the ascending
axis; when segmented into dissimilar members
it becomes a.stem” (Jackson, 1949). See CULM
SILICA BODIES; also silica corpuscles (G. Kieselkdrpe,).
Bodies of silica that are secreted and persist
within the cells of various tissues of the plant,
PHRAGM.
SHEATHIKG ORGAN.
SHOOT.
NUMBER 9
particularly any epidermal layer. In both size and
shape, they range widely. At one extreme, they are
small, lack characteristic shape, and occur solitarily
or in small numbers within a given cell. At the
other extreme, they may be large enough individually
to fill a cell more or less completely (see
SILICA CELLS) and have a characteristic shape.
Metcalfe (196O:xlii, fig. 1) lists 20 “types” (forms)
of silica bodies, some of which are illustrated. He
states (p. xx) that “the silica-bodies in silica-cells
assume very characteristic forms when the grass leaf
is mature, and are of considerable value for diagnostic
and taxonomic purposes.” Ohki’s (1932: 73f)
key to Japanese genera of bamboos refers to the
occurrence of “silica corpuscles” in the long epidermal
cells of the leaf in Dendrocalamus [D. Zatiflorus]
and in the articulation (bulliform) cells of the leaf
in Sinobambusa [ S . tootsik] and Chimonobambusa
[ Ch . q uadrangu laris].
SILICA CELLS. Epidermal short cells each of
which is more or less completely filled by a single
silica body (Metcalfe, 1960:xx).
SPICATE (L. spicatus, bearing spikes or ears).
Having some or all of the characters of a spike
(q.v.). A spicate raceme is a raceme in which the
pedicels are so short that the inflorescence resembles
a spike.
SPIKE (L. spica, an ear of grain). In the bamboos,
a spike is a determinate inflorescence in which
the sessile or subsessile spikelets are inserted on a
solitary rachis. Since the distinction between a spike
and a spicate raceme is, in the terms of their definitions,
only a matter of the relative length of “subobsolete”
pedicels, it is sometimes difficult to choose
between these terms for the description of a given
inflorescence (as in some species of the genus
Merostachys, for example). According to Jackson’s
(1949) definition (not applicable here) a spike is
an “indeterminate inflorescence with flowers sessile
on a common elongated axis.” A loose interpretation
of Jackson’s definition of a spike has resulted
in the occasional misapplication, in the literature,
of the term spike to a spikelet (q.v.) of bamboo.
A basic structural
component of every normal gramineous inflorescence,
comprising a segmented axis (the
rachilla) and its appendages. The appendages (beginning
with the lowermost) are: empty glumes
(usually two, rarely more, sometimes only one,
rarely lacking entirely), lemmas (either variable or
SPIKELET (L. spicula, little spike).
invariable in number, according to the taxon involved),
and branches of the rachilla (one subtended
by each lemma), each bearing a palea and the parts
of a flower. In some taxa, one or more of either
the distal or the proximal lemmas, or both, may be
sterile by virtue of either being empty or subtending
an incompletely developed flower. See FLORET;
FLOWER. [Empty glumes= transitional g1umes.-ed.]
* “SPODOGRAM (Gr. spodos, ashes; gramma, something
drawn). The “ash picture” of a plant part,
such as leaf epidermis, which is revealed on a slide
prepared by the following method: the plant part
is burned to the white-ash stage and the product
transferred intact to a glass slide and mounted in
an agent such as aniline or Canada balsam. Spodograms
were used for purposes of plant classification
by Molish (1920) and for identification of bamboos
by Ohki (1932).
SPORADIC (Gr. sporadikos, dispersed). “Widely
dispersed or scattered” (Jackson, 1949); dispersed,
or irregular, in time, as when the individual plants
of a given generation of bamboos (from seeds of
a common origin) enter the reproductive phase at
different times, or at irregular intervals. In either
case, the flowering is said to be sporadic (Gamble,
1896:viii). Cf. GREGARIOUS.
STOMA (Gr. mouth, opening; pl., stomata). A
functional organ found commonly, but not everywhere,
in the epidermis that covers photosynthetic
tissues. A stoma consists of two guard cells and (in
the bamboos, as far as known) two subsidiary cells
(Porterfield, 1937). Changes in the turgor of the
guard cells result in the opening and closing 06 a
slitlike aperture (the stoma of Jackson, 1949) between
them.
**STYLE. The description of the style as a separate
or distinct structure encounters difficulties
which grow out of peculiarities in the ontogeny of
the gynoecium. In many bamboos, the stigmas are
at first sessile at the apex of the ovary and there
is no discernible style. Usually, a style develops
later, but rarely (as in Bambusa multiplex), the
stigmas remain sessile. It does not apeear useful
or logical to call these separate styles. The style
when it appears seems to be an outgrowth of the
pericarp; but only by means of anatomical studies
of the pistil (gynoecium) at different successive
stages in its development can the details of the
origin, structure, and relationships of the compo138
SMITHSONIAN CONTRIBUTIONS TO BOTANY
nent structures traditionally given independent
names and status be clarified.
A prefix indicating either an approximation
or some reservation or limitation in the use
of the term with which it is combined, as in subdistal,
suberect, subfamily, subfusiform.
#*SUBEQUAL (as applied to the transitional
glumes of a spikelet, or the component axes of a
midculm branch complement). Nearly, but not
quite of equal size.
Of a shape suggesting the concept
fusiform, but not corresponding to it precisely; the
approximate general shape of a pachymorph rhizome.
See FUSIFORM.
Of the same order but of smaller
size; example: smaller members of the complement
of initial (primary) buds or branches at midculm
nodes in Chusquea and Swallenochloa. Cf. SECONBranches
of higher order
arising from buds at the base of a dominant primary
branch.
SUBTEND (L. subtendo, I stretch underneath).
To precede on a common axis, as a foliar organ
precedes (subtends) a bud or branch inserted immediately
above it.
“Grooved or
furrowed” (Jackson, 1949); as where the otherwise
approximately cylindrical shape of the surface of
the internode of a segmented axis is modified by one
or more longitudinal depressions. In most bamboo
genera the palea in functional florets is sulcate with
a single longitudinal dorsal depression. The furrow
or groove that marks, on the exterior of the caryopsis
in many bamboos, the interior locus of the
attachment of the seed to the pericarp. Cf. HILUM.
SUTURE (Fr., from L. sutura, from suere, to sew).
“A junction or seam of union; a line of opening or
dehisccnce” (Jackson, 1949).
““SYMPODIAL (Gr. sym or syn, together; pous,
podos, foot). Having the form of a sympodium; for
example, a system of related axes wherein successive
branches assume the role or position of effective
leadership or dominance so that the axis of a
sympodium is composite; a term used earlier (Mc-
Clure, 1925) to designate the branching habit of
the type of rhizome described herein as pachymorph
(q.v.). Cf. MONOPODIAL.
**TARDY (as applied to the proliferation of the
branch primordium at midculm nodes). Elaborat-
SUB- (L.).
SUBFUSIFORM.
*“SUBSIDIARY.
DARY.
SUBSIDIARY BRANCHES.
*+SULCATE (L. sulcatus, furrowed).
ing secondary branch primordia after the rupture
of the bud scale (prophyllum) that accompanies the
germination of the branch bud.
TAXON (neo-Gr., from taxis, arrangement). “A
taxoncmic group or assemblage of plants or animals
having certain characteristics in common,
which we take as evidence of genetic relationship,
and possessed of some degree of objective reality”
(Rickett, 1958). Rickett adds, “We can use the
word wherever we can use ‘Taxonomic group’ in
referring to the characteristics, dynamics, distribution,
or uses of such an assemblage.” Morton (1957)
cautions that “where the words ‘taxonomic group’
cannot be appropriately substituted, the word taxon
is misused.”
TERATIC (Gr. teras, teratos, monster, wonder).
Abnormal; teratological, in the sense of conventional
usage, with particular reference to marked
deviations from the normal, or expected, morphologica!
expression. See teratic example of prolification
under DETERMINATE.
