STUDIES IN THE FAMILY ORCHIDACE/E
OAKES AMES A.M.
' :
.
• A-V • . : . ' "
r ' ■
- '• . „
-
NUNC COCNOSCO EX PARTE
TRENT UNIVERSITY LIBRARY
PRESENTED BY
Mrs. H. H. Graham
To 7‘^o^k Oh
WiHL {j+aX Wsa£*M.
>j' >M~T (2* «»\Aa/». L-VwvW .
J
O (Wi /ivtn^. iawXC ^v**o* ^£v«. PB^’n^m- , Tv/u~
qHu^La 0^,0*. Tr^fiUon* ijv-W * w~‘~v>
dL w H ^ c—' e ^ ^ u^-
^ t£~ i 9 tL~Jc •***
0<~*UjUy i^vtf, /• ***•'-— tf
dA^M««U ,
'3 *J
Digitized by the Internet Archive in 2019 with funding from Kahle/Austin Foundation
https://archive.org/details/orchidaceaeillus0007unse
ORCHIDACEiE
VII
OR CH IDA C EJL
ILLUSTRATIONS AND STUDIES OF THE FAMILY ORCHID ACEJE
ISSUED FROM THE AMES BOTANICAL LABORATORY NORTH EASTON, MASSACHUSETTS
• •
•
POGONIA AND ITS ALLIES IN THE NORTHEASTERN UNITED STATES AND OTHER PAPERS
By OAKES AMES
WITH THIRTEEN PLATES BY BLANCHE AMES
Fascicle VII
BOSTON
THE MERRYMOUNT PRESS 1922
COPYRIGHT, 1922, BY OAKES AMES ALL RIGHTS RESERVED
PUBLISHED APRIL 5, 1922
D. B. UPDIKE • THE MERRYMOUNT PRESS • BOSTON
TO
GEORGE LINCOLN GOODALE, A.M., M.D., LL.D. TEACHER AND FRIEND THIS SEVENTH FASCICLE OF ORCHIDACEiE IS CORDIALLY DEDICATED
CONTENTS
PAGE
Preface jx
A Discussion of Pogonia and its Allies in the Northeastern United States with References to Extra-limital Genera and Species 3
Notes on Mexican Species ofTriphora 39
A Triandrous Form of Psilochilus macrophyllus 45
Studies of Otostylis brachystalix and the Species with which it has been confused 49
Notes on Erythrodes with Nomenclatorial Changes and Descriptions of Three New Species 63
Nomenclatorial Changes in Dendrochilum 79
New or Noteworthy Orchids from Different Parts of the World 83
Additions to the Orchid Flora of the Mountain Province, Luzon 141
Two New Species of Malaxis from Haiti 156
A New Oncidium from Haiti 159
Summary of New Genera, New Species and New Combi¬ nations proposed in the Present Volume 161
Index 169
PREFACE
IN the present volume of Orchidaceae I have brought together several short papers which represent studies of genera and species taken from a wide geographical range.
The paper on Pogonia and its Allies in the Northeastern United States may annoy those who are weary of the nomen- clatorial instability which seems more and more inseparable from systematic botany ; in fact, this paper may appear to be an at¬ tempt to justify a modern tendency that is positively dangerous in the realm of taxonomy. I have endeavored to show by plates of commendable completeness every structural character that seems important in the contemplation of generic cleavage. If these characters do not carry conviction, the plates that display them will, at least, furnish to opponents of change an accurate record on which to base objections and from which to draw conclusions.
The paper on Otostylis and Cyrtopodium does more than clarify a perplexing situation: it trenchantly reveals a difficulty that is always present when we attempt to identify plants. It indicates that many of our most carefully approached determi¬ nations may be faulty even when contrasted characters fit to¬ gether with delightful ease.
Oakes Ames
Bussey Institution for Applied Biology Harvard University
January, 1922
ORCHIDACEjE
.
A DISCUSSION OF POGONIA AND ITS ALLIES IN THE NORTHEASTERN UNITED STATES
WITH REFERENCES TO EXTRA-LIMITAL GENERA AND SPECIES
POGONIA taken in the broad sense is a group of rather sharply defined subgenera. That these subgenera should be raised to generic rank is, in the opinion of several authorities, very desirable. Consequently a diversity of treatment of Pogonia and its allies has resulted, not only in our local floras, but in more comprehensive works that deal with the flora of the world.
The present tendency is to make genera on rather narrower lines than prevailed when Bentham and Hooker published their Genera Plantarum, and to disregard the advantages that are of¬ fered by the maintenance of that very conservative though ad¬ mirable category known as the subgenus. The advocates of this tendency use a variety of arguments to defend their practice, one argument being that large genera are unwieldy and difficult to monograph, and that careful cleavage along subgeneric lines results in opportunities for comprehensive investigations that will hasten the arrival of that taxonomic stability which, in the final analysis, must rest on conscientious monographic work.
Conservative systematists view with justifiable horror and dis¬ tress the multiplication of synonyms that has attended recent activities in species making, and in their opinion the likelihood of an increase in this real evil is intensified by the difficulties that usually arise when critical species, owing to uncertainties as to their final resting-place, are passed from one genus to another. It is admitted by the conservatives that synonyms are unavoidable, but as they must always be with us it is argued that surely there is an advantage in the avoidance of generic synonyms by the
[ 3 ]
ORCHID ACEiE
maintenance of broadly characterized, easily defined and quickly recognized genera.
When we turn to Pogonia as usually defined and study it in¬ tensively, it becomes very evident that conservative systematists have carried to extremes the effort to keep the genus intact. In¬ deed, some of the subgenera are too clearly differentiated from one another to allow of any doubt as to the desirability of elevat¬ ing them to higher rank. Nervilia and Codonorchis are no longer regarded as subgenera of Pogonia ; they have been transferred to distinct subtribes, and there is no reason to question the treat¬ ment that has resulted in their removal from the Pogonia alli¬ ance. But there are other groups of species, those referred to Cleistes, Psilochilus, Isotria and Triphora, that cannot be so easily disposed of, and it is with regard to their treatment that opinions vary.
I propose to discuss here the subgenera of Pogonia that come within the range covered by the seventh edition of Gray’s Man¬ ual and, by means of detailed plates, to make clear those charac¬ ters that seem to be most important in considerations of generic segregation of the species.
The so-called extremists recognize Pogonia, Isotria and Tri¬ phora, but there is no thoroughgoing discussion of the charac¬ ters on which these genera are based, although in a paper on the anatomy of our native Pogonias Theo. Holm1 indicated differ¬ ences that he regarded as of sufficient importance to necessitate the recognition of three clearly differentiated groups of species. In his observations of the subterranean system of Pogonia trian- thophora he made very clear that there are morphological pecu¬ liarities that warrant the withdrawal of this species from Pogo¬ nia proper, and I believe for other reasons that it is wise to re-
1 American Journal of Science ser. 4, 9 (1900) 13-19.
[ 4 ]
ORCHIDACEdE
gard P. trianthophora as a representative of a very well defined group of which Triphora gentianoides (Sw.) A vies <% Schltr. covib. nov. ( Arethusa gentianoides Sw. FI. Ind. Occ. 3 (1806) 1436), may be taken as an example.
Cleistes has been looked on as a group referable to Pogonia. As a subgenus it is represented in our range by Pogonia divari- cata. If it should be deemed wise, as I believe it is, to treat P. ophioglossoides and its close allies in Asia as members of a distinct genus readily characterized by simple pollen grains, then P. di- varicata and the neotropical species to which it is related would constitute the genus Cleistes. Under such circumstances P. ver¬ ticil lata andP. ajfinis of the Manual should be segregated as mem¬ bers of a distinct group and transferred to Isotria, the most con¬ spicuous character on which to base segregation being the verti- cillate leaves and their behavior in vernation. (Cf. Plate 105, fig. 9.)
Just how much reliance to place on the character of the pollen grains is, of course, a debatable question. It would seem that the tendency of the pollen tetrads to break down early as in Cypri- pedium, Cephalanthera and a few other orchid genera ought to be taken as a fundamental characteristic in the contemplation of generic differences. In Pogonia ophioglossoides the pollen te¬ trads disintegrate into simple grains comparatively early in the development of the flower; in fact, they may be found thoroughly separated in buds that are still green. We find the same condi¬ tion in the Asiatic P. japonica Reichb. f. H. G. Reichenbach in his elaborate treatise on the structure of the pollen of the Orchi- daceae figured the pollen grains of P. ophioglossoides as simple at flowering time, not compound.1 InP. ophioglossoides var. bra- chypogon Fernald and in P. japonica var. minor Mak., the pollen
1 De Pollinis Orchidacearum Genesi ao Structura (1852).
[ 5 ]
ORCHIDACE^E
grains are simple, so that this character would seem to be con¬ stant in the Eupogonia group and may well be regarded as a weighty bit of evidence in favor of restricting to Pogonia only those species of the alliance which produce simple rather than compound pollen grains. It is true that Wettstein in a paper on Cephalanthera, Epipactis and Limodorum,1 in which he advo¬ cated the union of these genera under Epipactis, dismissed the evi¬ dence of simple pollen grains in Cephalanthera on the ground that in this genus they simply break down earlier than in Epipactis. Col. M. J. Godfery2 in a discussion of this paper maintains that Wettstein has treated the subject of the pollen grains in a some¬ what disingenuous manner and that, notwithstanding his conclu¬ sions, the differences between the pollen grains of Epipactis and Cephalanthera pointed out by Richard are undiminished in im¬ portance. In other words, Godfery is inclined to view the nature of the pollen as important in systematic work among the orchids, at least with regard to Epipactis and Cephalanthera.
Pogonia, Cleistes, Psilochilus, Isotria and Triphora are mem¬ bers of that large section of the Orchidaceae which is character¬ ized by mealy or powdery pollen, a section recognized by L. C. Richard and recently adopted by Dr. Rudolf Schlechter as the Polychondreae. Together with Pogoniopsis, Monophyllorchis, Xerorchis, Vanilla, Galeola, Eriaxis, Epistephium and Lecanor- chis they constitute the Vanilleae.
If we examine Triphora, Pogonia, Cleistes, Isotria and Psilo¬ chilus with a view to finding key characters to separate them clearly from one another, the following would stand out as the most weighty :
1 Oesterreiche botanische Zeitsclirift 39 (1889) 395-399, 422-430.
2 Journal of Botany 58 (1920) 69-74.
[ 6 1
ORCHID ACE^E
I. Column denticulate or paucidentate at the apex. Anther de¬ cumbent with the loculi facing downward. Pollen grains simple or compound with the extine free from pits or reti¬
culations.
A. Pollen grains simple, not cohering in tetrads . Pogonia
A. Pollen grains compound, cohering in tetrads.
a. Leaves solitary or alternate . Cleistes
a. Leaves (5 or 6) verticillate . Isotria
II. Column entire at the apex or simply lobed. Anther erect, with the loculi facing outward. Pollen grains compound with
the extine pitted or reticulated.
A. Anther rigidly attached to the top of the column . Triphora
A. Anther mobile at the top of the column and articulated
with it . Psi/ochilus
A glance at the key reveals a very important point, namely, that Triphora is set apart, not only by means of an easily determined character found in the pollinia, but by a peculiar structure of the column that does not occur in any of its allies. This character, in my estimation, is indicative of deep-seated differences which are accentuated when we examine the vegetative structure of the plants.
The subterranean organs of the four genera that occur within our range offer some very interesting differences. I believe these were first noticed in detail by Theo. Holm in his paper already referred to. Aside from fundamental differences, the gross vege¬ tative characters offer a means for dividing the alliance under consideration into two groups; Triphora possessing a most re¬ markable system of underground tubers and stolons which will receive attention on another page, Pogonia, Cleistes and Isotria propagating by means of root-shoots or successive buds, never by true stolons. I am unable to include Isotria ajfinis as a plant that propagates vegetatively by root-shoots, as this species is so rare that material is difficult to obtain in sufficient completeness to warrant conclusions regarding its vegetative system. The most
[ 7 ]
ORCHIDACEiE
complete specimens I have studied are clearly shown in the illus¬ tration. (Plate 107.) From this material one is only justified in making assumptions. It would be strange, indeed, if I. qffinis did not produce root-shoots, but conclusive proof that it does do so is still wanting.
Although it is not the purpose of this work to survey the fifty or more species that might properly come within its scope, some mention should be made of the Asiatic species that bears a strik¬ ing resemblance to Pogonia ophioglossoides and is sometimes referred to it. In 1858, when Asa Gray put forth his classic paper which contained observations upon the relations of the Japanese flora to that of North America,1 he included in the list of species common to New England and Japan two orchids, Liparis lilii- folia 2 and Pogonia ophioglossoides. More recent authors have found characters which they regard sufficiently significant to warrant the recognition of the Japanese Pogonia as a distinct species and the name P.japonica Reich b. f. has been accepted for it by the latest monographer of the Sino-Japanese orchid flora. I have examined herbarium material of the Chinese and Japanese Pogonia that has been referred to P.japonica and find that the most striking character is in the labellum. This organ is distinctly three-lobed. In our New England P. ophioglossoides the lip is sometimes obscurely three-lobed, but as a rule tenden¬ cies toward lobing are suppressed or obscured by a copious and deep fringe. Blume figured his P. similis, a species now merged with P. japonica, with a conspicuously three-lobed lip. Except for this lobing of the lip and a slight difference in the habit, and in the color of dried specimens, there is nothing to distinguish the Asiatic from the American species. Whether or not the dif-
1 Memoirs of the American Academy of Arts and Sciences n. s. 6, pt. 2 (1858) 377.
2 The Asiatic Liparis is now referred to L. Makinoana Scliltr.
[ 8 ]
ORCHIDACEiE
ferences that are emphasized will prove diagnostic of two distinct species remains for further consideration. It is true, however, that the aura of these species, widely separated geographically, is dis¬ tinctive.
If, at this point, we review the species of the northeastern United States included within the range covered by the seventh edition of Gray’s Manual, we find that the Pogonia alliance has five representatives. In the following pages these are arranged for discussion in the order that seems best suited to an illumi¬ nating survey of their differentiating characters.
TRIPHORA Nutt.
Triphora trianthophora (Sw.) Rydb. in Britton’s Man. (1901) 298. Helleborine mariana Pluk. Mant. (1700) 100, t. 348, f. 6. Arethusa trianthophora Sw. in K. Vet. Akad. Stockh. Nya Handl. 21 (1800) 230. Arethusa parviflora Michx. FI. Bor. Am. 2 (1803) 160. Triphora pendula Nutt. Gen. 2 (1818) 193. Po¬ gonia pendula Lindl. in Bot. Reg. (1825) t. 908. Pogonia trian¬ thophora BSP. Prelim. Cat. N. Y. (1888) 52.
Triphora trianthophora is the only species of Triphora within the territory covered by Gray’s Manual. In fact, with the excep¬ tion of a species recently discovered in Florida that appears to be closely allied with T. gentianoides (Sw.) Ames & Schltr., it is the only representative of Triphora in the United States. Its nearest ally is the Mexican T. meodcana (Wats.) Schltr. It dif¬ fers from all other New England orchids in its vegetative sys¬ tem, and has been very casually mentioned in botanical writings with regard to the subterranean phenomena that characterize it. Lownes studied the behavior of tubers in colonies under observation near Camp Algonquin, New Hampshire, and pub¬ lished a very illuminating note regarding them in Rhodora for
[9]
ORCHIDACEiE
March 1920. From my reading of Lownes’s article, however, I was unable to decide whether or not his observations agreed with mine; his drawings, diagrammatic though they were, failed to bring out clearly just what happens in the sequence of vege¬ tative multiplication. In the main, however, I believe the con¬ clusions at which he arrived are correct.
When a colony of this species is dissected, a process that de¬ mands much care and extraordinary patience, as the stems and stolons are exasperatingly fragile and brittle, it is found that the flowering stem arises from the broad end of an obconical or egg- shaped tuber, and that from the lower part of the flowering stem, above the tuber, stolons originate and grow out in a more or less horizontal direction, bearing at their end miniature replicas of the mature tuber. (Plates 102 and 103.) The exact significance of these miniature tubers is not at once apparent and perplex¬ ity may arise as to their function. If the mature tuber illus¬ trated (cf. Plate 102) is examined, a minute process will be ob¬ served at the basal end. This process is to be interpreted as the withered remains of a stolon and indicates the original point of attachment of the tuber to the stolon of the mother plant. This process is usually very distinct in fresh material, and yet might be passed over as of no consequence in a study of the life history of the species. If, now, we begin our study of development with a minute tuber that has been detached from a stolon, perhaps by frost or some other mechanical means, we are able to observe that it is at the distal end of this tuber that development takes place, and that the tissues in the apical region give rise first to a new stolon bearing a miniature tuber and then to a bud from which develops the new stem. This method of growth is clearly shown in the illustration (Plate 103) and suggests what may
be the early stages of development from a protocorm. The young
[ 10 ]
ORCHIDACEJE
tubers are provided with endotropic fungi (mycorrhiza), although the sections I have examined indicate that the fungal elements are restricted and are usually present only in the basal half, that is, near the end by which the tuber was originally attached to the stolon.
The flowers of Triphora trianthophora differ fundamentally from those of other New England orchids in their brief duration. The flowers seem to fade rapidly shortly after the sepals and petals expand, and it is very evident that floral perfection is a matter of hours rather than of days. The labellum is a very beau¬ tiful object, the brilliant green raised nerves, that traverse the middle longitudinally, forming a striking contrast with the whit¬ ish disc which seems to be composed of myriads of tiny crystals in piles of unequal density and depth. Shortly after anthesis the margin of the middle lobe becomes strongly involute, an indi¬ cation of the passing of floral perfection. The lip of T. triantho¬ phora is unlike the lip of its near allies, Pogonia and Isotria, and is devoid of calli or glands at the extreme base. In the flower the gynostemium or column is the structure of greatest interest, not only because it exhibits weighty characters for purposes of clas¬ sification, but because it is very unlike the gynostemium of Po¬ gonia, Isotria and Cleistes. At base it is semiterete; near the mid¬ dle it becomes laterally dilated. The clinandrium has a strongly deflexed margin and is flat or slightly rounded above. (Cf. Plate 102, figs. 4-6.) The most remarkable structure, however, is ex¬ hibited by the anther. This is subcylindrical, blunt at the apex and attached to the body of the column by an unjointed, rigid, semiterete base. From the white body of the anther the flanges of the cells that contain the pollinia extend forward. These are of a rich red-purple color, forming a strong contrast with adjoin¬ ing tissues and making the structure comparable, when seen
[ 11 1
ORCHIDACE^E
through a lens, to a ruby set in white enamel. The pollen co¬ heres in masses as shown in the illustration (Plate 102, fig. 9), and is violet or mauve, a very unusual color for this sub¬ stance. The pollen tetrads are characterized by a reticulated or pitted extine. (Cf. Plate 102, fig. 10.) I have observed germina¬ tion of the pollen tetrads in situ. The stigma is extraordinarily long. In fruit the perianth often persists and sometimes is to be observed when the capsules are mature and at the point of dehiscence.
Ttiphora trianthophora is normally a woodland species occur¬ ring in colonies. It exhibits extraordinary periodicity, and in a given locality may be plentiful in one year and then rare or absent for one or more years. Undoubtedly this peculiarity of periodicity is in some way bound up with the phenomena of vegetative reproduction referred to above. My supposition has been that there is a maximum size for the tubers that bearflower¬ ing stems, and that when the maximum is reached the tuber dies; or there may be a period during which stolon production by immature stems overbalances flower production. From the plate (Plate 103), in which the development of small tubers is illustrated, it will be seen that the formation of tubers capable of florification must require at least a growing season, and that a condition might exist in which numerous tubers approached the flowering age without any being present that were ready to produce flower-bearing stems, the mature tubers of preceding years having completed their period of activity.