TERETE (L., teres, well-turned). Circular in
cross-section, tapered progressively from one end to
the other, and smallest at the distal end or tip (for
example, a small bamboo culm). Cf. OBTERETE.
TERNATE (as applied to pleioclade midculm
branch complements). Typically consisting of only
three axes or branches.
TESTA (L., covering, in a poetical sense of the
substantive). The outer coat of a seed.
TILLERING. “Throwing out stems from the base
of a stem” (Jackson, 1949); proliferation of a culm
from its basal (subterranean) buds, without the intercalation
of a rhizome proper.
TRAFANT (Fr., running, creeping). Ranging
widely and freely; a term used by the Rivieres
(1879:321 et passim) to characterize (1) the slender,
elongate type of rhizome herein called leptomorph
(q.v.), and (2) the bamboos that spread by this
means.
A term newly coined
to designate those sheathing structures that occupy
the physiologically transitional zone located immediately
below the first fertile kmma of a bamboo
spikelet. In many of the tribes of conventional
grasses, and in some bamboos, it is possible to
focus on one or more features that reveal a clearly
perceptible distinction between two categories of
structures in this zone, namely, empty glumes and
sterile lemmas (q.v.). Agrostologists generally have
**TRANSITIONAL GLUMES.
NUMBER 9 139
made the most (in some cases too much) of these
structures as sources of taxonomic characters. In
many bamboos, either the total number or the
number of members of each category (or both of
these features) found in a given spikelet may vary
as between spikelets in a single specimen. Again, in
some species, one may look in vain for a clear
distinction (disjunction) between “empty glumes”
and “s:erile lemmas.” In such cases, their taxonomic
value may be greatly reduced. For this reason, in
describing bamboo genera, I have adopted the
expedient of treating both categories as members
of a single morphological category, namely, transitional
glumes. Besides avoiding a certain psychological
difficulty that arises in some cases this
emphasizes the existence of physiological gradients,
and places in proper perspective the frequent appearance
of morphological gradients, in place of
the sharp (abrupt) morphological transitions sought
by the taxonomist.
TRIADELPHOUS (Gr. treis, three; adelphos, brother).
Having “filaments in three brotherhoods”; an
adjective cognate with Triadelphia; the name of “a
Linnean order of plants with their stamens in three
sets” (Jackson, 1949), and conventionally used to
describe the stamen complement of a flower when
its members have their filaments connate to form
three distinct groups. In bamboos with triadelphous
androecia, the three “brotherhoods” commonly
comprise one, two, and three stamens, respectively.
Cf. ~IONADELPHOUS; DIADELPHOUS.
““TCTFTED (as applied to the component pleioclade
branch complements at midculm nodes of a
bamboo). Indicates more or less profuse branching
from the basal nodes of the primordial branch. The
component axes of tufted branch complements may
be subequal, or unequal (with the primordial axis
[branch] more or less strongly dominant).
TURGIDITY (L. turgidus, inflated). Turgor, turgescence;
the firmness imparted by “the distention
of a cell or cellular tissue by water or other liquid”
(Jackson, 1949).
““UNEQUAL (as applied to the transitional
glumes of a spikelet, or the component axes of a
midculm branch complement). Of more or less
strongly disparate size.
* ‘UNICESPITOSE (L. unus, one; caespitosus, from
cmspes, a tuft). Characterizes the clump habit of a
bamboo plant that embraces but a single, moresr
less compact, tuft of culms-as in Bambusa (Guadua)
amplexifolia. Cf. PLURICESPITOSE.
UNILATERAL (L. unilateralis, from unus, one;
latzis, a side). One-sided. See SECUND.
“‘UNISULCATE. Provided with, or marked by, a
single horizontal groove. See SULCATE; HILUM.
**UNRESTRICTED. Characterizes a branch complement
(monoclade or pleioclade) as possessing the
potential for proliferating by way or buds at the
proximal nodes of its component first-order member
(s).
“The disposition
or method of arrangement of foliage leaves
within the bud” (Webster, 1959). Jackson’s (1949)
definition, “the order of unfolding of leaf buds,”
is not applicable here. In most known bamboos, the
developing leaf blades are individually rolled up
tightly along their long axis, with one edge at the
center of the roll.
**VERSATILE (L. versatilis, adaptable, turning
with ease from one thing to another). Applied
herein to a rhizome system capable of producing
both leptomorph and pachymorph axes, as in
some species of Chusquea. Versatile is elsewhere
(Jackson, 1949) “turning freely on its support, as
many anthers on their filaments.”
ZYGOMORPHIC (Gr. zygos, yoke; morphb, form).
Symmetrically divisible by a single plane; “used of
flowers which are divisible into equal halves in
one p!ane only, usually the anteroposterior” (Jackson,
1949). Normal bamboo flowers conform to this
restricted criterion for zygomorphy (see McClure,
19GGb: 114). According to Jackson (1949), “Sachs
extends the meaning to such flowers as may be
bisected in any one plane.”
VERNATION (L. vernatio, renewal).
Literature Cited
Allan, H. H.
1949. Wild Species Hybrids in the Phanerogams, 11. The
Botanical Review (Lancaster) , 15 (2):77-105.
The Gramineae, a Study of Cereal, Bamboo, and
Grass. 480 pages. Cambridge: Cambridge University
Press.
Arber, A.
1934.
Arechavaleta, J.
1894-1897. Gramineas Uruguays. Anales del Museo Nacional
de Montevideo, volume I. [Parts 1 (Feb.
1894) :29-78, 2 (Nov. 1894) :79-212, 3 (Sept. 1895) :
213-292, 4 (March 1896) :293-372, 5 (Aug. 1896):373-
452, 6 (June 1897) :453-581.1
New Species from Colombia, Collected by Mr. K.
von Sneidern. Botaniska Notiser, 1939:796-804, fig.
2.
Asplund, E.
1939.
Auge, C., and M. de Tor0 y Gisbert, editors
1949. Pequelio Larousse Ilustrado, Nueuo Diccionario
Enciclopddico. 1528 pages, Pans: Libreria Larousse.
Bailey, L. H.
1914. The Standard Cyclopedia of Horticulture. 6 volumes.
New York: The Macmillan Company. [Bamboo,
vol. 1:444-449, by C. D. Beadle.]
1925. Manual of Cultivated Plants Most Commml,
Grown in the Continental United States and Canada.
Revised edition. New York: The Macmillan
Company. [Original published 1924.1
Monographie des Graminkes. Volume 12: 135-334 in
Baillon, Histoire des Plantes. 13 volumes. Paris:
Librairie Hachette & Cie.
Baillon, H. E.
1893.
Barnes, C. R.
Bentham, G.
1898. Plant Life. New York: Henry Holt & Co.
1861. Flora Hongkongensis. . . . 482 pages. London: L.
Reeve & Co.
1881. Notes on Gramineae. The Journal of the Linnean
Society (Botany), 19: 14-134.
Bentham, G., and J. D. Hooker
1862-1883. Genera Plantarum.. . . 3 volumes. London: L.
Reeve & Co. [Bambuseae by G. Bentham, 1883:
3 (2): 10941096 (key) and 1207-1215 (systematic
treatment) .]
Tertiary Fossil Plants from Colombia, South America.
Proceedings of the United States National
Museum, 75 (Art. 24) :2-12, 5 plates.
Berry, E. W.
1929.
Bertoni, M.
1918. Graminaiceas de las Regiones Forestales Litorales
del Alto Parana (Paraguaya, BrasileFa y Argentina)
. Anales Cientificos Puraguayos, series 2, no.
2:143-166.
Bibra, F. von
1853. Beitrage zur Naturgeschichte von Chile. Denkschriften
der Kaiserlichen Akademie der Wissenschaften.
Mathematisch-naturwissenschaftliche Klasse.
(Vienna), 5 (2):73-142, 5 plates. [Coliquea quila,
page 115, by E. G. Steudel.]