CANADA
Douglas fide Lindley Genera and Species of Orchidaceous Plants p. 413. MAINE
Brownfield, LeRoy Harris Harvey, 1899.
[ 12 ]
ORCHID ACEiE
NEW HAMPSHIRE
Wilton, Gray 505. Meredith, Edward H. Fogg, August 20, 1904; F. W. Batchelder, August 12, 1895. Bartlett, W. L. W., August 18, 1879. Chocorua, W. G. Farlow, 1904. Tubers hidden among decaying beech leaves; August 1912; September 1913. Intervale, Miss H. E. Freeman, August 1905; Miss Frances C. Prince; C. A. Hawes, August 18, 1901. Wiers, F. W. Batchelder, August 18, 1901. Asquam Lake, A. E. Lownes, August 2, 1917. In beech woods.
VERMONT
Newfane, Howe, August 30, 1897. Westminster, W. H. Blanchard, Au¬ gust 16, 1899. Putney, Blanchard, August 24, 1899.
CONNECTICUT
Granby, C. H. Bissell, August 12, 1904. Rich woodland hillside growing in moist humus; /. Holcomb, August 8, 1903. In moist woods; A. W. Driggs, August 23, 1901. New Haven, Specimen from Herbarium D. C. Eaton (in Herb. Gray).
NEW YORK
Conquest, L. Griscom 8? F. P. Metcalf 6863, August 12, 1916. Rich soil in Botrychium woods, Spring Lake. Syracuse, Miss E. G. Webster, August 1902. Vaughn, S. H. Burnham, August 1903. Olive, H. M. Denslow, Sep¬ tember 5, 1920. In low woods. Western New York, Sartwell.
NEW JERSEY
Closter, C. F. Austin. Fort Lee, William Bower.
PENNSYLVANIA
Chester County, J. Walton , Jr.
MARYLAND
Bradley Hills Park, E. T. Wherry , August 29, 1920. Herzog Island, E. T. Wherry <§■ H. C. Sheets, August 12, 1916. Low bank of Potomac above the island. Plummer’s Island, C. L. Pollard, September 22, 1903. Great Falls, Street car terminal, H. C. Sheets, August 28, 1916.
VIRGINIA
Bedford County, A. H. Curtiss, August 1873.
NORTH CAROLINA
Smoky Mountains, H. C. Beardslee <§* C. A. Kofoid, August 18, 1891. Rich damp woods. Mount Mitchell, August 7, 1897. Specimen distributed by
[ 13 ]
ORCHID ACEJE
Biltmore Herbarium (no. 781 b). Mount Sterling, A Ibert Ruth IfiO, Au¬ gust 20, 1897.
SOUTH CAROLINA
Santee Canal, H. W. Ravenel, September. Damp shady woods.
FLORIDA
ALABAMA
Peters.
TENNESSEE
Burbank, Roland Thaxter, September 1887. Wilkinson.
INDIANA
Whiting’s, E. J. Hill, August 6, 1886.
ILLINOIS
E. Hall. Urbana, II. A. Gleason, September 1902. Rich woods. Chicago. Moffatt, September 4, 1897. Tamarac swamp at the head of Lake Michi¬ gan. Athens, E. Hall, 1865.
MICHIGAN
Clarksville, I. W. Stacey, September 1901.
WISCONSIN
Madison, William Trelease, August 13, 1885.
MISSOURI
Dunklin County, B. T. Bush 131, September 17, 1893. Courtney, B. T. Bush 1788, 1877, September 18, 1903. Rare in rich woods. St. Louis, H. Eggert, August 5-September 14, 1892. Damp woods.
POGONIA Juss.
Pogonia ophioglossoides ( L .) Ker-Gawl. in Bot. Reg. (1816) t. 148. Arethusa ophioglossoides L. Sp. PI. 2 (1753) 951.
Pogonia ophioglossoides is too well known to deserve lengthy attention, but its means of vegetative reproduction differ so markedly from those of Triphora trianthophora that a detailed description is quite necessary for the purposes of this paper. In this species the rhizome is normally vertical, inconspicuous, and buried from view in deep sphagnum or boggy soil. From the
[ 14 ]
ORCHIDACEtE
vertical rhizome, root-shoots (cf. Plate 104) extend widely in a nearly horizontal plane and eventually produce new plants. It will be understood at once that in multiplication by means of root-shoots Pogonia ophioglossoides is fundamentally different from Triphora trianthophora which multiplies by means of true stolons that terminate in tubers.
The secondary stems of P. ophioglossoides are usually described as monophyllous, although the bract subtending the flower may be foliaceous and properly regarded as a second leaf. If two or more flowers are produced, an occurrence which is not at all rare in the Southern States, each flower is subtended by a bract.
The flower differs from that of Triphora trianthophora in du¬ ration and may remain in perfection for several days. At the base of the labellum there are two glands or calli, structures which are present in the tropical species of Cleistes and in Isot?ia verticil- lata and I. qffinis. (Cf. Plates 104, 105 and 107.) These glands, as has already been stated, are wanting in Triphora which has a very different lip-base, and their absence is still another indica¬ tion of generic differentiation between this genus and Pogonia. The column terminates in a deep clinandrium or excavation, and the anther is hinged in such a way that it is mobile and when in normal position hangs with the pollen cells looking at the floor of the clinandrium, not outward as in Triphora and Psilochilus. The margin of the clinandrium is toothed. In the structure of the an¬ ther, we find one of the most salient generic characters on which to rely in removing Pogonia from Triphora. The pollen of Po¬ gonia, as is shown in the illustration (Plate 104), differs funda¬ mentally from the pollen of Cleistes and Isotria. The grains con¬ sist of single cells held together by some viscid substance. I have been very much impressed by the constancy of this character in
all of the material I have examined, taken from the extremes of
[ 15 ]
ORCHIDACEdE
distribution. Furthermore I have observed that the tendency of the pollen of Pogonia to germinate in situ is very common, and pollen grains that have germinated within the anther may be ob¬ served even in dried material, when the grains are often in masses held together by pollen tubes. I have observed this phenomenon in Asiatic species as well as in our own P. ophioglossoides.
There is still another difference in addition to those described that serves to indicate generic distinction between Pogonia and Triphora; I refer to the behavior of the perianth after anthesis. In Pogonia, Isotria and Cleistes the perianth falls away shortly after the flowering season and is wanting at the time of dehis¬ cence of the capsule. In Triphora the perianth persists and may often be found on fruiting specimens at the time dehiscence takes place.
Canada to Florida westward to Minnesota.
ISOTRIA Raf.
Isotria verticillata ( Willd .) Raf. in Med. Repos. N. Y. 5 (1808) 357. Helleborine affinis, Planta Mariana, Herbae Para¬ dis facie quinquefoliata. Pluk. Mant. (1700) 101, t. 348. Arethusa verticillata Willd. Sp. PL 4 (1805) 81. Odonectis verticillata Raf. in Med. Repos. 5 (1808)357. Pogonia verticillata Nutt. Gen. 2 (1818) 192. Arethusa medeoloides Pursh Flora 2 (1814) 591. Iso¬ tria medeoloides Raf. FI. Tellur. 4 (1838) 47.
If Isotria verticillata is studied in the field in late April orearly May, it will be observed that the flower-bud is just pushing through the protective covering formed by a verticil of closely imbricating leaves. (Cf. Plate 105, fig. 9. ) The leaves at this stage of development closely resemble imbricating bracts and present a condition found in no other species, I. affinis excepted, of the Pogonia complex.
[ 16 ]
ORCHIDACEJE
Isotria verticillcita, lik ePogonia ophioglossoides, propagates by means of root-shoots. It has a vertical rhizome covered with fus¬ cous pubescence. The root-shoots, in this species, may attain an extraordinary length. (Cf. Plate 106.) The secondary stems are tubular with a few basal bracts and are terminated by a verticil of from five to six leaves, which at the time of anthesis are very slightly developed and only attain full size when the flowering season is drawing to a close. In the verticil two leaves are super¬ posed. In seedlings (cf. Plate 106) of this species the earlier stages are characterized by two, three or four leaves, and it would appear that vegetative maturity is not attained in the first two or three years of growth.
The attitude of the flower is characterized by several positions. At first the sepals, petals and lip point toward the zenith. This attitude owes its occurrence to the peculiar type of leaf verna¬ tion already mentioned. At this stage the upper sepal is bent sharply backward and the leaves are comparatively small, almost bract-like, but as the flower matures and the sepals and petals point toward the horizon, they develop rapidly and in a sur¬ prisingly brief time increase to their normal size. In a few days the flowers begin to droop, and if pollination has been effected and fertilization secured, the perianth falls away and the capsule assumes a vertical position ; if pollination has not resulted in fer¬ tilization, the perianth together with the pedicel and ovary drops to the ground.
Germination of the pollen before transference to the stigma has been observed in this species, but that self-fertilization is effected by pollen tubes reaching the stigma while the grains are still in situ has not been demonstrated. In Plate 105 a germi¬ nating tetrad is shown.
Isotria verticillata is a woodland species and is often associated
[ n ]
ORCHIDACEiE
with Medeola virginiana, a species which at certain stages of growth it resembles so closely, that references to the fact are very numerous. This association is observed not only in the New England States, but as far south as Virginia, and has been re¬ ported by Prof. Earle Jerome Grimes from Williamsburg, Vir¬ ginia, where Isotria verticillata prefers soils that are medium acid to methyl red.
If the flower of I. verticillata is compared in detail with a flower of Pogonia ophioglossoides, it will be found that they dif¬ fer markedly in the segments of the perianth. In the former the front of the clinandrium is smooth, in the latter it is conspicu¬ ously toothed. The differences that may be noticed in the lobing of the lip, in the nature and structure of the keel and in minor structures are specific rather than generic, therefore the weight¬ iest character aside from pollen structure, on which to rely for generic distinction from Pogonia, is the verticillate leaves and the behavior of these in vernation.
Westward from New England to Wisconsin and Indiana, southward to
Florida. Very much localized where found.
Isotria affinis ( Aust .) Rydb. in Britton’s Man. (1901) 297. Pogonia affinis Aust. ex Gray Man. ed. 5 (1867) 507. Odonectis affinis Schltr. in Engl. Bot. Jahrb. 45 (1911) 386.
Isotria affinis, ever since its discovery by C. F. Austin near Closter, New Jersey, has remained a rare species and is very spar¬ ingly represented in herbaria. When met with, the plants are reported as being in very much restricted areas where diligent search is rewarded by the discovery of very few specimens. In June 1920, the range of the species was extended southward to Virginia by Prof. Earle Jerome Grimes, who discovered a colony one and one-half miles southwest of Williamsburg. Prof. Grimes describes the locality where his specimens were observed as a flat,
[ 18 ]
ORCHID ACE,®
dry hardwood on a gently undulating interstream area, the soil well drained, composed of a gray, fine sandy loam over a yel¬ lowish sandy-clay or clay. The woodland consisted of white oak, beech, tulip and chestnut trees, with a few scattering loblolly pines and an abundance of flowering dogwood. Here the under¬ growth was sparse and the Isotria specimens were scattered over an area of about ten acres, occurring either singly or in open groups of from two to four plants. A composite soil sample of the habitat was tested and found practically neutral to Brom Thymol Blue, and soil shaken from the roots gave approximately the same reaction.
When we attempt to assemble data relative to Isotria affinis , we find ourselves in a well-nigh hopeless situation. In our largest herbaria there are but few sheets of this species; sometimes only a photograph rewards a search for specimens. When material is found, the plants are usually incomplete and inadequate for a thoroughgoing study, and examination of the flowers is precluded by the likelihood of damage being done when dissections are attempted. Now and again new stations are reported that it is hoped will relieve the situation, but investigation usually proves that some overzealous novice has confused Isotria verticillata with /. affinis. Those who have actually found I. affinis are among the favored few.
The following notes on the floral structure of Isotria ajfinis were taken from Virginian specimens supplied by Prof. Grimes. (Plate 107.) Lip three-lobed as in I. verticillata, but the nerves of the lateral lobes irregularly thickened, resembling somewhat the streams of tallow that flow down the sides of a lighted candle. The elevated callus that extends from the base to about the middle of the disc is widely and shallowly grooved, and at the tip terminates in blunt, elongated, wart-like processes that stand
[ 19 ]
ORCHIDACE^E
erect on the middle nerves. The upper half of the middle lobe is free from calli. The most apparent difference between I. qffinis and I. verticillata is found in the sepals. These are not elongated and resemble very closely the sepals of Pogonia ophioglossoides.
The plate of Isotria affinis (Plate 107) shows two specimens the roots of which deserve close attention. One of the plants is apparently seven years old, with a root system that exhibits very little similarity to that of I. verticillata. The other specimen is a younger plant and appears to be blooming for the second time. The roots of this specimen were intact, each one being provided with a growing tip. Although in such limited material it would be indiscreet to attempt definitive characterizations, it is well worth while to suggest that the development of root-shoots is either rare or wanting in Isotria affinis. Prof. Grimes’s account of the distribution of the plants he observed would lead one to believe that this species does not exhibit vegetative peculiarities that are comparable to those that characterize I. verticillata. An examination of the roots revealed the presence of mycorrhizal fungi in the cortical cells.
It is noteworthy that Isotria affinis frequently bears two flowers. It is not at all rare to find two capsules terminating the stems from the base of which a single flowered growth has sprung.
VERMONT
Burlington, Mrs. Henry Holt , June 1901.
MASSACHUSETTS
Mount Holyoke Range, East Hadley, H. L. Clark, May 31, 1899. CONNECTICUT
New Haven, Edit). Dana. Ledyard, B. J. Avery , Jr., June 3, 1909. Rather heavy, moist soil in woods. Stratford, C. K. Averill, October 6, 1891 (in fruit). In cool moist woods under beeches; E. II. Eames <$f C. K. Averill, June 3, 1892. In cool moist woods under beeches, all aerial parts glaucous
[ 20 ]
ORCHIDACEdE
in this species. Root system deep, distinct. Waterford, Mrs. F. S. But¬ ton, 1913. Small clearing in rich woods with Cypripedium acaule; Brown’s Meadow, Miss Florence Griswold, June 1, 1913.
NEW YORK
Long Island, Hempstead, Miss Harriet Midford, May 25, 1918.
NEW JERSEY
Closter, C. F. Austin, 1858. This station exterminated by 1873. Re¬ ported from Norwood and Trenton.
PENNSYLVANIA
Westtown, Edith Cheney, June 1, 1887. Reported from Monroe County, W ater Gap, and from Berks and Philadelphia Counties.
VIRGINIA
Williamsburg, E. Jerome Grimes 2637 , June 1, 1920. In high flat, dry open oak woods, in sandy soil; Grimes 3555, May 9, 1921. Flat white oak woods. Sandy soil and neutral.
CLEISTES L. C. Rich.
Cleistes divaricata ( L .) Ames comb. nov. Arethusa divari- cata L. Sp. PL 2 (1753) 951. Pogonia divaricata R. Br. in Ait. Hort. Kew ed. 2, 5 (1813) 203.
The last species for consideration is the Pogonia divaricata of the Manual. This is southern in its distribution, but reaches as far north as the pine barrens of New Jersey. It agrees vegeta- tively with Pogonia and Isotria, but seems to have more in com¬ mon with the neotropical species that have been referred to Cleistes, and should, if my views are adopted, be included in that genus. Although it has been placed in Pogonia by all mod¬ ern authors, it would seem that the characters emphasized in the key remove it clearly from that genus.
It is worthy of note that Isotria antedates Cleistes, and that should Cleistes not be taken up, the neotropical species that are not members of Triphora or Psilochilus would properly become members of the genus Isotria. (Plate 108.)
[ 21 ]
ORCHID ACELE
NEW JERSEY
Quaker Bridge, William H. Leggett, July 7, 1864.
DELAWARE
Ei.lendale, Bayard Long, June 28, 1909.
NORTH CAROLINA
Table Rock Mountain, J. K. Small A. A. Heller 285, July 2, 1891. Burke, M. A. Curtiss. Hickory, A. A. Heller, June 23, 1893. 1800 feet altitude. Hot Springs, J. R. Churchill, June 9, 1899.
SOUTH CAROLINA
Santee Canal, H. W. Ravenel, May. Summerville, Dr. Janet Perkins, April 29-May 10, 1918; Savannas, May 24, 1853. (Ex Herb. A. C. Hexamer & F. H. Maier.) Bluffton, W. H. Mellicliamp, 1886.
GEORGIA
Thomasville, Mrs. Augustus P. Taylor, May 1902. Bogs and swamps; Waite A Vasey, June 1889. Bullock, R. M. Harper 883, June 11, 1901.
ALABAMA
Mobile County, W. S. Dukes, May 1905.
FLORIDA
Apalachicola, May 1896. Distribution of duplicates of the Chapman Herbarium (no. 780 b). Palatka, William M. Canby, April 1869. Arling¬ ton River, G. Brown Goode, Summer, 1872. Jacksonville, A. H. Curtiss 2796, May. South Jacksonville, J. R. Churchill, April 17, 1897.
Summary
If, now, we sum up the salient characters relied on for the con¬ clusions arrived at in the foregoing pages, we find, with regard to the species of the northeastern United States, that the pol- linia furnish characters of great importance and give three groups of species. Pogonia is first set aside by means of simple pollen grains. The remaining groups have the pollen cohering in tetrads. Triphora comes out because of the reticulated pollen grains and rigidly attached anther, and is sharply differentiated from its allies by means of its vegetative peculiarities and differ-
[ 22 ]
ORCHIDACE/E
ent lip-base. Isotria and Cleistes, characterized in part by the unpitted extine of the pollen, remain. They are separated from each other by means of the foliage, Isotria having five or six leaves disposed in verticils, while in Cleistes the leaves are soli¬ tary or alternate. The verticils of leaves in Isotria are of extraor¬ dinary interest as they indicate a line of development that now lacks a parallel in the Orchidaceae, the nearest approach to it being found in the Antarctic Codonorchis. It is when we con¬ sider the significance of verticillate leaves in the evolution of the orchids that this peculiarity in Isotria assumes deep signifi¬ cance in attempts at generic segregation.
[ 23 ]
.
PLATE 102
ORCHID ACEiE
Plate 102: Triphora trianthophora 1, plant in bud, natural size, drawn from a specimen collected by Irving Holcomb, near Granby, Connecticut, in August 1920. 2, flower ( X 2) . 3, flowers, natural size from plant shown in f. 1. 4, upper part of column ( x 8|), pollinia removed. 5, upper part of column, viewed from the side ( x 8J). 6, column, drawn to exhibit semiterete base, sunken stigma below the de- flexed clinandrium and anther in normal posi¬ tion ( X 6f). 7, labellum, lateral lobes spread out. 8, labellum cut longitudinally in half, one keel removed. 9, pollinia ( x 13). 10, pollen te¬ trad highly magnified, median section, to show pitted extine. 11, capsule and persistent peri¬ anth, drawn natural size. 12, seed highly mag¬ nified.
[ 26 ]
PLATE 103
ORCHIDACEdE
Plate 103: Triphora trianthophora 1 and 2, fruiting specimens, drawn natural size, from specimens collected by Irving Hol¬ comb, near Granby, Connecticut, in October 1921. 3, corm and stolons, drawn natural size. 4, young tubers ( X 2) showing original point of attachment to a stolon, at the basal end, and illustrating the development of a stolon and minute tuber (lower left), and the formation of a bud from which a stem is about to arise (upper centre). 5, a later stage of development than that shown in f. 4; the stolon has elongated and the tuber has become enlarged. 6, longitudinal section through the middle of the uppermost tuber shown in f. 5 (semi-diagramatic).
[ 28 ]
PLATE 104
ORCHIDACEdE
Plate 104: Pogonia ophioglossoides 1, plant natural size, rooting base shown by f. 10, drawn from a specimen collected in North Easton, Massachusetts, in June 1921. 2, flower, natural size, from plant shown in f. 1, viewed from the side. 3, section taken from lip and col¬ umn, made longitudinally through the centre, to show the structure of the median keel, and the position of the anther in relation to the clinan- drium (the projection above at base of column represents the base of the upper sepal) ( X 2| ). 4, base of lip, drawn to show glands, one on each side, at the base of the lip( X 5|). 5, column with anther in normal position ( x 3f). 6, up¬ per part of column, anther turned back ( x 6f ). 7, ten pollen grains highly magnified. 8, seed highly magnified. 9, capsule, drawn natural size. 10, base of stem, showing vertical rhi¬ zome, roots and a root-shoot at the end of which may be seen a bud from which a new plant is about to develop (natural size).
[ 30 ]
V
—
PLATE 105
ORCHIDACEdE
Plate 105: Isotria verticillata 1, base of lip, drawn to show the glands, one on each side ( X 7J). 2, column, showing papil¬ lose stigma and the anther in normal position ( X 3). 3, a stellate trichome highly magnified, usually abundant on the lip in fresh specimens, but easily deciduous. 4, upper part of column with the anther turned back and the mealy pollen exposed. 5, pollen tetrad highly magni¬ fied, found germinating in situ. 6, pollen tetrad before germination, highly magnified. 7, label- lum spread flat, drawn to show the lobulate keel and lateral lobes ( x 2|). 8, pollen tetrad before germination, highly magnified. 9, flower bud invested by the verticil of leaves, drawn natural size. 10, labellum seen from the side ( X 2|). 11, flower and leaves, natural size, drawn from a specimen collected near North Easton, Massachusetts, in May 1920, upper sepal in position characteristic of floral maturity. (In younger stages of development the upper sepal is bent sharply backward.)
[ 32 ]
|
PLATE 106 |
|||
ORCHIDACEdE
Plate 106: Isotria verticillata 1 and 5, fruiting specimen, drawn natural size from a specimen collected near North Easton, Massachusetts, in 1920. 2, transverse section of capsule, magnified. 3, a seedling with complete root system, showing the stage of development characterized by three leaves, drawn natural size. 4, seed highly magnified. 5, lower part of stem, showing vertical rhizome with winter bud near the summit, several roots, and elongated root-shoot bearing the remains of a rooted rhi¬ zome and the winter bud and lower part of the stem of another plant, all drawn natural size.
[ 34 ]
a;
o P/-106
1SOT R,I A
vert ic i
i& d)
ORCHIDACEJE
Plate 107 : Isotria affinis 1, 2, 11 and 12, plant and root system, drawn natural size from material collected by Earle Jerome Grimes, near Williamsburg, Virginia, in May 1921. 3, lateral sepal ( x 3). 4, upper sepal ( X 3). 5, petal ( x 3). 6, lip, one half removed, drawn to show calli ( X 4). 7, lip spread out, drawn to show lateral lobes and calli ( X 4). 8, summit of column showing stigma, anther turned back, pollen removed. 9, column with anther in normal position. 10, pollen tetrad highly magnified.
[ 36 ]
PLATE 108
ORCHIDACEiE
Plate 108 : Cleistes divaricata 1, 2 and 3, plant natural size, drawn from spe¬ cimens collected by Mrs. Augustus P. Taylor, near Thomasville, Georgia. 4, petal, natural size. 5, lip spread out, drawn natural size. (In a series of specimens the margin of the trian¬ gular apical part of the lip, and the terminal part of the crest exhibit a wide range of varia¬ tion.) 7, rhizome and roots, drawn natural size.
[ 38 ]
CLUSTES
divcLriocctcL (,C-J
ist m-es
.
NOTES ON MEXICAN SPECIES OF TRIPHORA
Triphora yucatanensis Ames sp. nov. Herba terrestris tenera plus minusve 10 cm. alta, caudice subterraneo simplice scapo continuo perpendiculari e tuberi carnoso, ovoideo enato. Tuber intra terram alte sepultum, potius parvum, plus minusve 1.5 cm. longum. Caulis erectus, in speciminibus nostris tri- ad quadrifo- liatus, glaber, interne squamis binis ternisve brevibus cinctus, internodiis 1.5 cm. longis. Folia 1-1.7 cm. longa, 5-15 mm. lata, inferiora late ovata, acuta, superiora lanceolata, acuta, membra- nacea, glabra, in sicco atroviridia, sessilia, amplexicaulia. Flores in apice scapi plerumque duo vel tres, alterni, raro solitarii, pedi- cellati, ut videtur erecti, vix nutantes, pallide rosei, 9 mm. longi, vix contemporanei. Bracteae inflorescentiae foliaceae, plus mi¬ nusve 5 mm. longae, ovato-lanceolatae, patentes. Ovarium line- ari-fusiforme, rectum, obtuse trigonum. Sepala lateralia 8 mm. longa, trans medium 1.5 mm. lata, oblonga, subacuta, ad apicem minute cucullata, prope apicem valde incrassata, extus per me¬ dium usque ad apicem conspicue carinata, subpatentia. Sepalum dorsale 9 mm. longum, valde concavum, cymbiforme, ad apicem minute cucullatum, extus per medium usque ad apicem pro- minenter carinatum. Petala membranacea plus minusve 8.5 mm. longa, trans medium 1 mm. lata, oblonga, apice rotundato- obtusa, trinervia, erecta. Labellum trilobatum, 7 mm. longum, in basi gynostemii sine articulatione insertum, basi in unguem oblongum vix cuneatum le viter concavum angustatum, disco tricarinato e basi usque ad medium, carinis parallelis, carina in¬ termedia multo reducta et abbreviata; lobi laterales 1 mm. lati, rotundati; lobus medius vix 3 mm. longus, plus minusve 2 mm.
latus, oblongus, margine lacerato-dentato, ad apicem leviter re-
[39]
ORCHIDACE^E
tusus; discus muriculis sparsis ornatus. Columna omnino 6 mm. longa, clavata, sectioni propria. Anthera rigida, erecta, apiculata, 1.75 mm. longa. Massae polliniae ut in T. trianthophora.
The new species of Triphora described above was collected in Yucatan by G. F. Gaumer. The type preserved in the her¬ barium of the Field Museum of Natural History is composed of two plants, one in fruit, the other in flower. Originally this material was identified as Pogonia mexicana Wats. var. (a spe¬ cies now referred to Triphora), on the authority of Dr. Rudolf Schlechter of Berlin. From T. mexicana (Wats.) Schltr. the Yucatan species is widely separated by the oblong, denticulate middle lobe of the labellum and by the complanate-muriciform appendages that occur sparsely on the three middle nerves above the basal keels. The type of T. mexicana has a differently formed labellum, the middle lobe being cordate or subdeltoid rather than oblong and the claw more cuneate, while the ecarinate part of the disc-nerves lacks muriciform appendages. It is true that T. meocicana closely resembles T. yucatanensis and might readily be mistaken for it in the absence of flowers. (Plate 109.)
Yucatan, sine loco exactiore, Gaumer 1008. Floret Novembri.
Triphora mexicana {Wats.) Schltr. in Fedde Repert. 17 (1921) 139. Pogonia mexicana Wats, in Proc. Am. Acad. 26 (1891)154.
Pogonia mexicana Wats., now referred to Triphora by Schlech¬ ter, was originally described from very poor material gathered by Pringle in the State of San Luis Potosi, Mexico. The type consists of four plants, none of which is satisfactory for critical examination. Indeed, it would seem that Pringle collected his specimens at a time when the flowers were passing their prime. The only clue to the floral characters is contained in a carefully
[ 40 ]
ORCHIDACEJE
preserved labellum that was mounted on a small square of stiff paper and protected by a coating of some varnish-like substance. It is presumable that this labellum was saved by Watson from among his dissections at the time the species was described. The type was collected in August 1890 in the Tamasopo Canon, and in the following year from the same locality Pringle secured ma¬ terial which was in full flower in July.1 Although the plants of the second collection are taller and more robust than the plants of the first collection, they seem to represent Triphora meocicana and to furnish evidence to the effect that the species when well developed is very similar in habit to the northern Triphora tri- anthophora.
The labellum of Triphora meocicana as exhibited by the type seems to have had three green keels that extend to about the middle, the central one exceeding the lateral ones in length. There appear to be no flattened papillae on the central nerves, but in other Mexican material that I have examined, presum¬ ably referable to T. meocicana, the middle nerves break up into a few complanate processes and vanish toward the lower part of the middle lobe in tiny complanate emergences and minute pa¬ pillae that extend almost to the apex of the middle lobe. If such emergences occurred on the labellum of the type, it is plausible to suppose that they were obliterated by the fusing of the tis¬ sues under pressure and by the action of the preservative coating already referred to. The lip is 1 cm. long, the middle lobe 4 mm. wide across the broad base ; the lateral lobes are 1 mm. wide. In moistened material of specimens that I have identified as T. meocicana , the measurements slightly exceed those given for the type, but if due allowance is made for shrinkage in dried
1 It would seem that specimens of this second collection were distributed with labels that gave the name as Pogonia mexicana Wats. n. sp.
[ 41 ]
ORCHIDACEdS
material, the discrepancies noted are mechanical rather than botanical and should not be taken as seriously affecting the situation. (Plate 109.)
Mexico, State of San Luis Potosi, Tamasopo Canon, C. G. Pringle 3557 , August 5, 1890; Pringle 3787 , July 17, 1891, from the same locality is also referred to the species.
[ « ]
I
/
PLATE 109
ORCHIDACEjE
Plate 109
Triphora yucatanensis. 1, flower enlarged. 2 and 3, plant natural size, drawn from the type pre¬ served in the Herbarium of the Field Museum of Natural History, Chicago, Illinois. 4, label- lum enlarged, drawn to show keels and appen¬ dages.
Triphora mexicana. 5,labellum enlarged, drawn from the type preserved in the Gray Herba¬ rium, Harvard University. 6, plant natural size, drawn from the type.
[ 44 ]
'2$uts-J Sc Afar
TRIPHORA y uoa-6 anensi/s c- Aimes
.
A TRIANDROUS FORM OF PSILOCHILUS MACROPHYLLUS
Psilochilus macrophyllus (Lindl.) Ames comb. nov. Pogonia macrophylla Lindl. in Ann. & Mag. Nat. Hist. ser. 3, 1 (1858) 335. Pogonia physurifolia Reichb.f. in Nederl. Kruidk. Arch. 4 (1858) 324.
Among the specimens of Psilochilus macrophyllus from Gre¬ nada, B. W. I. a single plant was found in which the gynoste- mium was triandrous in a very interesting way, the supplemen¬ tary anthers being lateral, one on each side near the base of the stigma. These supplementary anthers probably represent nor¬ mally suppressed members of the outer androecial whorl, which often occur as staminodia in species of Orchis, Habenaria, etc. When in natural position the anther cells face inward toward the stigma and conceal it. Each anther is supplied by a vascular strand that continues into it as a median nerve. The pollen masses taken from the supplementary anthers consist of tetrads that appear to be quite normal, the extine exhibiting the retic¬ ulations or pits found in the pollen of normal anther cells of the species. Aside from the abnormal condition of the gyno- stemium the flowers examined appeared to be perfectly normal.
As far as I have been able to ascertain, a triandrous condition in Psilochilus macrophyllus has not previously been recorded. (Plate 110.)
Grenada, B. W. I., Azima Mountain, W. E. Broadivay, November 18,
1905. Woods, in damp flat ground. (Herb. Ames 7724.)
[ 45 ]
«
'
PLATE 110
ORCHIDACEJE
Plate 110: Psilochilus macrophyllus Plant natural size, drawn from a specimen pre¬ served in the Herbarium of Oakes Ames, col¬ lected by W. E. Broadway, in Grenada, British West Indies. 1, flower ( x 2). 2, column drawn to show supplementary anthers ( x 4). 3, lip spread out ( X 4). 4, supplementary anthers. 5, pollen tetrad showing pitted extine.
[ « ]
J°U 0
STUDIES OF OTOSTYLIS BRACHYSTALIX
AND THE SPECIES WITH WHICH IT HAS BEEN CONFUSED
Otostylis brachystalix (. Reichb.f ) Schltr. in Orchis 12 (1918) 39, 40, t. 5. Zygopetalum brachystalix Reichb.f. in Walp. Ann. 6 (1861) 660. Cyrtopodium cristatum Griseb. FI. Brit. W. Ind. (1864) 630, non Lindl. Cyrtopodium Grisebachii Rolfe ex Patter in Orch. Rev. 3 (1895) 276. Cyrtopodium cristatum Cogn.in Urb. Symb. Antill. 6 (1910) 579 in part as to Broadway 2336, exclude Broadway 2343. Aganisia brachystalix Rolfe in Orch. Rev. 22 (1914) 200. Koellensteinia brachystalix Schltr. in Orchis 9 (1915) 31, 32.
Otostylis brachystalix has had a remarkable history and has given rise to an extraordinary synonymy. It was originally col¬ lected in the Aripo Savannah of Trinidad and described by H. G. Reichenbach in 1861 as Zygopetalum brachystalix. In recent times it has been confused with Cyrtopodium cristatum Lindl., a species originally collected by Schomburgk in British Guiana. In fact Cogniaux in Symbolae Antillanae among the specimens cited under Cyrtopodium cristatum included number 2336 of W. E. Broadway’s collections, which is referable to Otostylis brachy¬ stalix.
In 1895, in a brief account of Trinidad orchids, Patter referred to a very beautiful white flowered plant with a lemon yellow crest on the lip under the name Cyrtopodium Grisebachii Rolfe. It would seem that this name was never properly published by Rolfe and depends wholly on Patter’s citation. In 1914 Rolfe studied material of this species from Trinidad and detected its affinity with orchids of the Zygopetaleae. He suggested that the species represented Zygopetalum brachystalix Reichb. f. As the charac¬ ters are not in conformity with Zygopetalum as now interpreted,
[ 49 ]
ORCHIDACE^E
Rolfe referred the species to Aganisia and made the combination Aganisia brachystalicc. At this point in the history of the species Rudolf Schlechter of Berlin studied it and concluded that it belonged to Koellensteinia. He made the new combination K. brachystalicc. This was in 1915. In 1918 Schlechter’s treatment of Aganisia and allied genera appeared and Koellensteinia bra¬ chystalicc was therein referred to a new genus under the name Otostylis brachystalicc.
This, briefly, is the history of the Trinidad plant that has passed as a white flowered form of Cyrtopodium cristatum Lindl.
Stems very short, about 2 cm. long, somewhat obpyriform, con¬ cealed by long bracts and by the sheathing bases of the equitant rather stiff and leathery leaves. Leaves four or more, up to 7 dm. long, 1-4 cm. wide, ligulate-lanceolate, acuminate, acute, taper¬ ing gradually toward the base and passing by imperceptible gra¬ dation into elongated, winged petioles that sheath the short stems, prominently three-nerved beneath, many nerved, obliquely as¬ cending, erect in the new growths. Scape wand-like, from the base of a new growth of closely equitant leaves and flowering be¬ fore the growth is completed, provided below the raceme with several tubular bracts that are about 1 cm. long and at intervals of 1 dm. more or less. Raceme simple, strict, up to 2 dm. long, bearing fifteen or more showy white flowers. Bracts of the in¬ florescence shorter than the pedicels, 5 mm. long, strongly con¬ cave, lanceolate, acute. Pedicel with the ovary about 16 mm. long, ascending obliquely. Flowers about 2.5 cm. across when spread out. Lateral sepals elliptical, 1.3 cm. long, 8 mm. wide, rather fleshy, main nerves seven in number. Upper sepal similar. Petals 1.2 cm. long, 6 mm. wide near the middle, spathulate, obtuse, main nerves five. Labellum 9 mm. long, inserted under the tip
of the foot of the column, strongly concave at the base, three-
[ 50 ]
ORCHID ACEdEC
lobed, the lateral lobes reduced to auricles, middle lobe 7 mm. long, about equally broad, suborbicular, apiculate, margin entire. Between the lateral lobes and uniting them in such a manner as to form a deep pocket in front of the column, there is a complex carunculated crest which passes into three short prominent keels at the base of the middle lobe and follows the course of the three middle nerves, otherwise the disc of the labellum is naked. Col¬ umn 5 mm. long to the base of the upper sepal, produced into a fleshy foot about 3 mm. long, broadly bialate at the summit, the wings 2.5 mm. long by 2 mm. wide; rostellum produced into an awl-shaped deflexed process. Anther two-celled, attached to the fleshy, triangular termination of the back of the column. Pollen masses two, in dried material readily disintegrating into tetrads; viscid disc elliptical. (Pi. ate 111.)
Trinidad, B. W. I., Aripo Savannah, W. E. Broadway 2 336 , April 16, 1908. Flowers white. Grows under the shade of trees on the outskirts of the Savannah. A very beautiful terrestrial. No pseudobulbs; Britton, Broad¬ way Sf Hazen 295, March 5, 1920. Grassy plains.
My studies of Otostylis brachystalioc led to a close examination of the plant that has passed unchallenged as Cyrtopodium cris- tatum Lindl. among Trinidad species. It is a true Cyrtopodium characterized by small flowers that are usually borne in simple racemes. When this species is compared closely with the type of Cyrtopodium crist atum collected in British Guiana by Schom- burgk it becomes very apparent that the Trinidad species is quite distinct from it. As far as I have been able to ascertain it is a new species and for it I propose the name Cyrtopodium Jdroad- wayi.
Cyrtopodium Broadwayi Ames sp. nov. Pseudobulbi fusi- formi-subcylindracei foliis coriaceis, linearibus, acutissimis. Sca- pus satis gracilis, plus minusve ramosus vel simplex, superne
[ 51 ]
ORCHIDACE^E
multiflorus, bracteis membranaceis, ovato-lanceolatis, setaceo- acuminatis. Flores purpureo-fusci etflavo-variegati. Pedicelli gra- ciles. Sepala lateralia ovato-lanceolata, acuta, leviter undulata. Sepalum dorsale simile. Petala elliptico-ovata apice rotundata et minute apiculata. Labellum trilobatum, anguste longeque un- guiculatum; lobi laterales erecti late falcati, ad apicem rotundati; lobus medius prope basim quadratus, superne dilatatus, late retu- sus ; discus inter lobos laterales crista oblonga carnosa quadrilo- bata ornatus. Columna clavata, crassa, exalata.