Bocher, T. W.
1963. The Study of Ecotypic Variation in Relation to
Experimental Morphology. Regnum Vegetabile, 27
(Symposium on Biosystematics, Montreal 1962, Proceedings)
: 10-16.
1967. Continuous Variation and Taxonomy. Taxon,
16 (4) :255-258.
Brade, A. C.
1956. A Flora do Parque Nacional do Itatiaia. Boletim
do Parque Nacional do Itatiaia, 5:l-85. [Also appeared
in 1957. Rodriguesia 32:151-166, 7 plates.]
Brandis, Sir Dietrich
1907. V: Remarks on the Structure of Bamboo Leaves.
The Transactions of the Linnean Society of Lon.
don, second series (1904-1913) (Botany), 7:69-92.
1929. Notes on Arundinaria. Bulletin of the Torrey BOBrown,
C. A.
tanical Club, 56 (6):315-318.
Brown, N. E.
1901. Report on Two Botanical Collections Made by
Messrs. F. W. McConnell and J. J. Quelch at Mount
Roraima in British Guiana. The Transactions of
the Linnean Society of London, second series (Botany),
6:l-107, 14 plates.
Brown, R. W.
1954. Composition of Scientific Words. 882 pages. (Revised
1956). Baltimore: Published by the author.
1912. Bambudes nouvelles. Notulae Systematicae, 2 (8) :
243-246.
1913. Les Bambusdes-monogaphie, biologie, culture,
pinciplux usages. 2 volumes: Text (215 pages)
and Atlas (101 plates). Paris: P. Lechevalier.
Camus, E.-G.
Carlquist, S.
1961. Comparative Plant Anatomy. 146 pages. New
York: Holt, Rinehart & Winston.
Chase, A.
1914. Field Notes on the Climbing Bamboos of Porto
Rico. Botanical Gazette, 58:277-279. plate 21.
140
NUMBER 9
Cronquist, A.
1969a. On the Relationship between Taxonomy and Evolution.
Taxon, 18 (2) :177-187.
1969b. Broad Features of the System of Angiosperms.
Taxon, 18 (2) :188-193.
Cuatrecasas, J.
1959. Aspectos de la Vegetacibn Natural de Colombia.
Revista de la Academia Colombiana de Ciencias
Exactas, Fisicas y Naturales, 10:221-268.
Davis, P. H., and V. H. Heywood
1963. Principles of Angiosperm Taxonomy. 556 pages.
Edinburgh and London: Oliver and Boyd.
Observations sur les GraminCes, et descriptions de
genres et especes nouvelles de cette famille. Me‘-
moires de la Socitte‘ d’dgriculture, Sciences et Arts
d’Angers, 1: 157-212, plates 7-9.
Desvaux, A. N.
1831.
Diels, L.
1937. Beitrage zur Kenntnis der Vegetation und Flora
von Ecuador. Bibliotheca Botanica, 29 (Heft 116):
1-190 [Gramineae, pages 57-60, by R. Pilger].
Diels, L., and E. Pritzel
1904. Fragmenta Phytographiae Australiae Occidentalis.
Beitrage zur Kenntnis der PflanLen Westaustraliens,
ihrer Verbreitung und ihrer Lebens-Verhaltnisse.
Botanische Jahrbuicher fur Systematik, Pflanzengeschichte
und Pflanzengeographie, 35 (1):55-662.
[Xerochloa, pages 64-68, by 0. Stapf.]
Dietrich, A.
1831-1833. Caroli a Linne Species Plantarum . . . olim
Curante Carolo Ludovico Willdenow. Editio sexta
. . . Auctore Dr. Albert0 Dietrich. 2 volumes.
Berlin. [Volume 2 published in 1833.1
Dutra, J.
1938. Les Bambusees de Rio Grande du Sud. Revista
Sudamericana de Botanica, 5 (5/6) :145-152.
1913. Die Graser des Brasilianischen Staates ParanP.
Arkiu for Botanik Utgibvet af K . Suenska Vetenskapsakademien
i Stockholm, 13 (10) :1-83, pls. 1-4.
1836-1840 [-18501. Genera Plantarum Secundum Ordlnes
Ekman, R. L.
Endlicher, S.
Naturales Disposita. Vienna.
Engler, H., and K. Prantl
1887-1915. Die Natuirlichen Pflanzenfamilien. Leipzig:
Verlag von Wilhelm Engelman. [The following
treatments on Gramineae by E. Hackel: 1887, 2 (2) :
1-97 (Tribe XIII: Bambuseae, pages 89-97); 1897,
Nachtrag (zum 2-4 Teil) 1:39-47 (bamboos, pages
46-47) ; 1900, Nachtrag (zum 2-4 Teil) 2:3-7 (bamboos,
page 7). R. Pilger, 1906, Nachtrag (zum 2 4
Tei1)1:12-22 (Gramineae) .]
Sertulum Naiguatense; Notes on a Small Collection
of Alpine Plants from the Summit of Naiguata, in
the Mountains of Caracas. The Journal of Botany,
British and Foreign, 10 (I new series, vol 1) :261-
264.
Ernst, A.
1872.
Ewan, J.
1969. Historical Problems for the Working Taxonomist.
Taxon, 18 (2) :194-203.
Catalogue du jardin des plantes de son Excellence
Monsieur le Conite Alexis de Razoumousky b Gorenki.
viii + 76 pages. Moscow: N. S. Vsevolojsky.
[Description of Triglossum on pages vi-viii.]
Trigonobalanus, A New Genus of Fagaceae, with
Notes on the Classification of the Family. Kew
Bulletin, 17 (3):381-396, figures 1-3.
Fischer, F. E. L. von
1812.
Forman, L. L.
1964.
Fosberg, F. R.
1944. El PPramo de Sumapaz, Colombia. Jouznal of the
New York Botanical Garden, 45:226-234.
De la modification des enveloppes florala des Graminkes
suivant le sexe de leurs flews. Comptes
Rendus Hebdomadaires des Seances de I‘dcademie
des Sciences, 84: 197-200.
1881. Mexicanas Plantas . . . Pars Secunda, Gramineae.
Paris: Imprimerie Nationale. 160 pages, 1 plate
[Bambusaceae, pages 129-132.1
Fournier, E.
1877.
Franchet, A.
1889. Note sur deux nouveaux genres de Bambusees.
Journal de Botanique [Edited by L. Morot, Paris],
3 (17) :277-284.
Freier, F.
1941. Contribucion a1 Estudio de la Anatomia Foliar de
las Especies del Genero “Chusquea” de la Flora Argentina.
Revista Argentina de Agronomia, 8 (4):
364-379.
1945. Relacion entre la Anatomia Foliar del Genero
Neurolepis (Gramineae) y su Posicion Sistematica.
Darwiniana, 7 (1):103-107.
Frizzell, D. L.
1933. Terminology of Types. American Midland Naturalist,
14 (6) :637-668.
Galeotti, H.
1842. Enumeratio Graminearum et Cyperacearum ab
Henrico Galeotti, in regionibus Mexicanis collectarum.
Bulletins de l‘dcaddmie Royale des Sciences
et Belles-lettres de Bruxelles, 9 (2):227-249.
Gamble, J. S.
1896. The Bambuseae of British India. Annab of the
Royal Botanic Garden, Calcutta, 7 (1) :xvii+1-133,
plates 1-119.
Some Additional Bamboos of the Philippine Islands.
The Philippine Journal of Science, section C
(Botany) ~ 8 (4):203-206.
1913.
Gay, C.
1845-1854. Historia Fhica y Politica de Chile, . . . 8
volumes. Paris: Santiago. [1854: Gramineas, vol.
6:233-469, plates 7483, by E. Desvaux.]
A Preliminary Investigation of the North American
Canes (Arundinaria). Bulletin of the Torrey Botanical
Ctub, 70 (3) :297-309.
Gilly, C. L.
1943.