Roots rather coarse, whitish in dried specimens, smooth. Pseu¬ dobulbs erect, 5-8 cm. long, ±1.5 cm. in diameter below the mid¬ dle, in dried specimens much wrinkled and with yellowish brown annulae at the five to eight points at which the leaves were borne, the fibrous remains of the leaves persistent. Leaves ±1.5 dm. long, ±5 mm. wide at flowering time, ±4.5 dm. long, ±1.5 cm. wide when mature, erect, coriaceous, many nerved, linear, acute, apicu- late, soon surpassed by the elongated scape with which they arise from the base of a mature and defoliated pseudobulb, middle nerve prominent beneath. Scape ±4 dm. tall, erect, with a few scattered bracts below the raceme, the bracts ample, the lower¬ most one sheathing and obliquely truncate, acute. Raceme ±1.5 dm. (sometimes exceeding 3 dm.) in length, for the most part simple sometimes branching below, many flowered. Bracts of the raceme about 1.5 cm. long, membranaceous, ovate-lanceolate, acute, a little shorter than the pedicellate ovary. Pedicels slender with the ovary up to 2 cm. long. Flowers about 2 cm. across, some¬ times 2.5 cm. across between the tips of the lateral sepals, brown¬ ish-purple with yellow spots and blotches. Sepals and petals spreading. Lateral sepals 9-12 mm. long, up to 6 mm. wide, ovate-lanceolate, acute, margins undulate. Petals 8-11 mm. long, ±7 mm. wide, elliptic-ovate, very slightly clawed, apparently little
[ 52 ]
ORCHID ACEiE
if at all undulated. Labellum three-lobed, suggesting in its form the labellum of C. punctatum; lateral lobes ±6 mm. long, about 3 mm. wide, erect, parallel with the column, rounded at the tip, conspicuously nerved ; middle lobe about 5 mm. long, 5 mm. wide at the base, below the middle subquadrate, above the middle di¬ lated, rounded on the distal margin, emarginate, conspicuously nerved; claw about 3.5 mm. long, nearly 2 mm. wide, oblong- cuneate, somewhat canaliculate and traversed longitudinally by raised nerves that terminate in the large lobed crest that is situ¬ ated on the disc between the lateral lobes. Crest four-lobed. Col¬ umn 7 mm. long,clavate,exalate, produced into a submembrana- ceous, mobile, oblong foot which is joined to the claw of the lip. Anther operculate, incumbent. Pollinia four, cohering in pairs.
In the structure of the flower this species shows close affinity with Cyrtopodium punctatum Lindl., but is clearly set apart by the smaller proportions of the perianth organs, more slender habit and by the simple, rarely branched, inflorescence. The lateral lobes of the labellum, as far as dried material allows one to judge, assume a different attitude from those of C. cristatum, and are not so nearly orbicular. F urthermore the crest is a differentiating character as a glance at the accompanying plate clearly shows. The flowers in dried specimens still retain indications of the red¬ dish-brown maculations and yellow bars and blotches that char¬ acterize the species when the flowers are fresh. Unfortunately Lindley did not make any reference to the color of the flowers of C. cristatum , and from the specimens I have examined it is not possible to detect any traces of the color that might have pre¬ vailed in living material. In the preparation of the plate that il¬ lustrates C. Broadway i an attempt has been made to restore the color areas, but due allowance should be made not only for nor¬ mal variations but for those crudities in restoration that must
[ 53 ]
ORCHIDACEiE
always attend efforts to represent, as it must have been when fresh, a flower that has long been dried. (Plate 112.)
Trinidad, B. W. I., Aripo Savannah, W. E. Broadway 2343, April 16,
1908. Flowers reddish brown and yellow.
Cyrtopodium cristatum Lindl. in Bot. Reg. (1841) sub t. 8 ; in Hook. Lond. Journ. Bot. 2 (1843) 672; Reichb. f. in Walp. Ann. 6 (1861) 667.
To complete the survey of the Trinidad species that have been referred to C. cristatuvi Lindl. it is now proposed to amplify the brief characterization which Lindley published in the Botanical Register for 1841. The remarks that follow are based on mate¬ rial in my herbarium that consists of a few flowers and a very clear tracing taken from Lindley ’s specimen of Schomburgk’s 628 preserved at Kew. For this material and for the opportunity to clear away ambiguity I am indebted to Sir David Prain.
The type specimen of Cyrtopodium cristatum was collected by Schomburgk in what is now British Guiana and bears the num¬ ber 628. Lindley in his list of Schomburgk’s orchids, a list which appeared in Bentham’s series of contributions toward a Flora of South America, gave the habitat of C. cristatum as rocks and trees.
In habit Lindley ’s type resembles C. Broadwayi very closely and might easily be mistaken for it until flowers of the two spe¬ cies are compared side by side; but it is very surprising that it should have been confused with Otostylis brachystalicc , which has white flowers and short scale-like bracts subtending the pedicels, not to mention the very different lip with much reduced lateral lobes.
Sepals ±14 mm. long, and about 8 mm. wide, conspicuously nine-nerved in dried specimens, elliptic-ovate, hardly acute, rather fleshy in texture, spreading, with undulate margins. Petals
[ 54 ]
ORCHIDACE^E
13.25 mm. long, 9 mm. wide above the middle, about 3 mm. wide near the base, obovate from a broad cuneate base, rounded at the apex, apparently not undulated on the margin, nine-nerved. La- bellum about 1 cm. long, conspicuously three-lobed, with a com¬ paratively short claw, articulated with the column foot, fleshy with coarse nerves. Claw 4 mm. long, 2.5 mm. wide near the mid¬ dle, oblong, cuneate, five-nerved ; lateral lobes 7 mm. long and wide, approximately 4 mm. wide where they join the main body of the lip, erect, partly concealing the column, round-falcate or orbicular, conspicuously nerved; middle lobe 5 mm. long, about 9 mm. wide, traversed longitudinally by about thirteen veins which are conspicuous in dried material, rounded at the apex, not emar- ginate nor thickened along the edge. Callus on the disc between the lateral lobes tuberculate, corrugated, extending onto the claw, 4 mm. long, 2 mm. wide at the truncated apex. Column fleshy, 6 mm. long to the point of insertion of the upper sepal, produced into a conspicuous foot. Pedicel and ovary 11 mm. long, slender. Flowers in dried material of a deep brown color. Habit as illus¬ trated. (Plate 113.)
British Guiana, Schomburgk 638. On rocks and trees.
[ 55 ]
'
■
.
PLATE 111
ORCHIDACEiE
Plate 111: Otostylis brachystalix Plant and inflorescence natural size, drawn from a specimen preserved in the Herbarium of Oakes Ames, collected by W. E. Broadway in Trinidad, British West Indies. 1, flower en¬ larged. 2, labellum and column, sepals and petals removed. 3, column enlarged.
[ 58 ]
PLATE 112
ORCHIDACEiE
Plate 112: Cyrtopodium Broadwayi Plant and inflorescence, drawn from specimens preserved in the Herbarium of Oakes Ames, collected by W. E. Broadway, in Trinidad, Brit¬ ish West Indies. 1, lip and column as seen from the side, enlarged. 2, flower enlarged, lateral lobes of the labellum spread out.
[ 60 ]
»
PLATE 113
ORCHIDACEiE
Plate 113: Cyrtopodium crist atum Plant drawn approximately natural size from a tracing taken from the type preserved in the Lindley Herbarium at Kew, England, collected by R. H. Schomburgk, in British Guiana. 1, 2 and 3, drawn from a flower of Lindley’s type preserved in the Herbarium of Oakes Ames.
1, labellum enlarged, lateral lobes spread out.
2, labellum and column enlarged, as seen from the side. 3, flower enlarged, lateral lobes of the labellum spread out.
[ 62 ]
NOTES ON ERYTHRODES
WITH NOMENCLATORIAL CHANGES AND DESCRIPTIONS OF
THREE NEW SPECIES
The genus Erythrodes founded by Blume and regarded as syn¬ onymous with Physurus L. C. Rich, has been reestablished as a living name by Fawcett and Rendle because Physurus as pub¬ lished by L. C. Richard is a nomen nudum.1
Dr. Rudolf Schlechter in his studies of the group to which Erythrodes belongs concluded that the palaeotropical species are clearly distinguished from the neotropical by a bilobed or didy- mous sac on the labellum. He proposed that Blume’s Erythro¬ des should be restricted to the palaeotropical species, the neotrop¬ ical species being set apart as a distinct genus under Physurus.
Fawcett and Rendle in their discussion of the value of a bi¬ lobed sac for generic characterization put Schlechter’s conclu¬ sions aside as in their judgment a bilobed sac is of trivial im¬ portance for the separation of the species of this group into dis¬ tinct genera, and they directed attention to the occasional occur¬ rence of a bilobed sac in the West Indian Erythrodes plantaginea Fawc. & Rendle.
Surely a bilobed sac by itself is a very superficial character for purposes of classification and should be disregarded in the defini¬ tion of genera, unless correlated with other substantial differ¬ ences. It is a character that is capable of extreme variation in species that exhibit it and may occur as a manifestation of abnor¬ mal development in a species which is norfnally without it. In other words, it is a character that may come and go and is plastic under those influences that govern the production of teratological formations in the orchid flower. Furthermore, the bilobed sac is
1 Flora of Jamaica 1 (1910) 28.
[ 63 ]
ORCHIDACEJE
not restricted to the species of the eastern tropics, as Erythrodes secunda Ames, a native of Mexico, has a bilobed sac in the flowers I have examined, and proves that this character is not, as Schlech- ter would have us believe, geographically limited in the group under consideration. The slightest tendency toward a lobing of the sac demands an arbitrarily fixed limit to guide us as to the point where Physurus should end and Erythrodes should begin.
In my studies of numerous specimens of Erythrodes from the tropics of both hemispheres, I have endeavored to find characters correlated with a conspicuously bilobed sac that would make cleavage possible along geographical lines and point the way to an absolute separation of the palaeotropical from the neotropi¬ cal species. The gynostemium, however, in which one would ex¬ pect to find the more conservative structures of the flower, is very similar in all the species examined and exhibits only those variations, such as differences in the length of the anther or ros- tellar processes, that would naturally be expected and that occur in different species irrespective of their origin. Schlechter con¬ cluded that the gynostemium is more slender in the Old World species than in those of the New World, but I have been unable to demonstrate that this difference exists.
It is noteworthy that the carinae, complanate or papilliform calli that occur on the anterior wall of the sac in many species of Erythrodes have received very little attention; in fact they have been overlooked or disregarded by systematists who have characterized the genus. Bentham and Hooker failed to mention them in the Genera Plantarum and in his work on the orchids for Engler and Prantl’s Das Pflanzenfamilien Pfitzer used the ab¬ sence of warts from the interior of the sac as a key character to distinguish this genus from Queteletia. The neglect of these structures is altogether surprising because their position on the
[ 64 1
ORCHIDACE.E
anterior wall of the sac and their form may be seen, in many cases, through the translucent lateral walls. In the separation of critical species I have found that these emergences or thickened tissues are really helpful, often decisive.
The sac of Erythrodes species is usually provided with three conspicuous nerves that pass down the posterior wall and as¬ cend by way of the anterior wall to the elongated limb of the labellum. Alternating with these nerves are the carinae or pa¬ pilliform calli just mentioned. They are usually four in number when they occur, but in some species two may be suppressed, or they may be broken up into cylindrical processes. Occasionally the outer ones, those situated on each side of the lateral nerves, are wanting; when present they are situated higher on the wall of the sac than the inner ones. The central pair, one on each side of the central nerve, may extend almost to the base of the sac as thickened lines, or they may be for the most part suppressed and appear simply as scale-like or papilliform calli, or, as in Ery¬ throdes mexicana Ames, they may be verruciform. It is note¬ worthy that these raised thickenings are not situated on the nerves as is usually the case with such structures, but arise from the interneural tissues.
I have observed that in some species, as for example in Ery¬ throdes vaginata (Hook.) Ames and E. querceticola (Lindl.) Ames, there are prominences alternating with the nerves of the sac, w'hile in E. secunda Ames, there may be twelve cylindrical, slender processes, that resemble hairs found in some species of Goody era. Four of these processes occur in a vertical line on either side of the central nerve and two on each side of the sac just below the point where the lateral nerves curve sharply up¬ ward in their course. In other words these twelve processes in E.
secunda, six to the right and six to the left of the centre of the
[ 65 ]
ORCHIDACEJS
sac, correspond in their position to the carinae or thickened lines of other species. They are interneural.
It would seem that conclusive proof that the palaeotropical and neotropical species constitute a single genus would be estab¬ lished, if in connection with the similarity of the structures of the gynostemium and perianth, carinae or papilliform emergences could be found in representatives of the group taken from both hemispheres. Erythrodes Wenzelii Ames from the Philippines has a conspicuously bilobed sac, and is provided with carinae which in form and position resemble very closely those of E. purpurea Ames from Guatemala. The carinae alternate with the central nerves, the outer pair situated higher up on the sac than the inner pair. In E. Weberi Ames, a native of the Philippines, the anterior wall of the sac is also provided with carinae which are similar in appearance and position to those of E. argyrosticta (Schltr.) Ames from Guatemala. It seems inconceivable that the evidence offered by these carinae should be disregarded. In my estimation they indicate a kinship that geographical bounds cannot alter.
The following changes in nomenclature are supplementary to those made in Fascicle V of this work on pages twenty-eight and twenty-nine.
Erythrodes aratanhensis ( Rodr .) Ames comb. nov. Physu- rus aratanhensis Rodr. Orch. Nov. 2 (1882) 290.
Erythrodes arietina (Reichb.f. § Warm.) Ames comb. nov. Physurus arietinus Reichb.f. <% Warm, in Reichb. f. Otia Bot. Hamb. 2 (1881) 52.
Erythrodes argyrosticta (Schltr.) Ames comb. nov. Physu¬ rus argyrostictus Schltr. in Fedde Repert. 16 (1920) 440.
In the structure of the flowers this species bears a close re¬ semblance to E. purpurea Ames. In habit, however, it is readily
recognized by the shortness of the peduncle between the lower-
[ 66 ]
ORCHIDACEiE
most flowers of the raceme and the uppermost leaf. In Schlech- ter’s type specimen the peduncle is about 3 cm. long between the raceme and the leafy portion of the stem and in my speci¬ mens exhibits an extreme range of 3-4 cm. In the type of E. purpurea the part of the peduncle that extends between the ra¬ ceme and the leaves is 13.5 cm. long and in duplicates of the type number preserved in the United States National Herba¬ rium it is 12-13.5 cm. long. Aside from habital differences there are other reliable guides to aid one in distinguishing between these species, one of the most important being the calli that are found on the anterior wall of the sac alternating with the three main nerves. In E. argyrosticta these calli are four elongated ridges, hardly thickened at the free basal tip, the two outer ones about 1 mm. long, the inner ones equally long but so situated that they extend nearly to the base of the sac. In E. purpurea there are also four ridges on the anterior wall of the sac, but they are more or less thickened and papilliform at the basal tip and in the type do not form such distinct carinae or raised lines as in E. argyrosticta. In both species the sepals are provided exter¬ nally with four or five-celled hairs, but in E. purpurea the hairs are more or less localized as tufts near the tip, only a few scat¬ tering ones occurring elsewhere.
Erythrodes bicolor ( Rodr .) Ames comb. nov. Physurus bi¬ color Rodr. Orch. Nov. 2 (1882) 290.
Erythrodes bifalcis ( Lindl .) Ames comb. nov. Physurus bi- falcis Lindl. Gen. & Sp. Orch. (1840) 502. Microchilus bifalcis D. Bietr. Syn. PL 5 (1852) 166.
In Lindley’s herbarium at Kew the flower of this species is shown by a rough sketch to have extraordinarily large falcate lobes at the end of the labellum. The spur is very long and slen¬ derly cylindrical.
[ 67 ]
ORCHIDACEAC
Erythrodes brachyrhyncha {Reichb. f) Ames comb. nov. Physurus brachyrhynchus Reichb. f. Xen. Orch. 2 (1873) 184.
Erythrodes calophylla {Reichb. f) Ames comb. nov. Physu¬ rus calophyllus Reichb. f. Beitr. Orch. Centr. Am. (1866) 64.
Erythrodes caucana ( Schltr .) Ames comb. nov. Physurus caucanus Schltr. in Fedde Repert. Beihefte 7 (1920) 70.
Erythrodes chicharrasensis^ mes sp. nov. Herba terrestris, erecta, 2-2.5 dm. alta, rhizomate decumbenti, cauliformi, radi- canti, mox in caulem erectum abeunte, radicibus villosis, cauli plerumque foliato, supra vaginis foliaceis obsesso, tereti, superne dense glanduloso, infra partem foliatam vaginis tubularibus. F o- lia usque ad 8.5 cm. longa et 2.8 cm. lata, in sicco atroviridia, oblique lanceolata, prope apicem valde attenuata, chartacea in sicco, clare trinervia, patentia, plus minusve 2 cm. inter se distan- tia. Petioli plus minusve 2 cm. longi, vaginantes, laxe appressi. Racemus plus minusve 7 cm. longus, multiflorus, 2 cm. per me¬ dium. Bracteae inflorescentiae plus minusve 9 mm. longae, quam pedicellus cum ovario multo longiores, triangulari-lineares, acu- tissimae, scariosae, clare trinerviae, glanduloso-pubescentes. Se- pala lateralia 4.5 mm. longa, 1 mm. lata, oblongo-ligulata, obtusa, uninervia, prope apicem glanduloso-villosa. Sepalum dorsale si¬ mile. Petala 4 mm. longa, vix 1 mm. lata, oblongo-spathulata. Labellum usque ad lobum terminalem 4 mm. longum, margini- bus incurvum; lobus terminalis 1 mm. longus et 3 mm. latus, oblongo-crescentiformis, obtuse mucronatus, minute papillosus ; calcar 3.5 mm. longum, complanato-cylindraceum, plus minusve 1 mm. per medium, intus quadricallosum callis papilliformibus. Columna 3 mm. longa, rostello profunde bifido utrinque angu- lato, divisionibus setaceis, acutissimis. Anthera 2 mm. longa, an- guste lanceolata. Ovarium glanduloso-pubescens.
Rhizome elongated, creeping, rooting at the nodes, the roots
[ 68 ]
ORCHIDACEvE
produced opposite the lowermost leaves, one at each node ; inter¬ nodes approximately 2-3 cm. long. Leaves at first crowded, the sheaths of the petioles entirely concealing the upper part of the stem, apparently variegated with oblique cross veins of a deep green when fresh ; where the stem is revealed it appears to have been suffused with purple when in life. Racemes slender, the period of anthesis of long duration as is shown in the type in which the lowermost flowers are expanded while the buds of the upper flowers are crowded in a pyramidal, dense cluster 2.5-3 cm. long. Rachis densely glandular pubescent. Peduncle about 3 cm. long between the uppermost foliaceous bract and the lowermost flower, densely glandular, brownish pubescent. Sepals with a few extraordinarily long glandular hairs scattered over the outer sur¬ face and with a tuft of hairs near the tip, the longest ones .5 mm. in length, margins smooth. Petals slightly dilated above the mid¬ dle, smooth, lightly adherent to the upper sepal. Labellum fleshy, densely papillose on the upper third, the dilated bilobed or cres¬ cent-shaped terminal portion being papillose on both the upper and lower surfaces; spur cylindrical, about equally thick from the base to the more or less abruptly rounded tip, on the anterior wall within there are four linear calli, — in parallel pairs, — alter¬ nating with the three main nerves, two (the outer ones) situated just below the middle of the spur, two (the inner ones) with their free tips almost reaching the tip of the spur. Ovary 5 mm. long, fusiform-cylindrical, densely glandular hairy.
The terminal lobes of the lip suggest very strongly the Peru¬ vian Erythrodes bifalcis (Lindl.) Ames which has a much more slender, spur-like sac. The type, according to notes in my her¬ barium's very similar in habit to Spruce 4464 and 4870 pre¬ served at Kew, specimens referable to E. repens (Poepp. & Endl.) Ames. In the form of the labellum, in the rostellum and elon-
[ 69 ]
ORCHIDACE.E
gated anther very similar to the Guatemalan^, lunifera (Schltr.) Ames, but differing from it in general habit and in the flowers being more sparsely pubescent.
Mexico, Chiapas, Chicharras, E. W. Nelson 3808, February 12-15, 1896.
6000 feet altitude.
Erythrodes clavigera ( Reichb.f '.) Ames comb. nov. Physu- rus claviger Reichb.f. in Bonpl. 4 (1856) 211.
Erythrodes commelinoides ( Rodr .) Ames comb. nov. Phy- surus commelinoides Rodr. Orch. Nov. 1 (1877) 193.
Erythrodes debilis (Lindl.) Ames comb. nov. Physurus de- bilis Lindl. Gen. & Sp. Orch. (1840) 503.
Erythrodes densiflora (Lindl.) Ames comb. nov. Physurus densiflorus Lindl. Gen. & Sp. Orch. (1840) 502.
Erythrodes dolichostachy a (AcMr.) Ames comb. nov. Phy¬ surus dolichostachys Schltr. in Fedde Repert. Beihefte 7 (1920)
70.
Erythrodes erythrodoides (Schltr.) Ames comb. nov. Phy¬ surus erythrodoides Schltr. in Fedde Repert. Beihefte 7 (1920)
71.
Erythrodes foliosa (Lindl.) Ames comb. nov. Physurus fo- liosus Lindl. Gen. & Sp. Orch. (1840) 503.
Erythrodes hetaerioides (Schltr.) Ames comb. nov. Physu¬ rus hetaerioides Schltr. in Fedde Repert. Beihefte 7 (1920) 72.
Erythrodes hyphaematica ( Reichb.f. ) A mes comb. nov. Phy¬ surus hyphaematicus Reichb.f. Xen. Orch. 2 (1874) 184.
Erythrodes humilis (Cogn.) Ames comb. nov. Physurus hu- milis Cogn. in Mart. FI. Bras. 3, pt. 4 (1895) 232.
Erythrodes juruenensis (Hoehne) Ames comb. nov. Physu¬ rus juruenensis iFoe/me in Comm. Linhas Teleg.Estrateg. Matto Gross. Annexo 5, pt. 1 (1910) 30, t. 10.
[ 70 ]
ORCHIDACE^E
Erythrodes lacteola ( Rodr .) Ames comb. nov. Physurus lac- teolus Rodr. Orch. Nov. 2 (1882) 292.
Erythrodes Lehmannii (Schltr.) Ames comb. nov. Physurus Lehmannii Schltr. in Beihefte Bot. Centralbl. 36, Abt. 2 (1918) 379.
This species in the note accompanying the original description is compared with E. secunda Ames, a Mexican species that is well characterized among its allies by the general structure of the labellum, the apical part being" reduced to a crenate margined semiorbicular lobe. In E. Lehmannii the tip of the labellum is also much less developed than is usual in the genus, the terminal lobe being obtusely angled on each side and mucronate in the middle. E. secunda would seem to differ from E. Lehmannii very markedly in having within the sac, near the base, twelve elongated processes.
Erythrodes Lindleyana ( Cogn .) Ames comb. nov. Physurus Liindleyanus Cogn. in Mart. FI. Bras. 3, pt. 4 (1895) 238, t. 55, f. II. Physurus roseus Lindl. in Herb. Monac., non Lindl. Gen. & Sp. Orch.
Erythrodes loxoglottis (Reichb.f.) Ames comb. nov. Phy¬ surus loxoglottis Reichb.f. Beitr. Orch. Centr. Am. (1866) 64; Xen. Orch. 2 (1874) 182, t. 178, f. I.
I have been unable to place this species satisfactorily. It resembles E. secunda Ames in the foliage but is distinct in the raceme and in the terminal lobe of the labellum. E. stictophylla (Schltr.) Ames is a very closely related species.
Erythrodes lunifera (Schltr.) Ames comb. nov. Physurus lu- niferus Schltr. in Fedde Repert. 15 (1918) 199.
In the lunate terminal lobe of the labellum and in the foliage not unlike E. chicharrasensis Ames, but with the lip-lobes nar-
[ H ]
ORCHIDACEiE
rower and more acute, and without angles on the sides of the rostellar process.
Erythrodes maculata (Hook.) Ames comb. nov. Physurus maculatus Hook, in Bot. Mag. (1862) t. 5305.
In floral structure similar to E. vaginata (Hook.) Ames, but different in habit.
Erythrodes major ( Presl ) Ames comb. nov. Microchilus major Presl Rel. Haenk. 1 (1827) 94. Physurus major Schltr. in Fedde Repert. Beihefte 9 (1921) 132. Physurus Preslei Lindl. Gen. & Sp. Orch. (1840) 501.
This is closely allied with E. procera (Schltr.) Ames and a study of more material may prove that they are conspecific.
Erythrodes Jdlayoriana(Kranzl.)Ames comb. nov. Physurus Mayorianus Krdnzl. in Mem. Sci. Nat. Neuchat. 5 (1914) 355 (Fuhrm. & May. Voy. Explor. Sci. Colomb. (1914) 355).
Erythrodes metallescens ( Rodr .) Ames comb. nov. Phy¬ surus metallescens Rodr. Orch. Nov. 2 (1882) 291.
Erythrodes mexicana Ames sp. nov. Herba terrestris, erecta, 3 dm. alta, rhizomate decumbenti, cauliformi, radicanti, mox in caulem erectum abeunte, radicibus elongatis, villosis; cauli prope basim quinquefoliato, supra vaginis paucis dissitis obsesso, tereti, glanduloso-puberulo, infra partem foliatam vaginis trun- catis scariosis. Folia plus minusve 5 cm. longa, usque ad 31 mm. lata, ovato-lanceolata, acuminata, acutissima, in sicco viridia. Pe- tioli vaginantes, 2 cm. longi. Scapus cum racemo 2.4 dm. longus, gracilis, bracteis vaginantibus tribus, 1.4-2 cm. longis, scariosis, glabris, plus minusve 4 cm. inter se distantibus. Racemus 6 cm. longus, plus minusve decemflorus. Bracteae inflorescentiae 8-10 mm. longae, scariosae, sparse pubescentes, marginibus ciliolatis, triangulari-lanceolatae, acuminatae, acutissimae, quam pedicel- lus cum ovario vix longiores. Flores in genere inter majores, illis
[ ™ ]
ORCHID ACEJE
E. Killipii fere aequimagni. Sepala lateralia 6.25 mm. longa, 2 mm. lata trans medium, oblonga, obtusa, conspicue uninervia, extus glanduloso-pubescentia. Sepalum dorsale simile, 5.5 mm. longum, 2 mm. latum. Petala 6 mm. longa, 2 mm. lata, oblongo- spathulata, obtusa. Labellum 7 mm. longum, marginibus incur- vum, apice dilatatum, deflexum, reniforme vel obtuse crescenti- forme, 1 mm. longum, 3 mm. latum, obscure mucronatum, mi¬ nute papillosum. Calcar 3 mm. longum, obtusissimum, intus squa- mis quatuor verruciformibus, breviter carinatis ornatum. Co- lumna 3 mm. longa, rostello profunde bipartito, anthera cordata, acuminata, acuta. Ovarium glanduloso-pubescens.
A rather robust plant among the Mexican and Central Amer¬ ican species of the genus, with flowers that suggest in size the flowers of E. Killipii Ames and E. vesicifera (Reichb. f.) Ames. Rhizome stout with a single root at each node, internodes 2.5 cm. long. Just below the leafly portion of the stem there are three or four scarious whitish or yellowish tubular sheaths, about 12 mm. long. Leaves more or less crowded near the ground; leafy part of the stem 5 cm. long, concealed by the sheathing, loosely appressed bases of the spreading leaves. Leaves five, ovate-lanceolate, acuminate, acute, about nine-nerved, from stud¬ ies of dried material apparently variegated or mottled when fresh. Peduncle with whitish glandular pubescence, with three bracts below the raceme, the bracts whitish, obliquely truncate, acute, with three prominent nerves. Raceme loose flowered, not unlike the raceme of E. plantaginea (Lindl.) Fawc. & Rendle, but with slightly larger flowers. The apical lobe of the labellum is sharply deflexed and minutely papillose. Within the sac, on the anterior wall, alternating with the three main nerves are four calli, the two outer ones j ust below the middle of the sac, the two inner
ones near the base of the sac.
[ 73 ]
ORCHIDACEiE
In the nearly related Erythrodes procera (Schltr.) Ames the calli on the anterior wall of the sac are small, the lowermost pair being very much reduced and consequently inconspicuous; in E.plantaginea the calli are represented by much elongated thick¬ enings, the inner pair extending nearly the whole length of the sac and appearing rather as thickened nerves than as calli. In E. vesicifera the thickenings of the sac are carinate and much longer in relation to their height than the nearly quadrate, com- planate calli of E. meocicana which are conspicuously protuber¬ ant. Erythrodes Killipii from Panama is a closely related spe¬ cies, but very easily distinguished by means of the thickened, elongated carinae which extend from above the middle and reach nearly to the base of the sac.
Mexico, Chiapas, Pinabate, E. W. Nelson 3777 , February 8, 1896.
6500-8000 feet altitude.
Erythrodes minor (Presl) Ames comb. nov. Microchilus minor Presl Rel. Haenk. 1 (1827) 94. Physurus minor Lindl. Gen. & Sp. Orch. (1840) 503.
Erythrodes mystacina ( Reichb.f ’.) Ames comb. nov. Phy¬ surus mystacinus Reichb.f. Xen. Orch. 2 (1874) 183.
Erythrodes nigrescens (Schltr.) Ames comb. nov. Physurus nigrescens Schltr. in Beihefte Bot. Centralbl. 36, Abt. 2 (1918) 380.
In general habit not unlike E. Killipii Ames, but with a dif¬ ferent terminal lobe on the lip. The plants when dried take on a blackish color. This characteristic has not been noted in E. Kil¬ lipii as represented by Ivillip’s specimens and by Maocon 5570 from the Upper Caldera Biver, Panama.
Erythrodes ovata (Lindl.) Ames comb. nov. Physurus ovatus Lindl. Gen. & Sp. Orch. (1840) 502.
[74]
ORCHIDACEdS
In general habit not unlike E. purpurea. The extraordinarily long peduncle in Lindley’s specimen of Matthews 1877 from Peru is a distinguishing character of this species. The sac of the labellum as figured by Lindley on the type sheet is bulbous.
Erythrodes paleacea ( Schltr .) Ames comb. nov. Physurus paleaceus Schltr. in Fedde Repert. Beihefte 7 (1920) 72.
Erythrodes pauciflora ( Poepp . § Endl.) Ames comb. nov. Pelexia pauciflora Poepp. Endl. Nov. Gen. ac Sp. 2 (1838) 17, t. 124 (excl. f. a-e). Physurus pauciflorus Lindl. Gen. & Sp. Orch. (1840) 504.
Erythrodes Petersiana ( Cogn .) Ames comb. nov. Physurus Petersianus Cogn. in Mart. FI. Bras. 3, pt. 4 (1895) 227, t. 52, f. I. Physurus roseus Reichb.f. in Linnaea 41 (1877) 125, non Lindl.
Erythrodes Pittieri (Schltr.) Ames comb. nov. Physurus Pit- tieri Schltr. in Fedde Repert. Beihefte 6 (1919) 31.
Erythrodes procera (Schltr.) Ames in Proc. Biol. Soc. Wash. 34 (1921) 150. Physurus procerus Schltr. in Fedde Repert. Bei¬ hefte 7 (1920) 73.
This species is closely allied to E. major and to E. Killipii. From the latter species it differs markedly in the calli on the anterior wall of the sac, the lowermost ones being very much reduced.
Erythrodes rariflora (Lindl.) Ames comb. nov. Physurus rariflorus Lindl. Orch. Linden. (1846) 28.
Erythrodes repens (Poepp. <$ Endl.) Ames comb. nov. Pe¬ lexia repens Poepp. Endl. Nov. Gen. ac Sp. 2 (1838) 17, 1. 124 f. a— e. Physurus repens Lindl. Gen. & Sp. Orch. (1840) 502. Mi¬ crochips repens D. Dietr. Syn. PL 5 (1852) 166.
Erythrodes rosea (Lindl.) Ames comb. nov. Physurus roseus Lindl. Gen. & Sp. Orch. (1840) 501.
Erythrodes stenocentron (Schltr.) Ames comb. nov. Phy-
[ 75 ]
ORCHID ACEvE
surus stenocentron Schltr. in Fedde Repert. Beihefte 9 (1921) 60.
Erythrodes stictophylla. (Schltr.) Ames comb. nov. Physurus stictophyllus Schltr. in Fedde Repert. 10 (1911) 249.
In the representatives of this species found in herbaria a sur¬ prising range of variation in size is to be observed, some of the more robust specimens resembling very closely E. Tuerckheimii (Schltr.) Ames. The smaller flowers and the erect flattened pa¬ pillae at the base of the sac are differentiating characters. I have been unable to detect in the flowers of E. Tuerckheimii any trace of calli at the base of the sac and the lateral ones are wanting or obscure. Apparently a close ally of E. loxoglottis (Reichb. f.) Ames, which is similar in habit to some of the smaller speci¬ mens observed.
Erythrodes tridax (Reichb. f.) Ames comb. nov. Physurus tridax Reichb. f. Beitr. Orch. Centr. Am. (1866) 64; Xen. Orch. 2 (1874) 182, t. 178, f. II.
Erythrodes trinitatis Ames sp. nov. AfF. E. querceticolae et E. vaginatae. Herba terrestris, ereeta, plus minusve 9 cm. alta, glabra, rhizomate decumbenti, cauliformi. Folia approximata, variegata, lamina 1-2 cm. longa, usque ad 11 mm. lata, ovato- lanceolata, acuminata, acuta. Petiolus plus minusve 1 cm. longus, infra medium laxe vaginans, supra medium contractus. Racemus plus minusve 2 cm. longus, densiflorus, glaber. Bracteae inflo- rescentiae usque ad 6 mm. longae, scariosae, ovato-lanceolatae, uninerviae, acuminatae, acutae. Flores plus minusve 4 mm. longi, albidi, mediocres. Sepala lateralia 4 mm. longa, 1.75 mm. lata, elliptico-oblonga, apice leviter cucullata,subobtusa, glabra. Sepalum dorsale 4 mm. longum, extus leviter carinatum, prope apicem minute papillosum. Petala 3.5 mm. longa, 1 mm. lata supra medium, inaequaliter spathulata, obtusa, uninervia. Label- lum 3.5 mm. longum, saccatum, lobo terminali trilobulato, 2 mm.
[ ™ ]
ORCHIDACE^E
longo, trans basim 2 mm. lato, lobulo intermedio triangulari; dis¬ cus lobi terminalis carina transversa ornatus. Calcar 2 mm. Ion- gum, subglobulum, intus glabrum. Gynostemium 1.75 mm. Ion- gum. Ovarium cum pedicello usque ad 4 mm. longum, longitu- dinaliter alatum.
Erythrodes trinitatis is closely related to the Central Ameri¬ can E. vaginata and to E. querceticola. From both it differs in the form of the terminal lobe of the labellum and in the more compact habit. The stems are leafy almost to the raceme.
Trinidad, B. W. I., St. Anne’s, W. E. Broadway s. n., 1888. A lovely
variegated terrestrial orchid. (Type in Herb. N. Y. Bot. Gard.)
Erythrodes Tuerckheimii ( Schltr .) Ames comb. nov. Physu- rus Tuerckheimii Schltr. in Fedde Repert. 2 (1906) 132. Physu- rus polygonatus Ames inDonn.-Sm. Enum. PI. Guatem. 7 (1905) 50, nomen; Orch. 2 (1908) 259; Orch. 3 (1908) 35, t. 35. Ery¬ throdes polygonata Ames Orch. 5 (1915) 28.
The type material of E. Tuerckheimii and E. polygonata is represented by the same number of Tuerckheim’s Guatemalan collection.
Erythrodes vaginata {Hook.) Ames Orch. 5 (1915) 29. Physurus vaginatus Hook. Ic. PI. (1842) t. 449.
Physurus humidicola Schltr. in Fedde Repert. 15 (1918) 198 is a very near relative of E. vaginata and in general habit re¬ sembles one of the larger specimens of Skinner’s Guatemalan collection on which Hooker based his description of Physurus vaginatus.
Physurus trilobulatus Schltr. in Fedde Repert. 15 (1918) 199 is surely a near ally of E. vaginata. In habit it resembles closely one of the smaller specimens of Skinner’s Guatemalan collection of Physurus vaginatus, and in the structure of the labellum closelv resembles that species.
[ ^ ]
ORCHID ACEiE
Erythrodes valida ( Rolfe ) Ames comb. nov. Physurus vali- dus Rolfe in Kew Bull. (1912) 134.
Erythrodes vesicifera {Reichb. f.) Ames comb. nov. Physu¬ rus vesicifer Reichb. f Beitr. Orch. Centr. Am. (1866) 63.
Erythrodes xystophylla {Reichb. f.) Ames comb. nov. Phy¬ surus xystophyllus Reichb. f. Xen. Orch. 2 (1874) 183.
Erythrodes zeuxinoides ( Schltr .) Ames comb. nov. Physu¬ rus zeuxinoides Schltr. in Fedde Repert. Beihefte 7 (1920) 74.
[ ^ ]
NOMENCLATORIAL CHANGES IN DENDROCHILUM
PSEUD ACORIDIUM A vies gen. nov.
Sepala libera similia, glabra, patentia, subaequalia, saccum la- belli amplectentia, oblique ascendentia. Petala sepalis similia, pa¬ tentia. Labellum saccatum, bilobatum, ad basim columnae ses¬ sile, utrinque callosum, lobis erectis columnae parallelis, cucul- lum formantibus. Saccus scrotiformis, intus per medium leviter carinatus. Columna gracilis, oblique erecta, semiteres, exalata, apicem versus sensim paulo dilatata, clinandrio prominenti mar- ginibus integro, canaliculate. Anthera intra clinandrium oper- cularis, incumbens, bilocularis. Pollinia quatuor, cerea, pyrifor- mia, per paria in loculis segregata. Ovarium pedicellatum, cla- vatum. Herba epiphytica, erecta, radicibus flexuosis, carnosulis. Pseudobulbi parvi, monophylli, vaginis nonnullis scariosis ob- tecti. Folium coriaceum, breviter petiolatum. Scapus lateralis racemo terminali, erectus quam folium longior ; floribus distichis in racemo denso dispositis. Bracteae inflorescentiae glumaceae, rigidae, ovarium excedentes. Species singula adhuc nota silva- rum Mindoro indigena.
Pseudacoridium Woodianum Ames comb. nov. Dendro- chilum Woodianum Ames Orch. 2 (1908) 80 cum ic.
The bilobed labellum, its lobes erect and forming a hood over the column, the deep scrotiform sac with a keel inside, and the lateral scape arising from the base of a pseudobulb are charac¬ ters that set this genus apart from Acoridium; from Dendrochi- lum the exalate column, prominent clinandrium and saccate bi¬ lobed labellum separate it clearly.
[ TO ]
0RCHIDACEJ5
ACORIDIUM Nees Meyen.
From the abundance of material that has been examined in collections from the Philippine Islands it is clear that the sec¬ tion Acoridium of Dendrochilum should be elevated to generic rank. The exalate column, prominent rostellum and saccate or concave labellum which is rigidly attached to the column are differentiating characters that indicate the line of cleavage. The following nomenclatorial changes are proposed.
Acoridium affine Ames comb. nov. Dendrochilum affine Ames Orch. 2 (1908) 95.
Acoridium anfractoides Ames comb. nov. Dendrochilum anfractoides Ames Orch. 3 (1908) 13, t. 28.
Acoridium auricular e A ?nes comb. nov. Dendrochilum auri- culare Ames in Philipp. Journ. Sci. C. Bot. 4 (1909) 595.
Acoridium binuangense Ames comb. nov. Dendrochilum binuangense Ames Orch. 6 (1920) 281.
Acoridium cinnabarinum ( Pfitz .) Ames comb. nov. Dendro¬ chilum cinnabarinum PJitz. in Engl. Pflanzenreich IV. 50. II. B. 7 (1907) 104.
Acoridium confusum Ames comb. nov. Dendrochilum con- fusum Ames in Philipp. Journ. Sci. C. Bot. 6 (1911) 41.