142 SMITHSONIAN CONTRIBUTIONS TO BOTANY
Godman, F. Ducane, and 0. Salvin, editors
1879-1888. Biologia Centrali-Americana; or, Contributions
to the Knowledge of the Fauna and Flora of
Mexico and Central America. 4 volumes. London.
[Botany by W. B. Hemsley. Vol. 3(1885): Gramineae,
pages 475-588; Tribe XIII: Bambuseae, pages
586588.1
Aulonemia, nouveau genre de la tribu des Bambusees.
Annales des Sciences Naturelles (Botanique),
series 3, 5:75-77, plate 4.
Chromosome Numbers in Southwestern Grasses, 11.
American Journal of Botany, 47 (10) :873-877.
Goudot, J.
1846.
Gould, F. W.
1960.
Gould, F. W., and T. R. Soderstrom
1967. Chromosome Numbers of Tropical American
Grasses. American Journal of Botany, 54 (6):676-
683.
1970. Chromosome Numbers of Some Mexican and Colombian
Grasses. Canadian Journal of Botany,
48: 1633-1639.
Grisebach, A. H. R.
1862. XVIII: Plantae Wrightianae e Cuba Orientali, Pars
11: (Monopetalae et Monocotyledones) . Memoirs
of the American Academy of Arts and Science,
new series, 8 (part 2):503-536. [Gramineae:531-
534.1
Flora of the British West Indian Islands, parts 6-7,
pages 507-789. London: Love11 Reeve and Company
*
Plantae Lorentzianae; Bearbeitung der Ersten und
Zweiten Sammlung Argentinischer Pflanzen des
Professor Lorentz zu Cbrdoba. A bhandlungen der
Koniglichen Gesellschaft der Wissenschaften xu
Gottingen, 19:49-280, 2 plates. [Gramineae: 249-
264.1
1864.
1874.
Hackel, E.
1899. Enumeratio Graminum Japoniae. Verzeichnis der
Graser Japans. Bulletin de l'Herbier Boissier, old
series, 7 (10) :701-726. [XII: Bambuseae:716-721.]
1902. Neue Graser aus Mexico. Annalen des K . K. Naturhistorischen
Hofmuseums, 17:254-256.
1903a. Neue Graser. Tribus: Bambuseae. Ueber Arthrostylidium
und Arundinaria. Bsterreichische Botanische
Zeitschrift, 53 (2) :67-76.
1903b. Neue Graser. Bsterreichische Botanische Zeitschrift,
53 (4) :153-159.
1903c. Neue Graser. Bsterreichische Botanische Zeitschrift,
53 (5) :194-197.
1903d. Neue Graser. Osterreichische Botanische Zeitschrift,
53 (index) :516.
1908. Gramineae Novae. V. Reptertorium Specierum Novarum
Regni Vegetabilis, 6: 153-161.
1909a. Gramineae Novae. VI. Repertorium Specierum
Nouaruin Regni Vegetabilis, 7:311-327.
1909b. Ex Herbario Hassleriano: Novitates Paraguarienses.
111. XII. Gramineae 11. Repertorium Novarum Specierum
Regni Vegetabilis, 7:369-374.
Hance, H. F.
1862. Manipulus Plantarum Novarum, Potissime Chinensium,
adjectis Notulis Nonnullis Affinitates, caet.
Respicientibus. Annales des Sciences Naturelles,
quatrieme serie (Botanique), 18:217-238.
1876. A New Chinese Arundinaria. Journal of Botany,
British and Foreign, 14 (new series 5):339-340.
Preliminary Report on the Flora of Southern Hidaka,
Hokkaido(Yezo), V. The Botanical Magazine
(Tokyo) 49 (578) :69-85. [A description of
Merostachys sikokianus by T. Nakai appears on
page 74.1
Hara, H.
1935.
Harper, R. M.
1928. Economic Botany of Alabama, Part 2: Catalogue of
the Trees, Shrubs and Vines of Alabama, with their
Economic Properties and Local Distribution. Geological
Suroey of Alabama Monograph, 9:357.
[Arundinaria, pages 72-77, figures 18-20.]
Hayata, B.
1907. Supplements to the Enumeratio Plantarum Formosanarum.
Botanical Magazine (Tokyo) , 21 (242) :
49-55.
Heiser, C. B., Jr.
1949. Natural Hybridization with Particular Reference
to Introgression. The Botanical Review (Lancaster),
15 (10) :645-687.
Herter, W. G.
1939-1943. Estudios Botnaicos en la Region Uruguaya, 14
(Flora Ilustrada del Uruguay). i-xvi+256 pages;
index (pages 1-13) . Montevideo. [Bambusaceae,
pages 4849, published in 1941.1
1940. Plantae Uruguayenses Novae vel Criticae. Revista
Sudamericana de Botdnica, 6 (5-6): 129-155.
Die von Dr. Th. Herzog auf seiner zweiten Reise
durch Bolivien in den Jahren 1910 und 1911 gesammelten
Pflanzen. Teil V. Mededeelingen van's
Rijks Herbarium (Leiden), 4O:l-77. [Bamboos,
pages 75-77, by J. T. Henrard.]
Herzog, T.
1921.
Hitchcock, A. S.
1905. The Identification of Walter's Grasses. Annual Report
of the Missouri Botanical Garden, 16:31-56.
1908. Types of American Grasses: A Study of the American
Species of Grasses Described by Linnaeus, Gronovius,
Sloane, Swartz, and Michaux. Contributions
from the United States National Herbarium,
12 (3) :v+ 158+v.
1913. Mexican Grasses in the United States National
Herbarium. Contributions from the United States
National Herbarium, 17:181-389+xiv.
920. The Genera of Grasses of the United States with
Special Reference to the Economic Species. United
States Department of Agriculture Bulletin, 772: 1-
307. [Revised 1936 by Agnes Chase, 302 pages and
20 plates.]
927a. New Species of Grasses from Central America.
Proceedings of the Biological Society of Washing
ton, 40:79-88.
NUMBER 9
192713. The Grasses of Ecuador, Peru, and Bolivia. Contributions
from the United States National Herbarium,
24 (8) :v+291-556+xx.
1935. Manual of the Grasses of the United States. United
States Department of Agriculture Miscellaneous
Publication, 200: 1-1040. [Revised 1951 by Agnes
Chase, second edition, 1051 pages.]
1936. Manual of the Grasses of the West Indies. United
States Department of Agriculture Miscellaneous
Publication, 243: 1-439.
Hitchcock, A. S., and A. Chase
1917. Grasses of the West Indies. Contributions from the
United States National Herbarium, 18 (7) :v+261-
47 1 + xviii.
195.5. Growth-habits of Monocotyledons-Variations on a
Theme. Phytomorphology, 5 (4) :399-413.
1958. The Bamboos of the Malay Peninsula. The Gardens’
Bulletin (Singapore) , 16:l-135.
Icones Plantarum. Volume 3, plates 273-274. London:
Longman, Orme, Brown, Green, and Longmans.
Holttum, R. E.
Hooker, W. J.
1840.
Huber, J.
1904. Miscellaneas Menores, V: Guadua superba Hub.
n. sp., a Taboca Gigante do Alto Rio Purus. Boletim
do Museu Goeldi (Museum Paraense) de Historia
Natural e Ethnographia, 4 (2 and 3):479-480.
Humboldt, F. H. A. von, and A. J. Bonpland
1808. Plantae Aequinoctiales . . ., volume 1, 232 pages.
Paris. [For bamboos see pp. 63-69. Plantae Aequinoctiales
. . . consists of two volumes, parts of
which appeared between 1805-1817, which comprise
part 6 (Botany), section 1, of Humboldt and
Bonpland, Voyage aux re‘gions e‘quinoctiales du
nouveau continent, fait en 1799-1804.1
Humboldt, F. H. A. von, A. J. A. Bonpland, and C. S. Kunth
q1816. Nova Genera et Species Plantarum . , ,, volume 1.