Acoridium Curranii Ames comb. nov. Dendrochilum Cur- ranii Ames Orch. 3 (1908) 15, t. 29.
Acoridium Elmeri Ames comb. nov. Dendrochilum Elmeri Ames in Elmer Leaf!. Philipp. Bot. 5 (1912) 1558.
Acoridium exile Ames comb. nov. Dendrochilum exile Ames Orch. 2 (1908) 84.
Acoridium Foxworthyi Ames comb. nov. Dendrochilum Foxworthyi Ames Orch. 3 (1908) 8, t. 26.
Acoridium hastatum Ames comb. nov. Dendrochilum has-
tatum Ames Orch. 2 (1908) 99 cum ic.
[ 80 ]
ORCHIDACEJE
Acoridium Hutchinsonii Ames comb. nov. Dendrochilum Hutchinsonii Ames Orch. 2 (1908) 96 cum ic.
Acoridium irigense Ames comb. nov. Dendrochilum irigense Ames Orch. 5 (1915) 58.
Acoridium Loheri Ames comb. nov. Dendrochilum Loheri Ames Orch. 3 (1908) 12, t. 27, f. I.
Acoridium longibulbum Ames comb. nov. Dendrochilum longibulbum Ames in Philipp. Journ. Sci. C. Bot. 7 (1912) 26.
Acoridium lucbanense Ames comb. nov. Dendrochilum luc- banense^wes in Elmer Leafl. Philipp. Bot. 5 (1912) 1559.
Acoridium luzonense Ames comb. nov. Dendrochilum lu- zonens e Ames Orch. 2 (1908) 121 cum ic.
Acoridium maleolens ( Krdnzl .) Ames comb. nov. Dendro¬ chilum maleolens Krdnzl. in Orchis 2 (1908) 63.
Acoridium McGregorii Ames comb. nov. Dendrochilum McGregorii Ames in Philipp. Journ. Sci. C. Bot. 6 (1911) 42.
Acoridium microchilum ( Schltr .) Ames comb. nov. Platy- clinis microchila Schltr. in Bull. Herb. Boiss. ser. 2, 6 (1906) 302. Dendrochilum microchilum Ames Orch. 2 (1908) 87.
Acoridium mindorenseAfraescora6.7ion.Dendrochilum min- dorens e Ames Orch. 2 (1908) 91 cum ic.
Acoridium pulogense Ames comb. nov. Dendrochilum pulo- gens eAmes in Philipp. Journ. Sci. C. Bot. 4 (1909) 594.
Acoridium pumilum ( Reichb.f '.) Ames comb. nov. Dendro¬ chilum pumilum Reichb.f. in Bonpl. 3 (1855) 222.
Acoridium purpureum Ames comb. nov. Dendrochilum pur- pureum Ames Orch. 6 (1920) 286.
Acoridium quadrilobum Ames comb. nov. Dendrochilum quadrilobum Ames Orch. 5 (1915) 61.
Acoridium reniforme Ames comb. nov. Dendrochilum reni-
iorme Ames Orch. 5 (1915) 63.
[ 81 ]
ORCHIDACEiE
Acoridium simulacrum Ames comb.nov. Dendrochilum sim¬ ulacrum Ames Orch. 2 (1908) 92 cum ic.
Acoridium Vanoverberghii^mes comb . nov. Dendrochilum Vanoverberghii Ames in Philipp. Journ. Sci. C. Bot. 7 (1912) 27.
Acoridium Wenzelii Ames comb. nov. Dendrochilum Wen- zeAii Ames Orch. 5 (1915) 65.
[ ]
NEW OR NOTEWORTHY ORCHIDS
FROM
DIFFERENT PARTS OF THE WORLD
In the following pages new species from the Philippines, Bor¬ neo, Samoa, Central America, South America, Trinidad and Cuba are described. I have been unable to refer these species to any that have been reported heretofore. Notes that throw light on the history and distribution of orchids or that increase our knowledge of the orchid family have been included. To facili¬ tate reference, the genera have been arranged in alphabetical sequence.
Species from Bukidnon Subprovince, Mindanao, collected on Mount Lipa and Mount Candoon by Ramos and Edano indi¬ cate that as explorations in this region are carried on intensively many additions to the orchid flora of the Philippines may be expected. Among the species from the mountains named are several which lack close allies in those parts of the Philippines that are now botanically well known.
BULBOPHYLLUM Thou.
Bulbophyllum cubicum Ames sp. nov. Radices fibratae. Pseudobulbi caespitosi, 11 mm. alti, in sicco rugosi, anguste py- riformes, in sicco plus minusve 4 mm. in crassitudine infra me¬ dium, monophylli. Folium ligulatum, 3-6 cm. longum, circa 6 mm. latum trans medium, usque ad basim subtiliter attenuatum, apice acuto, minute apiculatum, textura coriaceum, in sicco nervosum. Scapi plus minusve 3 cm. longi, pergraciles, numerosi, folio multo breviores, infra florem nudi, in sicco circa .25 mm. in crassitudine, uniflori. Bractea pedicellum subtendens scariosa, infundibuliformis, 1 mm. longa. Flos 7 mm. longus, sepalis vix
[ 83 ]
ORCHIDACEiE
patentibus. Pedicellus cum ovario 8 mm. longus, pergracilis. Sepala lateralia circa 7 mm. longa, 2 mm. lata, oblongo-triangu- laria, sensim acuta, trinervia, in sicco pellucida. Sepalum dorsale simile, 6 mm. longum, basi concaviusculum. Petala 3.5 mm. longa, 1 mm. lata, oblonga, uninervia, margine praecipue prope apicem ciliata. Labellum mobile, 4 mm. longum, in circuitu elliptico-lanceolatum, vix 2 mm. latum trans medium, apice sub- acuto, basi valde cubico-incrassata, truncata, in basi bilobulatum, lobis circa 1 mm. latis, obtusis, leviter recurvis, elevatis, post lobos callo papilliformi ; unguis brevis, angustatus. Columna vix 1.5 mm. alta, in pedem 1.5 mm. longum producta, utrinque ad apicem in stelidium sigmoideum, antheram superans, vix 1 mm. longum producta. Anthera apice incrassata.
The labellum is thickened at the base so as to form a cube in front of the membranaceous elliptic-lanceolate lamina. At the summit of this cubically thickened base there is a crescent¬ shaped appendage — its rounded tip curved backward — that gives a two-lobed appearance to the basal part of the labellum. In addition to the crescent-shaped appendage there is a small papilliform or umbonulate protuberance, posteriorly situated, which points toward the column.
Near allies of B. cuhicum are B. grciciliscapum Ames & Rolfe which has broader leaves and a very different labellum; B. sap- phirinum Ames which has shorter scapes, more falcate lateral sepals and a very different labellum and B. angulatum J. J. Sm. which has ovate not oblong petals, shorter scapes and a very different labellum.
Philippines, Mindanao, Bukidnon Subprovince, Mahilucot River, M.
Ramos G. Edano Bar. Sci. 38633 , June-July 1920. On tree in forest.
“Flowers blue and brown.” 4200 feet altitude.
[ 81 ]
ORCHIDACE.E
Bulbophyllum erratum Ames sp. nov. Rhizoma longe repens, flexuosum, vagans, circa 1 mm. in crassitudine, prope pseudobulbos radicans, glabrum. Radices fibratae, abbreviatae, usque ad 2.7 cm. longae, .5 mm. per medium. Pseudobulbi 1.4-3. 5 cm. inter se distantes, 5-7 mm. longi, plus minusve 4 mm. in crassitudine, obpyriformes, usque ad apicem attenu¬ ate in sicco valde sulcati, flavescentes, nitidi, oblique erecti, monophylli. Folia 1.5-2. 5 cm. longa, usque ad 6 mm. lata, ob- longa vel oblongo-lanceolata, obtusa, vix apiculata, in petiolum brevem, rigidum, 1 mm. longum, sulcatum contracta, erecta, lamina coriacea, in sicco nervulosa. Scapi 6-9 mm. longi, e basi pseudobulbi vel a rhizomate in medio inter pseudobulbos orti, basi bractea 2 mm. longa, tubulata vaginati, uniflori vel biflori. Flos 6 mm. longus, bractea infundibuliformi, oblique truncata, 2 mm. longa subtentus. Pedunculus cum ovario plus minusve 2 mm. longus, pergracilis. Sepala lateralia usque ad 6 mm. longa, prope basim 2 mm. lata, triangula, acuta, apice subcarinata, trinervia, textura membranacea, mentum 1.25 mm. longum formantia. Sepalum dorsale 6 mm. longum, 1.5 mm. latum, ob- longo-lanceolatum, apice subcaudatum, acutum, trinervium, concavum. Petala quam sepala multo minora, vix 2 mm. longa, .6 mm. lata, oblonga vel oblongo-spathulata, apice rotundata, uninervia, glabra. Labellum 2 mm. longum, 1.5 mm. latum inter loborum lateralium apices extenditos, cum pede columnae articulatum, mobile; lobi laterales semiorbiculares, membranacei, in sicco pellucidi ; lobus medius vix 1 mm. longus, 1 mm. latus, valde incrassatus, apice late rotundatus, margine leviter revolu- tus, in sicco superficie corrugatus; discus inter lobos laterales linea vel carina incrassata ornatus. Columna carnosa, ebrachiata.
In habit very similar to B. alagense Ames, from which it is
distinguished by its apiculate leaves, less caudate sepals and
[ 85 ]
ORCHIDACEiE
broader labellum. Closely allied to B. longerepens Schltr. of the Celebes, from which it is separable by the longer scapes, by a distinctly three-lobed labellum and by slightly larger flowers.
Philippines, Mindanao, Bukidnon Subprovince, Mahilucot River, M. Ramos 8f G. Edano Bur. Sci. 38685, July 14, 1920. Along streams in forest. Flowers yellowish. 4100 feet altitude.
Bulbophyllum nigroscapum Ames sp. nov. Herba epiphy- tica. Rhizoma repens. Pseudobulbi subobsoleti, monophylli. Fo¬ lium erectum, ligulatum, coriaceum, petiolatum, utrinque atten- uatum, costa media supra sulcata, subtus carinata. Scapus sub- arcuatus, in sicco niger, plus minusve quindecimflorus, infra racemum bracteatus, bracteis vaginantibus. Bracteae inflorescen- tiae anguste lanceolatae. Sepala lateralia triangulari-lanceolata, ad basim mentum formantia. Sepalum dorsale simile. Petala se- palis multo breviora, oblonga, obtusa. Labellum simplex, ecallo- sum. Columna brevis, ad apicem bicuspidata vel bidentata.
Rhizome creeping, whitish when dry, copiously rooted, the roots coarse and fibrous. Pseudobulbs inconspicuous, reduced to raised discs on the rhizome, 1 mm. high, about 2 cm. apart, mono- phyllous, yellowish. Leaves 8-13.5 cm. long including the peti¬ ole, about 1.5 cm. wide at the middle, coriaceous, oblong, ligu- late, hardly bidentate at the rounded tip, sulcate above along the mid-nerve. Petiole short, about 1.5 cm. long, sulcate, rigid. Scape ±8 cm. long, with its loose raceme of cream colored flowers shorter than the leaves, blackish when dry, sheathed below the raceme by several tubular scarious bracts ±1 cm. long, the uppermost one obliquely infundibuliform, whitish when dry. Raceme about 5 cm. long, flowers about 3 mm. apart, subtended by linear-lanceolate, acuminate bracts ±5 mm. in length, whitish when dry, which exceed the pedicellate ovaries. Flowers about fifteen, yellowish in dried material, cream colored when fresh.
[ 86 ]
ORCHIDACE/E
Lateral sepals approximately 6 mm. long, 2 mm. wide near the middle, triangular-lanceolate, acute, somewhat thickened near the tip, decurrent on the curved foot of the column at base, form¬ ing an obtuse mentum 2 mm. long. Dorsal sepal 6 mm. long, oblong, ligulate, acute, conspicuously one-nerved along the mid¬ dle, similar to the lateral sepals. Petals 2-2.5 mm. long, about 1 mm. wide below the middle, much smaller than the sepals, ob¬ long, obtuse, conspicuously one-nerved. Labellum about 4 mm. long and 2 mm. wide, simple, linguiform, elliptical-oblong when spread out, obtuse, conspicuously three-nerved, deflexed above the middle, basal half with the sides erect, ecallose. Gynoste- mium very short, from the tip to base of column-foot scarcely 1 mm. long, terminating on each side in a triangular tooth ; foot 1.5 mm. long. Anther with a cucullate tip. Pollinia two.
Bulbophyllum nigroscapum is a member of that puzzling group of the genus for which Dr. Rudolf Schlechter proposed the section Aphanobulbon. It is closely related to B. anguipes Schltr., a native of the Celebes, and to B. dasypetalum Rolfe of the Philippine Islands. The section Aphanobulbon is character¬ ized by the racemose inflorescence and by extraordinarily re¬ duced pseudobulbs which are represented by disc-like thicken¬ ings on a more or less elongated rhizome. B. nigroscapum, as its name implies, possesses in dried specimens a blackish scape which forms a striking contrast with the scarious floral bracts. Whether or not this characteristic is to be noticed in fresh material is not reported by the collector. (Plate 114.)
Samoa, Tutuila Island, “Vao sosolo i le la’ou.” On trees at about 1800 feet altitude. Flowers cream color. Collected by M. C. Collarino, July 19, 1920. (No. 383 of W. A. Setchell’s distribution of plants collected on Tu¬ tuila Island under the auspices of the Department of Biology of the Car¬ negie Institution of Washington, D. C.)
[ 87 ]
ORCHID ACEiE
CAMPYLOCENTRUM Benth.
Campylocentrum panamense Ames sp. nov. Herba epiphy- tica, plus minusve 3 dm. longa, foliosa. Caules dependentes, plus minusve 3 mm. in crassitudine, rigidi ut videtur, vaginis foliorum obtecti. Folia valde coriacea, plus minusve 8 cm. longa, usque ad 22 mm. lata, apice inaequaliter bilobatavel oblique et profundere- tusa, lobis terminalibus 4-8 mm. longis, nervo medio subtus levi- ter carinato. Vaginae foliorum plus minusve 2 cm. longae, in sicco valde nervosae,coriaceae,rigidae.Pedunculicum racemo plus mi¬ nusve 5 cm. longi, validi, multiflori, laterales, racemo 4 cm. longo. Flores parvi, breviter pedicellati. Sepala lateralia 5.5 mm. longa, vix 1 mm. lata, oblongo-spathulata, carnosa, obtusa, trinervia, basi dilatata. Sepalum dorsale simile. Petala plus minusve 4 mm. longa, 1 mm. lata prope apicem, spathulata, prope medium con- stricta.Labellum 5.25 mm. longum, 1.5 mm. latum prope basim, triangulari-lanceolatum, utrinque lobulo prope basim obsessum, acuminatum, acutum, nervosum, in saccum 4 mm. longum pro- ductum. Saccus 1 . 5 mm. in diametro, scrotiformis, le viter curvatus, valde obtusus. Gynostemium generis. Capsula 10-11 mm. longa.
The type consists of a single plant in which the racemes are well advanced toward the production of mature capsules. From C. micranthum (Lindl.) Rolfe it differs in having longer leaves, longer racemes and more or less spathulate perianth segments. It differs from C. Schiedii Benth. in the form of the leaves and perianth segments.
Panama, near Gatun, Sutton Hayes 988, January 10, 1860. On trees.
(Type in Herb. N. Y. Bot. Gard.)
COELOGYNE Lindl
Coelogyne longirachis^m^ssp. nov. Radices glabrae, plus minusve 2 mm. in crassitudine, carnosae ut videtur, in sicco pal-
[ 88 ]
0RCHIDACEJ5
lidae. Rhizoma valid um, plus minus ve 8 mm. crassum. Pseudo- bulbi in serie, approximate usque ad 14 cm. alti, anguste ampul- lacei, plus minus ve 12 mm. in crassitudine infra medium, usque ad apicem attenuati, diphylli, in rhizomate crasso repenti plus minus ve 1 cm. inter se distantes; juniores vaginis coriaceis acutis fulti. Folia cum petiolo usque ad 24 cm. longa, plus minusve 4.5 cm. lata, elliptico-lanceolata, sensim acuta, utrinque attenu- ata, subtus prominenter quinquenervia, in petiolum 3 cm. Ion- gum contracta. Scapus initio foliis juvenilibus synanthus, folium demum superans, basi nuda, multiflorus, compressus, conspicue bialatus, antheseos finem versus plus minusve 80 cm. longus, gra¬ cilis, arcuatus, rachide fractiflexa. Bracteae inflorescentiae ovaria pedicellata superantes, cum floribus deciduae, supra flores arc- tissime imbricatae, plus minusve 8 mm. longae. Rachis usque ad 56 cm. longa, demum antheseos finem versus valde arcuata. Flores subrubri, succedanei. Pedicellus cum ovario 1.7 cm. lon¬ gus, in fructu 4.5 cm. longus. Sepala lateralia usque ad 7 mm. longa, 6 mm. lata, anguste elliptica, abrupte acuta, subcarnosa, in sicco subfusca. Sepalum dorsale simile. Petala 11 mm. longa, 1 mm. lata prope basim, usque ad apicem sensim attenuata, lev- iter curvata, linearia, acuta. Labellum vix 12 mm. longum, sub- carnosum, conspicue trilobatum; lobi laterales subquadrati, apice rotundati, circa 3.5 mm. longi ex apice loborum lateralium usque ad basim lobi medii, 3.5 mm. lati prope apicem, margine inae- qualiter triangulari-dentata ; lobus medius 6.5 mm. longus, 3 mm. latus, basi leviter rotundata, apice triangulari-acuto ; dis¬ cus callo alte sulcato, apice tridentato ornatus. Columna 8 mm. longa, carnosa, apicem versus alata, ala paucidentata.
Among Philippine species closely allied to C. bilamettata Lindl. from which it is distinguished by the broader lateral lobes
of the labellum which are blunt at the tip and irregularly
[ 89 ]
ORCHIDACEiE
toothed, and by the tridentate callus which extends from the base of the column onto the lower third of the middle lobe. The scapes attain an extraordinary length as the flowering season advances and would seem to droop in such a manner that the tip of the raceme must hang far below the point where the plants are attached. As many as sixty flowers are produced in succes¬ sion, approximately one centimeter apart on the rachis.
Philippines, Mindanao, Bukidnon Subprovince, Mahilucot River, M.
Ramos <$• G.Edano Bur. Sci. 38671, July 15, 1920. Along streams in forest.
4000 feet altitude.
Coelogyne Ramosii A vies sp. nov. Aff. C. sulphur eae (Bl.) Reichb. f. Herba epiphytica. Radices carnosae. Pseudobulbi plus minusve 4 cm. alti, anguste obpyriformes vel late semifusifor- mes, in sicco longitudinaliter rugosi; juniores vaginis ovatis, ob- tusis, 2-8 cm. longis, siccis papyraceis fulti. Folium usque ad 3 dm. longum, plus minusve 4 cm. latum, in petiolum lamina multo breviorem angustatum, coriaceum, late oblongum, usque ad basim attenuatum, apice abrupte rotundato, vix acutum, nervo medio solo subtus prominenter praeditum. Petiolus plus minusve 4 cm. longus,rigidus. Scapus proteranthus sive folii apice protruso synanthus, erectus, 1.5 dm. longus, sex ad octoflorus, folium non aequans. Bracteae naviculares, vix 2 cm. longae, pallidae, subscariosae, deciduae, ovarium longe pedicellatum superantes. Pedicellus cum ovario usque ad 1.7 cm. longus. Flores in racemo laxo dispositi, 13-14 mm. longi, subvirides, textura valde mem- branacea. Sepala subaequalia, 12.5 mm. longa, 3 mm. lata, acuta, basi manifeste saccata, extus valde carinata. Petala plus minusve 11 mm. longa, 1 mm. lata. Labellum 13 mm. longum, prope apicem 9 mm. latum, basi lata gynostemio insertum; lobi lat¬ erals parvi, triangulares, 1 mm. longi, acuti, prope medium labelli ; lobus medius ex isthmo distincto valde dilatatus, trans-
[ 90 ]
ORCHID ACE^E
verse reniformis, 8.5 mm. iongus, emarginatus. Carinae duae. Gynostemium e basi angustiore valde alato-dilatatum, conchi- forme, ala apice subtrilobata, margine erosa.