302 pages. Paris. [Nova Genera . . . consists of
seven columes, published between 1816-1825, which
comprise part 6 (Botany), section 3, of Humboldt
and Bonpland, editors, Voyage aux regions equinoctiales
du nouueau continent, fait en 1799-1804.1
Jackson, B. D.
1949. A Glossary of Botanic Terms. Fourth edition, 481
pages. London: G. Duckworth & Co.. Ltd.
Janaki Ammal, E. K.
1959. A Cyto-Systematic Survey of Bambuseae, I: The
Slender Bamboos of Asia and S. America. Bulletin
of the Botanical Survey of India, l(1) :78-84.
Notes on the Flora of Southeastern Kentucky, with
a List of Plants Collected in Harlan and Bell Counties
in 1893. Bulletin of the Torrey Botanical Club,
Kearney, T. H.
1893.
20 (12) :474-485.
Kendrick, W. B., and J. R. Proctor
1964 Computer Taxonomy in the Fungi Imperfecti.
Canadian Journal of Botany, 42 (1) :65-86.
Keng, K. H.
1957. One New Genus and Two New Species of Chinese
Bamboos. Acta Phytotaxonomica Sinica, 6 (4) :355-
360+plates 56 and 57.
Bemerkungen zu den “Gramineae Exsiccatae.” Allgemeine
Botanische Zeitschrift fur Systematik,
Floristik, Pflanzengeographie etc., 14: 160-161.
Gynaeceum oder Gynoecium? und anderes Sprachliche.
Verhandlungen der Physikalisch-Medicinischen
Gesellschaft in Wiirzburg, new series, 39:9-14.
1822a. Notice sur le genre Bambusa. Journal de Physique,
de Chimie, d’Histoire Naturelle et des Arts, 95:
1822b. Synopsis Plantarum . . ,, volume 1. 491 pages.
Paris. [Synopsis Plantarum . . . consists of four
volumes, published between 1822-1826, which comprise
part 6 (botany), section 5, of Humboldt and
Bonpland, editors, Voyage aux regions equinoctiales
du nouveau continent, fait en 1799-1804.1
1829-1834. Revision des Graminees. Part 6 (Botany), section
6, of Humboldt and Bonpland, editors, Voyage
aux regions equinoctiales du nouveau continent,
fait en 1799-1804. Paris: Librairie-Gide. [Bambuseae:
1829:137-141: 1830:323-334, plates 73-79;
1831:483-484; 1832:553-554; 1834:610-611, plates
155, 156, 191.1
1833. Enumeratio Plantarum’. . . . Agrostographia Synoptica
. . . Volume 1.606 pages. Stuttgart, Tubingen:
J. G. Cottae.
Enumeratio Plantarum . , . Agrostographia Synoptica
, . . Volume I1 (Supplement to Volume I).
436 pages + 40 plates. Stuttgart, Tubingen: J. G.
Cottae. [Arundinaria, pages 348-349.1
Kneucker, A.
1908.
Kraus, G.
1908.
Kunth, C. S.
148-151,
1835.
Kuntze, C. E. 0.
1891-1898. Revisio Generum Plantarum. . . . 3 volumes.
Leipzig. [vol. 1 (1891) :i-clvi+1-374; vol. 2 (1891):
375-1011; vol. 3 (1) (1893) :clvii-cccxx; vol. 3 (2)
(1898) :1-202; v01. 3 (3) (1898):l-576.1
Kurz, S.
1876. Bamboo and Its Use. Indian Forester, 1 (3):219-269,
plate 1, 2; 1 (4):335-362, plates 3, 4.
Lamarck, J. B. A. P. M. de
Tableau Encycloptdique et Mtthodique . . . Botanique.
Volume 1, part 1, 200 pages, 100 plates.
Paris: Panckoucker.
1791.
Lamson-Scribner, F.
1899. Recent Additions to Systematic Agrostology.
United States Department of Agriculture, Division
of Agrostology, Circular, number 15 (Agrostology
49) : 1-10. [Includes an English translation (by
Lamson-Scribner published anonymously) of the
item listed under Hackel, 1897.1
Some Mexican Grasses Collected by E. W. Nelson
in Mexico, 1894-95: Studies on American Grasses.
Lamson-Scribner, F., and J. G. Smith
1897.
144
United States Department of Agriculture, Division
of Agrostology, Bulletin, 4:ll-16.
Lamson-Scribner, F., and E. A. Southworth
1890.
Lechler,
1857.
Lindley,
1835.
The True Grasses, viii + 228 pages. New York:
Henry Holt & Company. [Includes an English
translation composing principally the item under
Hackel, 1887. References to Hackel, 1887, in the
present work are given as pages in the English
translation by Lamson-Scribner and Southworth.]
Berberides Americae Australis. Accedit Enumeratio
Plantarum, Quas in America Australis Detexit
W. Lechler. 55 pages, Stuttgart: E. Schweizerbart.
[Chusquea intermedia, nomen nudum, page 52, by
E. G. Steudel.]
In Penny Cyclopaedia of the Society for the Diffusion
of Useful Knowledge, 1833-1843. 27 volumes.
London: Charles Knight. [Botanical articles as far
as the letter “R” prepared by J. Lindley, teste
Agnes Chase. Description of Bambusa pubescens,
vol. 3 (1835) :357, apparently by Lindley with name
provided by G. Loddiges.]
W.
J.
Lindman, C. A. M.
1900. Beitrage zur Gramineenflora Siidamerikas. Kongelige
Svenska Vetenskaps-akademiens Handlingar,
34 (6) :I-52, 15 plates.
McClure, F. A.
1925.
1934.
1941.
1942.
1944.
1945.
1954.
1957.
1959.
1960.
1962.
Some Observations on the Bamboos of Kwangtung.
Lingnaam AgricuEtural Review, 3 (1) :40-47.
The Inflorescence in Schizostachyum Nees. Journal
of the Washington Academy of Sciences, 24(12):
On Some New and Imperfectly Known Species of
Chinese Bamboos. Sunyatsenia 6 ( 1 ) :28-51, 12
plates.
New Bamboos from Venezuela and Colombia.
Journal of the Washington Academy of Sciences,
Western Hemisphere Bamboos as Substitutes for
Oriental Bamboos fm the Manufacture of Ski Pole
Shafts. Project QMD-24, National Research Council
Committee on Quartermaster Problems. [Mimeographed.]
The Vegetative Characters of the Bamboo Genus
Phyllostachys and Descriptions of Eight New Species
Introduced from China, Journal of the Washington
Academy of Sciences, 35 (9):276-293.
A New Bamboo and a New Record for Guatemala.
Phytotogia, 5 (3) :81-82.
Typification of the Genera of the Bambusoideae.
Taxon, 6 (7) :199-210.
Typification of the Genera of the Bambusoideae,
11. Taxon, 8 (6) :208-209.
Typification of the Genera of the Bambusoideae,
111. Taxon, 9 (6) :194.
Typification of the Genera of the Bambusoideae,
IV. Taxon, 11 (4) :141.
541-548.
32 (6):167-183.
1963a.
1963b.
1966a.
1966b.
SMITHSONIAN CONTRIBUTIONS TO BOTANY
A New Feature in Bamboo Rhizome Anatomy.
Rhodora, 65: 134-136.
Typification of the Genera of the Bambusoideae,
V. Taxon, 12 (3):127.
A Glossary of the Bamboos. Taxon, 15 (6) :220-235.
The Bamboos-A Fresh Perspective. xv + 347
pages. Cambridge: Harvard University Press.
Maekawa, F.
1961. A New Dwarf Bamboo Genus Matudacalamus from
Mexico. Journal of Japanese Botany, 36 (10):343-
346.
Maguire, B., et al.
1969. The Botany of the Guayana Highland, Part 8.
Memoirs of the New York Botanical Garden,
lS(2):l-290. [Gramineae, pages 11-22, by T. R.
Soderstrom.]
Maguire, B., J. A. Steyermark, J. J. Wurdack, et al.
1957. Botany of the Chimanti Massif, 1: Gran Sabana,
Venezuela. Memoirs of the New York Botanical
Garden, 9 (3):393-439. [Gramineae, pages 393408,
by J. R..Swallen.]
Maguire, B., J. J. Wurdack, et al.
1957. The Botany of the Guayana Highland, Part 2.
Memoirs of the New York Botanical Garden, 9 (3) :
235-392. [Gramineae, pages 237-278, by J. R. Swallen.]
Botany of the Guayana Highland, Part 5. Memoirs
of the New York Botanical Garden, lO(5) :1-278.
[Bambusoideae, pages 1-6, by F. A. McClure.]
1964.
Marchant, J. R. V., and J. E. Charles [revised by]
1958. Cassell‘s Latin Dictionary. 927 pages, New York:
Funk & Wagnalls Company.
Markgraf, F.
1958. (Review of) Brade, A. C. A Flora do Parque
Nacional do Itatiaia. Bol. 5 do P. N. do Itatiaia.
Rio de Janeiro 1956. Botanische jahrbucher fur
Systematik, Pflanrengeschichte und Pflanzengeographie,
78 (Literaturbericht, Pflanzengeographie) :
1-2.
Martius, C. F. P. von
1829. Flora Brasiliensis, Seu Enumeratio Plantarum . . .
pol. 2, part 1, Agrostologia Brasiliensis, by C.
G. D. Nees von Esenbeck. 608 pages. Stuttgart,
Tubingen: J. C. Cottae.]
Flora Brasiliensis.. . . 14 volumes (in 40), published
in 130 fascicles. Munich, Vienna. [Bambuseae
by J. C. Doell, 1880: volume 2 (3) :161-219 +
plates 44-56.]
1840-1906.
Meisner, D. F.
1843-1846. Plantarum Vasculariurn Genera. . . . Leipzig:
Libraria Weidmanniana. pol. 1, “Tabulis Diag
nosticis,” pages 425-426 Tribus X: Festucaceae,
Subtribus 11: Bambuseae. Vol. 2: “Commentarius,”
pages 325-326: Neurolepis.]
Anatomy of the Monocotyledons, I: Gramineae. lxi
+ 731 pages. London: Oxford University Press.
[Some passages from the text are reproduced with
Metcalfe, C. R.
1960.
NUMBER 9 145
the permission of both the author and the publisher.]
Michaux, A.
1803. Flora BoreaH-Americana. . . . 2 volumes. Paris
and Strasbourg.
Mitford, A. B. Freeman
1896. The Bamboo Garden. 224 pages. New York and
London: Macmillan and Company.
Mohr, C. T.
1901. Plant Life of Alabama. Contributions from the
United States National Herbarium, 6: 1-921.
1782. Saggio sulla Storia Naturale del Chili. 367 pages,
folded map. Bologna: Tommaso d’Aquino.
1920. Aschenbild und Pflanzenverwandtschaft. Akademie
der Wissenschaften in Wien. Sitzungsberichte.
Mathematisch-naturruissenschaftliche Klasse. Abteilung
I (Vienna), 129:261-294, 2 plates. [Bamboo,
page 277, pl. 2.1
New Species of Plants from Guatemala. Phytologia,
1 (4): 145-150. [Chusquea lanceolata, pages 145-146,
by A. S. Hitchcock.]
The Misuse of the Term Taxon. Taxon, 6 (5) :155.
Muhlenberg, G. H. E.
Catalogus Ptantarurn Americae Septentrionalis , . .
iv+ 112 pages. Lancaster, Pennsylvania: William
Hamilton.
1817. Descriptio Uberior Graminum. , . . 295 pages.
Philadelphia: Solomon W. Conrad,
A Monograph of the Bambusaceae, Including Descriptions
of all the Species. Transactions of the
Linnean Society of London, 26: 1-157, 6 plates.
Arthrostylidium Presto% Munro [Gramineae Bambuseae].
Bulletin of Miscellaneous Information
[Royal Botanic Gardens, Kew, England], 1895:186.
Nipponobambusa Muroi, a New Genus of the Bambusaceae.
Hyogo-Ken chuto Kyoiku Hakubutsugaku
Zasshi, 6:88-90.
Molina, G. I.
Molisch, H.
Morton, C. V.
1935.
1957.
1813.
Munro, W.
1868.
1895.
Muroi, H.
1940.
Nakai, T.
1925. Two New Genera of Bambusaceae, with Special
Remarks on the Related Genera Growing in Eastern
Asia. Journal of the Arnold Arboretum, 6 (3) :145-
153.
1933. Bambusaceae in Japan Proper (I). Journal of
Japanese Botany, 9 (1) :5-34.
1942. Nipponocalamus, Genus Novum Bambusacearum.
Journal of Japanese Botany, 18 (7) :347-368.
1951. Miscellaneous Notes on Scientific Names of Japanese
Plants. Journal of Japanese Botany, 26 (11):
321-328. [Pleioblastus and Nipponocalamus, pages
326-327: in Japanese.]
Nash, G. V.
Nees von Esenbeck, C. G. D.
1903. A New Bamboo from Cuba. Torreya, 3:172-173.
1834. Bambuseae Brasilienses: Recensuit, et Alias in India
Orientali Provenientes Adjecit. Linnaea,
9[1835]:461-494.
The Genera of North American Plants, and a Catalogue
of the Species, to the Year 1817. 2 volumes.
Philadelphia: Published by the author.
Collections Towards a Flora of the Territory of
Arkansas. Transactions of the American Philosophical
Society held at Philadelphia for Promoting Useful
Knowkdge, new series, 5:139-203.
Nuttall, T.
1818.
1837.
Nybakken, 0. E.
1959. Greek and Latin in Scientific Terminology. 321
pages. Ames: The Iowa State University Press.
On the Systematic Importance of Spodograms in
the Leaves of the Japanese Bambusaceae. Journal
of the Faculty of Science, Imperial University of
Tokyo, Section 3: Botany, 4 (1) :1-130.
1934. On the Spodograms of the Leaves of the Genus
Shibataea Makino. Journal of Japanese Botany, 10
(Jan.):42-45.
Las Bambuseas Indigenas en la Mesopotamia Argentina.
Revista Argentina de Agronomia, 3 (4):229-
244.
Estudio Preliminar Sobre el Wnero “Chusquea” en
la Argentina. Revista Argentina de Agronomia,
8 (4) :331-345, plates 22-25.
1944. Arthrostylidium harmonicum, Nueva Especie de
Bambdsea del Perd. Physis, 19 (54) :478-481.
1945. Sinopsis de las Gramineas Chilenas del Genero
Chusquea. Revista Universitaria (Santiago), 30:61-
71.
1955. Significado de Algunos Terminos Usados en Botinica.
Revista Argentina de Agronomia, 22 (3) :
La Taxonomfa de las Gramineae Argentinas a la
Luz de las Investigaciones Mas, Recientes. Volume
1:125-130, in Recent Advances in Botany (From
Lectures & Symposia Presented to the IX International
Botanical Congress, Montreal, 1959). Toronto:
University of Toronto Press.
Ohki, K.
1932.
Parodi, L. R.
1936.
1941.
157-159.
1961.
Persoon, C. H.
1805-1807. Synopsis Plantarum. . , . 2 volumes. Pans,
Tiibingen. polume 1 published in 1805.1
1858. Plantarum Novarum Chilensium, Centuriae Sextae
Pars. Linnaea, 29 (1) :96-110.
1859. Plantarum Novarum Chilensium Centuriae Sextae
Continuatio. Linnaea, 30:185-212. [Bamboos, pages
206-207.1
1864. Plantarum Novarum Chilensium Centuriae inclusis
quibusdam Mendocinis et Patagonicis. Linnaea,
33: 1-308. [Bamboos, pages 298-300.1
1865. Excursion Botanica en Valdivia desde Los Cuncos
en el departamento de la Union, a traves de la
Cordillera de la Costa, hasta la Mar, por Federico
Philippi i descripcion de las Especies Nuevas de
Plantas Halladas en Ella por Rodolfo Armando
Philippi, R. A.