Coelogyne Ramosii is a near relative of C. sulphur ea (Bl.) Reichb. f. from which it differs in the smaller side lobes of the labellum and in the texture of the flowers.
Unfortunately my materials for comparative studies have been limited so that I have been unable to judge of the variation which is characteristic of Coelogyne sulphured. My studies of Javan specimens distributed by J. J. Smith, and of Bornean specimens, indicate that the lobes of the labellum should have some weight in determinations ; furthermore the Philippine ma¬ terial which I have proposed as a new species differs from both the Javan and Bornean representatives of C. sulphured in hav¬ ing extremely membranous floral segments which in dried ma¬ terial become so agglutinated that the usual methods of prepa¬ ration make examination extremely difficult.
Philippines, Mindanao, Bukidnon Subprovince, Mount Candoon, M.
Ramos # G. Edauo Bur. Sci. 389 JJ, June 27, 1920. Flowers green with
red interior. 5000 feet altitude.
CRANICHIS Sw.
Cranichis monophylla Lindl. Orch. Linden. (1846) 27; in Ann. & Mag. Nat. Hist. ser. 3, 1 (1858) 334 as to name, excl. plant. Sauroglossum monophyllum Griseb. Cat. PI. Cub. (1866) 269 as to synonymy, excl. plants. Spiranthes monophylla Cogn. in Urb. Symb. Antill. 6 (1909) 339 as to synonymy, excl. plants. Cyclopogon monophyllus Schltr. in Beihefte Bot. Centralbl. 37, Abt. 2 (1920) 391 as to synonymy, excl. Cuba.
Schlechter in his monograph of the Spiranthinae unaccount¬ ably referred to Cyclopogon the Crdnichis monophylld of Bind¬ ley, a Venezuelan plant collected by Linden in the Province
[ 91 ]
ORCHIDACEJE
of Merida. In his geographical note he refers the species to Cuba.
In Lindley’s herbarium at Kew there are three specimens mounted on the sheet of Cranichis monophylla, one, the type, from Venezuela, the others from Cuba, where they were collected by Charles Wright in 1856-57. The Cuban specimens are com¬ parable to Cranichis diphylla Sw. from the Blue Mountains of Jamaica. These Cuban plants are apparently conspecific with C. tenuijlora Griseb. Wright’s specimens belong to the series collected in eastern Cuba, and in pencil the number 621 has been inscribed on the label that accompanies them. There is absolutely no doubt as to the type, because Lindley glued the labels to the sheet in such a manner that the stem of the Vene¬ zuelan plant is covered by Linden’s label.
Cogniaux in Urban’s Symbolae Antillanae in the synonymy given under Spiranthes monophylla (Lindl.) Cogn. cites the ori¬ ginal description of Lindley’s Cranichis monophylla and in his ge¬ ographical notes gives Wright’s 621 and 1480 as the specimens that represent the Cuban species. Now, it is perfectly clear from the specimens in the Lindley Herbarium and in the Gray Her¬ barium, that Wright’s 621 collected in eastern Cuba in 1856-57 is a species of Cranichis. Wright’s 1480, on the other hand, on the evidence of two specimens preserved in the Gray Herbarium is a species of Spiranthes, but as the plants bearing this number were collected in 1859 they may be disregarded for the present. (See Spiranthes under S. Wrightii.)
The type of Cranichis monophylla Lindl., as has been stated already, was a Venezuelan plant. Why it should have been re¬ duced to synonymy because specimens from Cuba were later referred to it by Lindley is difficult to understand. It is true
that Grisebach in his Catalogus Plantarum Cubensium, pub-
[ 92 ]
ORCHIDACEdE
lished in 1866, referred Wright’s 1480 and 621 to Sauroglossum monophyllum Griseb., citing Lindley’s Cranichis monophylla, based on Wright’s Cuban specimens, as a synonym, but such a procedure had nothing to justify it, as C. monophylla was a species in good standing prior to the time when Lindley published his paper on W right’s Cuban collection of orchids, and as W right’s 621 which Lindley confused with C. monophylla is a true Cra¬ nichis.
It would seem that the transference of Cranichis monophylla Lindl. to Cyclopogon, made by Schlechter, resulted from a de¬ pendence on the synonymy cited by Cogniaux, and that attempts to identify Lindley’s Venezuelan species were based on a study of Wright’s 1480 rather than on Linden’s material from Venezuela. Cranichis tenuiflora Griseb. Cat. PL Cub. (1866) 268.
Wright’s specimens of 621 that I have examined are diphyl- lous, one leaf being much larger than the other, somewhat ovate- cordate in outline, with a petiole about 3 cm. long. The smaller leaf is lanceolate, acuminate, acute and gives to the plant a very distinctive appearance. When compared with Swartz’s type of Cranichis diphylla it becomes very apparent that the Cuban plants are clearly related to it. Wright’s 621 is the same as his 3292 which is the type number of Cranichis tenuiflora Griseb.
Cuba, Orientale, Wright 621 , 1856-57. Pinal, Monte Verde, February
1860-64; Wright 3292, Pinal, Monte Verde, February 1860-64.
DENDROBIUM Sw.
Dendrobium (§Pedilonum) appendiculoides Amessp. nov. Herba epiphytica, gracilis. Radices fibratae, carnosae. Caules a g- gregati, fasciculati, plus minusve 2.5 dm. alti, 2 mm. in crassitu- dine,multiarticulati, demumflavescentes. Folia disticha, usque ad 3.5 cm. longa, 5-6 mm. lata trans basim, apice inaequaliter bilo-
[ 93 ]
0RCHIDACEJ5
bata, demum retusa, in sicco subcoriacea, nervosa, plus minusve 1 cm. inter se distantia, vaginis inserta ; internodia leviter nodosa, demum fibrillis vaginarum dejectarum vestita. Racemi e caulibus aphyllis breves, pauciflori, plus minusve 1.5 cm. longi. Flores membranacei, tenues, albidi, roseotincti. Bracteae inflorescentiae minutae, triangulares, acutae, 1.75 mm. longae. Pedicellus cum ovario usque ad 18 mm. longus, gracilis, glaber. Sepala lateralia 5 mm. longa, triangulari-oblonga, subacuta, mentum trilobatum, valde elongatum, 16 mm. longum, 1 mm. in crassitudine, cum labello formantia. Sepalum dorsale elliptico-ovatum, obtusum, 5.25 mm. longum, trans medium 2 mm. latum. Petala 5 mm. longa, 2 mm. lata trans medium, elliptico-oblonga, obtusa, tri- nervia, membranacea, in sicco pellucida. Labellum simplex, lam¬ ina vel pars libera labelli oblanceolata e basi lata, ecallosa, 7 mm. longa, 3 mm. lata. Columna brevis, apice utrinque lobulata, lobulo rotundato.
Dendrobium appendiculoides is related to D. Hasseltii Lindl., but is a more slender and compact species with much smaller flowers. In habit the slender leafy stems resemble Appendicula lucbanensis Ames.
The mentum is very slender and about equally thick from base to tip, and is closed in front up to the point where the lamina of the lip begins to dilate.
Philippines, Mindanao, Bukidnon Subprovince, Mount Candoon, M.
Ramos 4" G. Edaiio Bur. Sci. 3894-0, June 27, 1920. Flowers pink and
white. On tree in forest. 5000 feet altitude.
Dendrobium (§ Desmotrichum) candoonenseAwe,?.sp.7zot;. AfF. D. conveoco Lindl. Rhizoma longe repens. Caules penduli, ramosissimi, in nodis infimis radicantes, fibrillis foliorum vetus- tiorum obsiti, articuli ultimi in pseudobulbos clavatos, 2-3 cm. longos, plus minusve 3 mm. crassos incrassati, monophylli. Pseu-
[ 94 ]
ORCHID ACE M
dobulbi usque ad basim attenuati, in sicco longitudinaliter sul- cati, flavescentes, plus minusve 3 cm. inter se distantes, caulibus subparallelis. Folia 4-8 cm. longa,4-7 mm. lata,ligulata, oblongo- linearia, coriacea, apice inaequaliter bilobata, lobulis obtusis in petiolum brevem contracta. Racemi ex axilla folii unici orien- tes, basi squamis scariosis obsiti, perbreves. Flores succedanei, semper singuli, fugacissimi, vix aperti. Pedicellus cum ovario circa 3 mm. longus. Sepala lateralia 6 mm. longa, mentum aper- tum, obtusum, 3 mm. longum formantia, oblique triangula, ob- tusa, 4 mm. lata trans medium. Sepalum dorsale 4-5 mm. lon¬ gum, circa 2.5 mm. latum, e basi lata ellipticum, subacutum. Petala membranacea, circa 5 mm. longa, linearia, subacuta, tri- nervia. Labellum 1 cm. longum, a basi lineari cuneatum, trilo- batum, lobo medio bifurcato, inter apices loborum lateralium 4 mm. latum, prope apicem 5 mm. latum ; lobi laterales minuti, valde obtusi, vix 1.5 mm. longi; lobus medius quadratus, deinde profunde bilobulatus vel bifurcatus, e basi usque ad apicem lobu- lorum 4.5 mm. longus, infra lobulos vix 3 mm. latus; lobuli 2 mm. longi, 2 mm. lati, subelliptici, obtusi ; lamellae in disco tres, quarum mediana brevissima; lamellae laterales in isthmo lobi medii valde curvatae, manifeste elevatulae. Gynostemium breve.
Dendrobium candoonense, among Philippine species, is most closely allied to D. pseudoconvexum Ames from which it is sep¬ arated by its oblong linear leaves, three carinae on the disc in¬ stead of two, and by the smaller lateral lobes of the labellum which are not porrect as in D. pseudoconvexum.
Philippines, Mindanao, Bukidnon Subprovince, Mount Candoon, M. Ramos Sp G. Edano Bur. Sci. 38716, June 27, 1920. On tree trunk on for¬ ested slopes. Flower with red stripes. 5000 feet altitude.
Dendrobium crumenatum Sw. in Schrad. Journ. Bot. 2 (1799) 237.
[ 95 1
ORCHIDACEtE
In the Philippine Islands the yellow stems of this plant are used for purposes of ornamental weaving. In the economic plant collection of the Bussey Institution of Harvard University there is a riding whip the handle of which is covered with interwoven strands taken from the stems of Dendi'obium crumenatum and from a species of fern, probably from Lygodium circinnatum (Burm.) Sw.
Dendrobium Schuetzii Rolfe in Orch. Rev. 19 (1911) 224; in Bot. Mag. (1913) t. 8495.
A specimen of this species from the Province of Surigao in northwestern Mindanao, where it was collected in company with D. Dearei Reichb. f., is in my herbarium. Although D. Schuetzii was originally published without exact localization, the geographical origin being given as the Philippine Islands, it is now known to be a native of Mindanao. Dendrobium Sanderae Rolfe, a near relative of D. Schuetzii, is a native of Luzon, where it has been found in the Province of Benguet and in Bontoc Subprovince. Dendrobium Dearei is represented in my herba¬ rium by specimens from Mindoro, and from Dinagat Island, a small island just north of Mindanao.
Dendrobium (§ Aporum) sinuosum Ames sp. nov. Radices fibratae, albidae. Caulis plus minusve 30 cm. altus, e basi tereti paulum complanatus, medium usque conspicue foliatus, interno- diis 4 cm. longis. Folia cum vaginis usque ad 5 cm. longa, 5 mm. lata, bene longiora quam altiora, lineari-lanceolata a latere visa, acuta, disticha, oblique ascendentia, supra medium caulis subito in foliolos complanato-subulatos decrescentia, foliolis plus mi¬ nusve 4 cm. inter se distantibus, cum vaginis plus minusve 2.5 cm. longis, pars caulis aphylla sinuosa. Flores prope apicem caulis succedanei ut videtur, albidi et purpurei. Glomeruli florum certe
pauciflori ; bracteae griseae, quam ovarium longius pedicellatum
[ 96 ]
ORCHIDACEJE
bene breviores. Pedicellus cum ovario plus minus ve 8 mm. lon- gus, gracilis. Sepala lateralia triangula, 8 mm. longa, mentum apertum 5.5 mm. longum formantia, acuminata, apice breviter subulata. Sepalum dorsale 8 mm. longum, 2 mm. latum infra medium, anguste lanceolatum, acuminatum, acutum, apice brev¬ iter subulatum. Petala membranacea, 6 mm. longa, plus mi- nusve 1 mm. lata supra medium, lanceolato-acuminata, e basi ob- longa. Labellum vix trilobatum, prope medium in unguem 4.5 mm. longum contractum ; lamina libera 6 mm. longa, 6 mm. lata trans basim, rotundato-triangula, subacuta, membranacea, mar- gine irregulariter denticulato et crenulato ; lineae tres breviter elevatulae per discum, in medio disci crista auctae. Gynoste- mium perbreve, 1 mm. altum in pedem 5 mm. longum produc- tum. Anthera subconica, postice retusa, basi utrinque in auricu- iam producta, antice obtusa, apice bilobata, curvata.
Among Philippine species most closely related to Dendrobium ventricosum Kranzl. which has a very different labellum and larger more erect leaves. In habit it bears some resemblance to the Celebesian JD. chrysotainium Schltr. which is a more robust species and is very different in the structure of the labellum.
The lower third of the stem is occupied by from six to thir¬ teen obliquely spreading distichous leaves of about equal length. Above the lower third of the stem the leaves become suddenly reduced to form linear limbs about five in number that are approximately 13 mm. long to the tip from the point of contact with the stem and scarcely 2 mm. wide. The stem that bears these reduced leaves is complanate and more or less sinuous, and near its summit gives rise to clusters of bracts from among which the flowers are produced in succession. In height the plants examined range from eighteen to thirty -five centimeters. The
anther is produced posteriorly, at the base, into a distinct auricle
[ 97 ]
ORCHIDACE.E
on each side, and at the shallowly bilobed tip is somewhat bent backward.
Philippines, Leyte, Tacloban, C. A. Wenzel 0996, May 5, 1916. Flower with white sepals and petals and a white, purple and yellow labellum. At sea level.
DENDROCHILUM BL
Dendrochilum (§Platyclinis) prodigiosum Ames sp. nov. Herba epiphytica. Radices fibratae, ramosae, glabrae. Pseudo- bulbi caespitosi, plus minusve 1 cm. longi, subfusiformes, in sicco flavidi, rugosi, monophylli. Folium cum petiolo 2.5-4.5 cm. longum, usque ad 11.5 mm. latum, anguste ellipticum, obtu- sum, bene marginatum, valde coriaceum. Petiolus 3-7 mm. lon- gus, rigidus. Pedunculus terminalis, cum racemo 5.5-6 cm. lon- gus, gracilis, usque ad racemum nudus, infra inflorescentiam bractea glumacea vestitus. Racemus densiflorus, 3 cm. longus, 1 cm. per medium. Pedicellus cum ovario plus minusve 2.5 mm. longus. Bracteae inflorescentiae glumaceae, nervosae, quadrato- ovatae, margine erosae, 5 mm. longae, ovarium excedentes. Se- pala lateralia 6 mm. longa, 3 mm. lata trans basim, lanceolata, acuta, cum apiculo sub apicem, carnosa, trinervia, mentum bre- vem formantia. Sepalum dorsale 5 mm. longum, valde concavum, ovato-lanceolatum,trinervium. Petala ligulata, vix 5 mm. longa, 1 mm. lata, subacuta, trinervia. Labellum carnosum, 6 mm. longum, subsaccatum, basi gynostemii insertum, utrinque prope basim lobulo minuto instructum; hypochilium 4 mm. longum, bicarinatum, oblongum, marginibus erectis, apice dilatatum; lobus terminalis 2 mm. longus, 3 mm. latus transverse ellipticus. Discus lobi terminalis tricarinatus. Columna plus minusve 2 mm. longa. Brachia erecta, recurva, acuta, e parte superiore gyno¬ stemii orientia, anguste linearia.
Dendrockilum prodigiosum is without close allies in the Phil-
[ 98 ]
ORCHIDACEJE
ippine Islands. The structure of the labellum sets it clearly apart from all other members of the genus. The lip is subsaccate at the base and produced into a long semitubular hypochil which is ter¬ minated by a transversely elliptical plate or epichil. On either side of the hypochil, near the margin, a thickened keel extends from the base of the lip and terminates in a distinct callus. The terminal plate is tricarinate, the middle carina being longer than its com¬ panions and reaching nearly to the tip of the lip. Where the hypo¬ chil joins the column there is a tiny triangular lobe on each side.
Philippines, Mindanao, Bukidnon Subprovince, Mount Lipa, M. Ra¬ mos 4* G. Edano Bur. Sci. 88521 , July 13, 1920. On tree in mossy forest. Flowers brownish red. 6500 feet altitude.
EPIDENDRUM L.
Epidendrum strobiliferum Reichb. f in Nederl. Kruidk. Arch. 4 (1858) 333.
This species has recently been collected in Guatemala. I do not find any record of this plant, heretofore published, that brings it within the Central American flora.
Guatemala, Department of Alta Verapaz, Cham a, Harry Johnson 861, October 15, 1920. On tree trunks, usually in sun. Flowers white. 900 me¬ ters altitude.
Epidendrum sulcatum Ames sp. nov. Aff. E. gracillimo Lehm. & Kranzl., speciei quamhabituet structura haec species conspicue simulat. Radices albidae, ramosae, glabrae, carnosae. Caules ramosi, 8.5-12.5 cm. longi, plus minusve 4 mm. in cras- situdine, quinquefoliati, foliis distichis, infra folia vaginis persis- tentibus, plus minusve 2.5 cm. longis, arete appressis, complana- tis, pars caulis aphylla usque ad 7.5 cm. longa. Folia coriacea, plus minusve 4.5 cm. longa, plus minusve 7 mm. lata, oblongo- linearia, acuta, nervo medio subtus carinato. Pedunculus elonga- tus, plus minusve 22 cm. longus, satis gracilis, basi bractea spatha-
[ 99 ]
ORCHIDACEdS
cea instructus, infra racemum bractea vaginanti, plus minusve
2 cm. longa, acuminata, acuta ornatus. Racemus usque ad 14 cm. longus, multiflorus, laxus, arcuatus, gracilis. Bracteae inflores- centiae elongatae, plus minusve 1 cm. longae, anguste lanceo- latae, valde acuminatae, acutissimae, ovarium aequantes vel ex- cedentes. Pedicellus cum ovario 1 cm. longus, gracilis, oblique ascendens ut videtur. Sepala lateralia plus minusve 1.3 cm. longa,
3 mm. lata, conspicue quinquenervia, nervis prominentibus, ex- tus nervo medio carinato,spathulato-lanceolata, acuminata, apice conduplicata. Sepalum dorsale 1.4 cm. longum, 2 mm. latum, lineari-oblongum,acutum, prominenter trinervium. Petala usque ad 12 mm. longa, 1 mm. lata supra medium, lineari-spathulata, acutissima, prope basim valde angustata, vix .5 mm. lata, nervo medio prominenti. Labellum 12-13 mm. longum, usque ad 7 mm. latum trans medium, valde trilobatum, circuitu hastatum, columnae appressum et cum eo in tubum connatum ; lobi later- ales semiorbiculares, margine denticulati, 4 mm. longi, vix 3 mm. lati ; lobus intermedius primo oblongus deinde triangulus,acutus,
4 mm. longus, 3 mm. latus trans basim, margine minute denticu- latus vel integer; discus prope apicem columnae conspicue bi- callosus, inter callos breviter carinatus, carina carnosa. Columna carnosa, 7 mm. longa, usque ad apicem ungui labelli adnata, in sicco alte bisulcata.