146 SMITHSONIAN CONTRIBUTIONS TO BOTANY
Philippi. Anales de la Universidad de Chile,
27 (3):289-351.
1873. Descripcion de las Plantas Nuevas Incorporadas
Ultimamente en el Herbario Chileno. Anales de la
Universidad de Chile, 43:479-583.
1896. Plantas Nuevas Chilenas de las Familias que
Corresponden a1 Tom0 VI de la obra de Gay.
Anales de la Universidad de Chile, 94:341-362.
Pilger, R.
1898.
1899.
1905.
1920.
1921.
1923.
1924.
1927.
Plantae Stiibelianae Novae. Botanische Jahrbucher
fur Systematik, Pflanzengeschichte und Pflanzengeographie,
25: 709-72 1.
Gramineae Lehmannianae et Stiibelianae Austro-
Americanae Additis Quibusdam ab Aliis Collectoribus
Ibi Collectis Determinatae et Descriptae. BOtanische
Jahrbucher fiir Systematik, Pflanzengeschichte
und Pflanzengeographie, 27:35-36.
Gramineae Andinae. I. Bambuseae. Repertorium
Specierum Novarum Regni Vegetabilis (F. Fedde,
editor), 1 (10):145-152.
Gramineae Austro-americanae Imprimis Weberbauerianae,
V. Botanische Jahrbiicher jiir Systematik,
Pflanrengeschichte und Pflanzengeographie
56, Beiblatt, 123 (2) :23-30.
Drei Neue Andine Graser. Repertorium Specierum
Novarum Regni Vegetabilis (F. Fedde, editor) ,
Gramineae Nonnullae Austro-Americanae. Notizblatt
des Botanischen Gartens und Museums zu
Berlin-Dahlem, 8 (76):452-456.
Plantae Tessmannianae Peruvianae, V. Notizblatt
des Rotanischen Gartens und Museums zu Berlin-
Dahlem, 10:124-126.
Plantae Tessmannianae Peruvianae, V. Herausgegeben
von J. Mildbraed. Gramineae. Notizblatt des
Botanischen Gartens und Museums zu Berlin-
Dahlem, 10 (91):114-115.
17: 445-448.
Pittier, H.
1892. Viaje de Exploraci6n a1 Valle del Rio Grande de
Terraba (Enero-Febrero 1891) . Anales del Znstituto
Fisico-Geogra’fico y Del Museo Nacional de Costa
Rica, 3:61-106. [Primera Parte. Diario de la EXpedici6n.
Capftulo 1. De San Jod a Santa Maria
de Dota, 15-16 de Enero, pages 61-65.]
Poiret, J. L. M.
1804. Encyclope‘die Mdthodique Botanique, 6 (1) :I-384.
1817. Encyclopddie Me‘thodique Botanique, supplement,
Paris.
5:l-780. Paris.
Histogenesis in the Bamboo with Special Reference
to the Epidermis. Bulletin of the Torrey Botanical
Club, 64 (7):421-432.
Porterfield, W. M., Jr.
1937.
Prat, H.
1936. La Systematique des Graminees. Annales des
Sciences Naturelles, 1Oe serie: Botanique, 18: 165-
258.
Presl, K. B.
1825-1835. Reliquiae Haenkeanae, , . . 2 volumes.
Prague: J. G. Galve. [Vol. 1(4-5) :207-355, Gramineae,
by J. S. Presl, published in 1830. Includes
“Bambusaceae,” 1 (4) :256-257.1
Agrostografia Brasiliensis. 58 pages, 1 table, 1 plate.
Lucca: Tipografia Ducale.
Raddi, J.
1823.
Rafinesque Schmaltz, C. S.
1808. Prospectus of Mr. Rafinesque Schmaltz’ two Intended
Works on North-American Botany; the
First on the New Genera and Species of Plants
Discovered By Himself, and the Second on the
Satural History of the Funguses, or Mushroom-
Tribe of America. Medical Repository, 2352-353.
Caratteri di Alcuni hruovi Generi e Nuove Specie
di Animali e Piante detla Sicilia. 105 pages, 20
plates. Palermo: Sanfilippo.
The Embryo in Grass Systematics. American Journal
of Botany, 44 (9) :756-768.
1810.
Reeder, J. R.
1957.
Reitz, R., editor
1967. Flora Ilustrada Catarinense. Gramineas-Suplemento
(Bambdseas), by F. A. McClure and L. B. Smith.
78 pages. Itajai: Tipografia e Livraria Blumenauense
S. A.
Rickett, H. W.
Rickett, H. H., and W. H. Camp
1958. So What Is a Taxon? Taxon, 7 (2) :37-38.
1950. The Application and Use of Botanical Names. Bulletin
of the Torrey Botanical Club, 77 (4) :245-261.
1879. I.es Bambous. Bulletin de la SociCtC Nationale
d’AccZirnatation de France, 5:221-253, 290-322, 392-
[Preprinted (1878) as a separate, continuously
paged publication: citations in the present work
refer to the 1879 edition, which contains a few
minor corrections and emendations.]
Rivibre, A. and C.
421, 460-478, 501-526, 597-645, 666-721, 758-828.
Roemer, J. J ~ , and J. A. Schultes
1817-1830. Caroli a Linnd, Equitis, Systema Vegetabilium.
. . . 7 volumes. Stuttgart: J. G. Cottae. [Bamboos,
Volume 7 (2):1336-1361, published in 1830, by J. A.
and J. H. Schultes.]
Addenda ad Floram Regionis Chaco Australis (pars
secunda) . Bulletin de l’Acaddmie Internationale de
Ge‘ographie Botanique, series 4, 28:155-165. [Two
title pages appear in the copy of the National
Agricultural Library, United States Department of
Agriculture. One bears “Tome XXVIII” and the
other “Tome XXVII,” although both are dated 1918;
28 is assumed to be the correct number as the
volume for 1919 is known to be volume 29, even
though it is incorrectly printed “Tome XIX.”]
Rojas Acosta, N.
1918.
Roxburgh, William
1832. Flora Zndica. 2nd edition, edited by W. Carey. 3
volumes. Serampore.
Ruprecht, F. J.
1839. Bambuseae Monographice Exponit. 75 pages, 18
plates. St. Petersburg: Typis Academiae Caesareae
NUMBER 9
Scientiarum. [Reprinted in 1840 under title “Bambuseas
Monographice Exponit” in Memoires de
I’Academie Impdriale des Sciences de Saint-Petersbourg,
sixikme serie (Sciences Naturelles), 3 (I):
91-165, plates 1-18.]
Some Account of the Curate, a Grass of the Tribe
of Bambuseae, of the Culm of Which the Indians
of Guiana Prepare Their Sarbacans or Blowpipes.
The Transactions of the Linnean Society of London,
18 (4) :557-562. [Note by J. J. Bennett with description
of Arundznaria schomburgkii, page 562.1
Some Mexican Grasses Collected by E. W. Nelson
in Mexico, 1894-95: Studies on American Grasses.
United States Department of Agriculture, Division
of Agrostology Bullatin, 4: 11-16.
Beitrage zur Wachstumsgeschichte der Bambusgewachse.
Journal of the Faculty of Science, Zmperial
University of Tokyo, 13:427-496, plates 22-24.
Algumas Especies Novas da Flora de Minas Geraes.
Archivos do Museu Nacional do Rio de Janeiro,
22:99-102 + 2 figures.
Sodiro, L. [Padre Luis (Aloysio)]
Una Excursidn Botdnica. 25 pages. Quito: Imprenta
Nacional (privately printed).
Gramineas Ecuatorianas de la Provincia de Quito.
Anales de la Universidad Central del Ecuador
(Quito), 3 (25) :474-484. [Chusquea quitensis,
nomen nudum, page 484, by E. G. Hackel.]
1955. Plants Collected in Ecuador by W. H. Camp.
Gramineae. Memoirs of the New York Botanical
Garden, 9 (2):133-145. [Bamboos identfied by F. A.
McClure.]
Schomburgk, R. H.
1841.
Scribner, F. L., and J. G. Smith
1897.
Shibata, K.
1900.
da Silveira, A. A.
1919.
1881.
1889.
Sohns, E. R., and J. R. Swallen
Sprengel, A.
1825. Caroli Linnaei . , . Systema Vegetabilium. Editio
Decima Sexta. Volume 2, 939 pages. Gottingen. [The
complete work consists of 5 volumes, published
between 1825 and 1828.1
Standley, P., and J. A. Steyermark, editors
1955. Flora of Guatemala. Fieldiana (Botany), 24, part
XI, Grasses of Guatemala (by J. R. Swallen) , ix +
390 pages. [Bamboos by F. A. McClure. The genera
of bamboos and other grasses arranged in one
alphabetical sequence.]
Stapf, 0.
1904a. On the Fruit of Melocanna bambusoides, Trin., an
Endospermless, Viviparous Genus of Bambuseae.
The Transactions of the Linnean Society of London
(Botany) , 6 (9) :401-425, plates 45-47.
1904b. Himalayan Bamboos, Arundinaria Falconeri and
A. falcata. Gardeners’ Chronicle, series 3, 35 (907-
910):305-306. [Reprint (undated) paged 1-41.
1913. Arthrostylidium angustiflorum, Stapf [Gramineae-
Bambuseae]. Bulletin of Miscellaneous Znformation
[Royal Botanic Gardens, Kew, England], 1913:268-
269.
Stearn, W. T.
Botanical Latin: History, Grammar, Syntax, Terminology
and Vocabulary. 566 pages. New York:
Hafner Publishing Company.
Postulates Basic to the Formulation of a Species
Concept. 1 page, mimeographed.
1966.
Stebbins, G. L., Jr.
1940.
Steudel, E. G.
1854. Synopsis Plantarum Glumacearum, Pars I: Synopsis
Plantarum Graminearum. 475 pages. Stuttgart:
J. B. Metzler. [Pages 32S339, Tribus XI: Bambusaceae.]
Flora del Auyan-tepui. Acta Botinica Venezuelica,
2:5-370. [Gramineae, pages 13C-135, by J. R.
Swallen.]
Steyermark, J.
1967.
Steyermark, J. A., et al.
1951. Contributions to the Flora of Venezuela: Botanical
Exploration in Venezuela-I. Fieldiana (Botany) ,
28 (1) :I-242. [Arthrostylidium, pages 31-33, by F. A.
McClure; Myriocladus, pages 34-36, by J. R.
Swallen.]
1931. Five New Grasses from Colombia. Journal of the
Washington Academy of Sciences, 21 (1) :14-16.
1938. Additions to the Grass Flora of British Honduras.
Journal of the Washington Academy of Sciences,
28 (1) :6-11.
1940. Miscellaneous New American Grasses. Journal of
the Washington Academy of Sciences, 30 (5) :209-
217.
1943. Merostachys pauciflorus Swallen, sp. nov. The
American Midland Naturalist, 29 (2) :469470, figure
1.
Grasses of Guatemala. Fieldiana (Botany), 24 (part
2) : ix + 390. [Bamboos, by F. A. McClure.]
Swallen, J. R.
1955.
Takenouchi, Y.
1931a. Systematisch-Vergleichende Morphologie und Anatomie
der Vegetationsorgane der Japanischen Bambus-
Arten. Memoirs of the Faculty of Science and
Agriculture, Taihoku Imperial University, 3 (1): 1-
60, 3 plates.
1931b. Morphologische und Entwicklungs-mechanische
Untersuchungen bei Japanischen Bambus-Arten.
Memoirs of the College of Science, Kyoto Imperial
University, series B (Biology) , volume 6 (3) :!09-160,
plates 18-20.
1820. Fundamenta Agrostographiae. . . . x + 214 pages,
index, 3 plates. Vienna: J. G. Heubner.
[1823-11828-1836. Species Graminum: Iconibus et Descriptionibus
ZElustravit [Auctor]. 3 volumes. St. Petersburg:
Impensis Academiae Imperialis Scientiarum.
1835. Bambusaceas Quasdam Novas Describit. Mtmoires
de I’Acaddmie Zmptriale des Sciences de Saint-
Pe‘tersbourg, sixieme serie (Sciences Naturelles) ,
Trinius, C. B. von
1 : 613-629.
148 SMITHSONIAN CONTRIBUTIONS TO BOTANY
1836. Species Graminum lconibus et Descriptionibus Illustravit
[Auctor]. Volume 3. St. Petersburg: Impensis
Academiae Imperialis Scientiarum. [This work consists
of three volumes, issued in separate parts
between 1823-1836. The title page of volume 3
bears the date 1836, although separate parts of
this volume appeared at earlier undetermined
dates.]
Urban, I., editor
1898-1928. Symbolae Antillanae, Seu Fundamenta Florae
Indiae Occidentalis. 9 volumes. Berlin: Fratres
Borntraeger, [The following contributions by R.
Pilger. 1900-1901, vol. 2:336343, “Arthrostylidium;”
1903, vol. 4 108-109, Arthrostylidium and
Bambusa, this volume also representing the Flora
Portoricensis; 1907, vol. 5:288-289, “VIII: Nova
Genera et Species Gramineae.”]
Eighteen Chromosome Pairs in an American Bamboo,
Chusquea subtessellata Hitchc. The Journal of
Agriculture of the University of Puerto Rico,
Virkki, N.
1963.
47 (2) :98-101.
Walker, E. R.
1906. On the Structure of the Pistils of Some Grasses.
The University of Nebraska University Studies,
6:203-218, plates 1-5.
Ftora Caroliniana . . . 263 pages. London: J.
Fraser.
Weberbauer, A.
Die Pflanzenwelt der Peruanischen Anden. [Translated
and revised, 1945, EZ Mundo Vegetal de 10s
Andes Peruanos, 776 pages, 39 plates. Lima:
Walter, T.
& 1788.
191 1.
Estacion Experimental Agrfcola de la Molina, Ministerio
de Agricultura.]
Webster, N.
1959. Webster’s New International Dictionary of the
English Language. Second edition, unabridged.
Springfield, Mass.: Merriam. [In the 1959 printing
here cited, and used, the section entitled “Addenda,”
expanded to incorporate definitions of many words
not found in earlier printings, was copyrighted in
1954.1
West, E. M.
1935. Canebrakes of the Southeastern United States. The
Ohio State University Press. Abstracts of Doctors’
Dissertations, number 16253-265.
Ergebnisse der Botanischen Expedition der Kaiserlichen
Akademie der Wissenschaften Nach Sudbrasilien
1901, I: Pteridophyta und Anthophyta. Denkschriften
der Kaiserlichen Akademie der Wissenscha
ften. Mathematisch - naturwissenscha ftliche
Klasse, 79:vi + 311 pages, 26 plates. [Gramineae,
pages 62-83, including bamboos, pages 81-83, by
E. Hackel. This appeared in preprint form in 1906
as pages 1-22.]
Wettstein, R. von, editor
1908.
Willdenow, C . L.
1808. Beschreibung eines Neuen Baumartigen Grases.
Der Gesellschaft Naturforschender Freunde LU
Berlin Magazin fur die Neuesten Entdeckungen in
der Gesammten Naturkunde (Berlin), 2:320.
Linnaeus Species Plantarum. 6th edition,
The American Botanist and Florist. . . 449 pages.
New York: A. S. Barnes & Company.
Willdenow, C. L., editor
1833.
Wood, A.
1871.
* U. S. GOVERNMENT PRINTIN6 OPPlCEs lD73-484-323/43