Closely allied with Epidendrum gracillimum Lehm. & Kranzl. which is much smaller in all its parts, with the lateral lobes of the labellum more conspicuously and coarsely dentate. The specific name alludes to the deep longitudinal grooves on the back of the column which are separated by a rounded rib.
Colombia, State of Cauca, headwaters of Rio Lopez, Rio Palo basin,
Tierra Adentro, H. Pittier 1060, January 1906. Epiphyte, flowers greenish
yellow. 2500-3000 meters altitude. (Type in U. S. Nat. Herb. No. 531282.)
[ 100 ]
ORCHID ACEdS
ERIA Lindl.
Eria (§ Eriurae) candoonensis Ames sp. nov. Herba epiphy- tica, plus minusve 3 dm. alta. Radices fibratae, ramosae. Caules erecti, plus minusve 8 mm. in crassitudine, rigidi, usque ad api- cem foliosi, vaginis foliorum tecti. Folia disticha, 8-11 cm. longa, 10-12 mm. lata, coriacea, articulata, oblonga, apice valde inae- quilobata, lobis obtusis. Vaginae foliorum 2 cm. longae, persisten- tes. Racemi terminales, in speciminibus nostris duo, plus minusve 1.5 dm. longi. Pedunculus infra racemum 4 cm. longus, dense lanuginosus, 2 mm. crassus. Racemus multiflorus, plus minusve 1 dm. longus. Bracteae inflorescentiae dependentes, plus minusve 7 mm. longae, lanceolatae, acuminatae, pubescentes, in sicco ver- ruculosae. Flores plus minusve 4 mm. inter se distantes. Pedi- cellus cum ovario 1-1.4 cm. longus, dense lanuginosus. Sepala lateralia 5.25 mm. longa, 3.25 mm. lata prope basim, carnosa, ovato-lanceolata, obtuse acuta, quinquenervia, extus dense pu- bescentia. Sepalum dorsale 5.75 mm. longum, 2 mm. latum trans medium, anguste elliptico-lanceolatum, obtusum, extus pubes- cens. Petala 5 mm. longa, 1.25 mm. lata, prope apicem spathu- lata, obtusa, trinervia. Labellum unguiculatum, plus minusve 6 mm. longum, trilobatum, in circuitu semiorbiculatum, 6 mm. latum ; unguis 1 mm. longus ; lobi laterales labelli 2 mm. longi, 1.5 mm. lati trans basim, semiovati, obtusi, lobum intermedium aequantes; lobus intermedius 2 mm. longus, 5.5 mm. latus, levi- ter retuso-apiculatus, transverse ellipticus. Calli sex, tres prope basim labelli, in disco utrinque callus complanatus, breviter cari- natus prope sinum, in medio lobi terminalis tuberculum permag- num, erectum, complanato-conicum. Columna brevis.
A very distinct species well characterized by its distichous, rigid leaves. The calli of the labellum in the less mature flowers are covered with a farinaceous substance that is formed by the
[ ioi ]
ORCHID ACEtE
breaking down of trichome-like structures at an early period in the development of the flowers. In mature flowers this farinaceous substance soon passes away. The central callus of the basal trio is curved backward and is much taller than its companions; the lat¬ eral calli are membranaceous, triangular, obtuse, purplish in color and posteriorly shortly carinate. The large fleshy callus on the mid¬ dle lobe is flattened laterally, is 2 mm. tall and is blunt at the apex.
Philippines, Mindanao, Bukidnon Subprovince, Mount Candoon, M. Ramos # G. Edano Bur. Sci. 389J/.1, June 27, 1920. Flowers whitish yel¬ low, labellum purplish. 5000 feet altitude.
Eria hirsutipetala Ames sp. nov. Rhizoma validum, to- mento rufo vestitum, radicans, circa 1 cm. per medium. Caules erecti, usque ad 16 cm. alti, plus minusve 6 mm. in crassitudine, in sicco longitudinaliter sulcati, pilosi, prope medium vagina inflata, 3.5 cm. longa, laxe appressa vestiti. Folia prope apicem caulis conferta, plerumque tria, usque ad 17 cm. longa, 1.5-2. 4 cm. lata, utrinque attenuata, coriacea, anguste elliptica, apice inaequaliter bilobata, lobulis obtusis, in petiolum brevem con- tracta, nervo medio subtus prominenti. Racemi quam folia brevi- ores, plus minusve 2 cm. longi, pauciflori, rubro-villosi, flori- bus plerumque glomeratis. Bracteae inflorescentiae circa 8 mm. longae, dense rufo-villosae,concavae,rotundatae, margine erosae, quam pedicellus cum ovario vix breviores. Flores circa 1 cm. longi, tomento rufo vestiti. Sepala lateralia circa 8 mm. longa, carnosa, triangulari-lanceolata, subacuta, intus pubescentia, extus villosa, mentum 4 mm. longum, obtusum formantia. Sepalum dorsale 9-10 mm. longum, oblongo-ellipticum, valde concavum, obtusum, extus dense villosum, intus pubescens. Petala valde carnosa, circa 8 mm. longa, 2 mm. lata, oblonga, extus pubescen¬ tia, intus hirsuta, apice attenuata et incrassata, obtusa vel sub¬ acuta. Labellum 8 mm. longum, usque ad 4.5 mm. latum prope
[ 102 ]
ORCHIDACEiE
basim, breviter unguiculatum, circuitu ovato-lanceolatum, obtu- sum, valde carnosum, extus pubescens. Columna 4 mm. longa, pubescens, in pedem 4 mm. longum producta.
This is a very remarkable species which differs conspicuously from its Philippine allies in the short stems with from two to four leaves near the summit, in the manner of species of section Hymeneria, in the stout rhizomes which at the base of young shoots are covered by closely appressed tubular bracts with a dense reddish tomentum, and in the fleshy petals, thickened at the tip, which are densely pubescent on both sides.
Philippines, Mindanao, Bukidnon Subprovince, Tangculan and vicinity, M. Ramos 8f G. Edano Bur. Sci. 89121^ July 2, 1920. On tree in mossy forest. Flowers yellow and coffee color. 5500 feet altitude.
Eria (§ Hymeneria) macera Ames sp. nov. Herba epiphytica. Rhizoma repens. Radices fibratae, ramosae, pubescentes. Caules aggregati, graciles, plus minusve 2 dm. longi, in sicco plus mi- nusve 2.5 mm. in crassitudine, pauciarticulati ut videtur, vaginis tubulatis vestiti, plerumque trifoliati. Folia plus minusve 12 cm. longa, usque ad 1 cm. lata, lineari-oblonga, in sicco charta- cea, utrinque attenuata, acuta, oblique erecta. Raeemi subapi- cales, laxiflori, floribus 8 mm. inter se distantibus, cum pedun- culo usque ad 5 cm. longi, plus minusve quadriflori, glabri. Brac- teae inflorescentiae usque ad 5 mm. longae, in sicco chartaceae, ovatae, acutae, margine erosae. Pedicellus cum ovario 6-13 mm. longus, gracilis, glaber. Sepala lateralia plus minusve 11 mm. longa, usque ad 2.5 mm. lata trans basim, triangulari-oblonga, usque ad apicem attenuata, acuta, mentum obtusum 2 mm. longum formantia. Sepalum dorsale simile, plus minusve 1 cm. longum, 3 mm. latum. Petala 9 mm. longa, 2-2.75 mm. lata, ligulata, acuta, trinervia, apice incrassata. Labellum 4 mm. longum, 3 mm. latum infra medium, 2 mm. latum trans lobum
[ 103 ]
ORCHIDACE,®
medium, circuitu ovatum, vix unguiculatum, obtusum, triloba- tum ; lobi laterales e medio labelli orientes, minuti, obtusi ; lobus medius ellipticus, 2 mm. longus, margine ciliolata, per medium disci carina carnosa usque ad basim lobi medii extendit, utrinque prope marginem labelli carinis duabus prostratis. Columna 2 mm. longa in pedem 1 mm. longum producta. Pollinia octo.
The elongated stems with as many as six tubular sheaths which loosely invest them at the period of anthesis ; the slender peduncles and few-flowered loose racemes which arise from among the sheaths just below the obliquely erect, oblong-linear leaves, give to this species a very distinct aspect among its near¬ est allies in the Philippines. From .Ena retroflexa Lindl. it differs in the few-flowered racemes, three-lobed lip with distinct carinae and in the linear-oblong leaves.
Philippines, Mindanao, Bukidnon Subprovince, Mount Lipa, M. Ramos
4* G. Edafio Bur. Sci. 38512 , July 12, 1920. On trees on mossy forest slope.
Flowers yellowish white. 6400 feet altitude.
Eria (§ Hymeneria) microchila Ames sp. nov. Planta ha- bitu E. philippinensis. Caules ramosi, vagi, oblique ascendentes, cataphyllis laxe imbricantibus, obtusis, in sicco fulvis tecti. Caules foligeri usque ad 1.5 dm. longi, super vix 1 cm. in cras- situdine, rugosi, infra medium attenuati, plus minusve 5 mm. crassi. Vaginae caulis in sicco inflatae, plus minusve 2 cm. longae, in sicco chartaceae, prope basim caulis imbricatae, apice foliatae. Folia plus minusve 1.5 dm. longa, 10-18 mm. lata, utrin¬ que attenuata, acutissima, in sicco chartacea, oblique erecta, quinque ad septem. Racemi nonnulli ex axillis superioribus, plus minusve 1.5 dm. longi, pubescentes, tomento brunneo ves- titi. Bracteae racemi reflexae, plus minusve 6 mm. longae, lanceo- latae, acutae, glabrae. Pedicellus cum ovario arcuatus, usque ad 9 mm. longus, dense pubescens, tomento brunneo tectus. Flores
[ 104 ]
ORCHID ACE
extus sparsim brunneo-pubescentes. Sepala lateralia triangula, 9 mm. longa, mentum 3 mm. longum formantia, margine infra medium ciliata. Sepalum dorsale ligulatum, 1 cm. longum, plus minusve 1.5 mm. latum, margine ciliata. Petala oblonga, utrin- que leviter attenuata, 9 mm. longa, 2.25 mm. lata, obtusa, tri- nervia. Labellum breviter unguiculatum, 3 mm. longum, 2 mm. latum trails basim laminae, cordato-lanceolatum, apice obtu- sum, utrinque prope basim callo breviter carinato instructum; unguis vix .5 mm. longus. Columna 1.5 mm. longa in pedem 2.5 mm. longum producta.
Eria microchila is closely allied to E. philippinensis Ames, but is distinct from it in the smaller flowers. From E. Hutchin- soniana Leavitt it is in part separated by the different foliage and bicallose lip.
In young flowers of E. microchila the lip is distinctly bicallose near the base, close to the margin, and supplementary calli maybe detected near the middle nerve in front of the claw. It would seem that these calli are evanescent, as in older flowers they are wanting.
Philippines, Luzon, Bontoc Subprovince, Mount Cauca, M. Ramos <§•
G. Edano Bur. Sci. 37997, March 7, 1920. On tree trunk in mossy forest.
5600 feet altitude.
Eria ( § Aeridostachyae) propinqua A mes sp. nov. AfF. E. aeri- dostachyae et E. Whitfordio. Radices fibratae, ramosae, glabrae. Pseudobulbi cauliformes, 3-4 cm. longi, plus minusve 1.5 cm. in crassitudine, diphylli, vaginis magnis laxisque omnino tecti. Folia plus minusve 2.5 dm. longa, usque ad 3.9 cm. lata, valde coriacea, ligulata, utrinque angustata, quam racemus longiora, nervo medio prominenti. Pedunculus cum racemo 1.5 dm. longus, pubescens. Rachis inflorescentiae dense pubescens, to- mento flavido perbrevi. Racemus dense multiflorus, cylindra- ceus, arcuatus, 1.8-1. 9 cm. trans medium. Bracteae inflorescen-
L 105 ]
ORCHID ACEdE
tiae flores subtendentes, squamiformes, late ovatae, cucullatae, 1 mm. longae. Pedicellus cum ovario 3.5 mm. longus ad basim menti ; mentum 2 mm. longum. Flores extus dense flavido- pubescentes. Sepala lateralia vix 3 mm. longa, 4 mm. in diame- tro maximo, falcata, subacuta. Sepalum dorsale 3.5 mm. longum, valde concavum vel cochleatum, ovato-oblongum, trinervium. Petala ligulata, curvata, 3 mm. longa, 1 mm. lata, obtusa, tri- nervia. Labellum simplex, 4 mm. longum, 2 mm. latum, ali- quid incrassatum, oblongo-ovatum, acutum, margine crenulata, tri- vel quinquenervium, ecallosum, vix unguiculatum. Columna in pedem crassum producta.
Smaller in all floral parts than E. Wliitfordii Leavitt to which it is closely allied. In dried specimens the flowers appear to have been yellowish with purplish petals. In herbarium specimens easily separated from its near allies by the yellowish rather than rusty pubescence of the peduncle and flowers. From E. Mearnsii Leavitt clearly distinguished by the proportions of the labellum.
Philippines, Mindanao, Bukidnon Subprovince, Mount Candoon, M.
Ramos <$(■ G. Edano Bur. Sci. 38939 , June 27, 1920. On tree in forest.
5000 feet altitude.
Eria (§Hymeneria) vagans Ames sp. nov. Planta habitu E. philippinensis Ames. Caules ramosi, vagi, lignosi, plerumque abbreviati, demissi ut videtur, caulibus foligeris plus minusve 4 cm. inter se, usque ad 1 dm. longis, cataphyllis laxe tectis. Folia plerumque anguste lanceolata, approximata, plus minusve 7 cm. longa, usque ad 14 mm. lata, utrinque attenuata, subcori- acea, apice inaequaliter bilobata. Racemi ex axillis superioribus, foliis longiores, plus minusve 8 cm. longi, laxiflori, rachide pu- bescenti, pilis rufis. Bracteae inflorescentiae plus minusve 4 mm. longae, ovato-ellipticae, acutae, subglabrae vel sparsim pubes- centes, trinerviae. Pedicellus cum ovario 6-9 mm. longus, leviter
[ 106 ]
ORCHIDACEtE
arcuatus, tomento denso rufo. Flores 10 mm. longi, prope basim sepalorum pubescentes, plus minusve 8 mm. inter se distantes. Sepala lateralia usque ad 10 mm. longa, mentum 3 mm. Ion- gum formantia, triangularia, subacuta. Sepalum dorsale 7 mm. longum, 3 mm. latum, ligulatum, ad basim vix attenuatum, ob- tusum. Petala 9 mm. longa, vix 3 mm. lata, anguste lanceolata, obtusa, trinervia, in sicco pellucida. Sepala petalaque mem- branacea. Labellum 5 mm. longum, trans basim plus minusve 4 mm. latum, late ovatum, obtusum, margine irregulariter den- ticulata, utrinque supra basim callo carinato ornatum; unguis perbrevis. Columna circa 3 mm. longa in pedem 3 mm. longum producta. Anthera transverse elliptica.
JEria vagans belongs to the same category with E. philippi- nensis Ames and E. dagamensis Ames, but differs from them in its smaller flowers and much shorter racemes. In general aspect it is a more compact species than either E. philippinensis or E. dagamensis, the flowering stems being much shorter than in E. dagamensis to which it is most closely allied.
From the specimens examined E. vagans is of decidedly straggling habit, the flowering stems arising as shortened growths from an elongated, primary, leafless stem.
Philippines, Luzon, Bontoc Subprovince, Bauco, Vanoverbergh 3896 , January 12, 1914. Epiphyte on trees. Plant 5 dm. high. 1700 meters alti¬ tude; Vanoverbergh 2223, January 1913. Here the following specimens should be referred, Luzon, Province of Benguet, Baguio, Elmer 8801, March 1907.
HABENARIA Willd.
Habenaria bicornis Lindl. Gen. & Sp. Orch. (1835) 309. Originally described from Cuban material. This species has , not been reported from Central America heretofore.
Panama, Canal Zone, Las Sabanas, H.Pittier 6792, September 10, 1914.
[ 107 ]
ORCHID ACEiE
LIPARIS L. C. Rich .
Liparis (§ Distichae) magnicallosa Ames sp. nov. Aff. L. propinquae Ames. Herba epiphytica. Radices fibratae, glabrae, albidae,prope pseudobulbos orientes.Rhizoma repens, valde elon- gatum, plus minusve 2 mm. in crassitudine, vaginis scariosis, tubulatis vestitum. Pseudobulbi plus minusve 1 cm. alti, pyri- formes, usque ad 7 mm. in crassitudine trans basim, 2-4 cm. inter se distantes, monophylli, in sicco valde rugosi, erecti. Folium plus minusve 12 cm. longum, usque ad 9 mm. latum, in sicco subchartaceum, utrinque attenuatum, acutum, lineari-oblon- gum,erectum, prominenter quinquenervium,nervulis interjectis. Scapus erectus, plus minusve 12 cm. longus, bialatus, infra race- mum nudus, complanatus, circa 1 mm. per medium, folio bre- vior. Racemus usque ad 4 cm. longus, multiflorus, floribus flavi- dis, perpaucis eodem tempore apertis. Bracteae inflorescentiae distichae, usque ad 6 mm. longae, circa 2 mm. latae a latere visae, demum deciduae. Pedicellus cumovario usque ad 15 mm. longus, valde gracilis, bractea subtendenti multo longior. Sepala latera- lia dependentia, 5 mm. longa, vix 2 mm. lata, oblongo-ellip- tica, acuta, apiculata, extus nervo medio prominenti, in apicu- lum producta. Sepalum dorsale dependens, 5 mm. longum, circa 1 mm. latum, anguste elliptico-lanceolatum, apiculatum, extus nervo medio prominenti. Petala dependentia, 4 mm. longa, .5 mm. lata prope apicem, linearia, acuta, usque ad basim attenu- ata, membranacea. Labellum 4 mm. longum, usque ad 2 mm. latum trans medium, dependens, oblongo-ovatum, leviter obcu- neatum, obtuse et breviter apiculatum, haud retusum, infra api¬ cem 1.25 mm. latum, 2.25 mm. latum trans basim, ad basim callo permagno ornatum, eo usque ad rostellum columnae pertinenti, plus minusve 1 mm. alto, erecto, transverse 1 mm. lato, curvato,
utrinque callo minore, complanato, carinato instructum. Co-
[ 108 ]
ORCHID ACEJE
lumna 2 mm. alta, valde arcuata, utrinque prope apicem breviter alata. Anthera antice truncata.
In habit similar to L. cyclostele Schltr. and L. gibbosa Finet. From L. propinquci Ames it differs conspicuously in the form of the labellum and petals. Liparis Aviesiana Schltr. is very similar in habit, but is clearly set apart by the cuneate labellum which is much broader at the tip than at the middle.
Philippines, Mindanao, Bukidnon Subprovince, Mount Candoon, M. Ramos A G. Edailo Bur. Sci. 38789, June 27, 1920. On trees on forested slopes of the mountain. 5000 feet altitude.
Liparis (§ Distichae) prava^wes sp.nov. Herba epiphytica usque ad 16.5 cm. alta, gracilis. Radices fibratae, prope pseudo- bulbos orientes. Rhizoma repens, lignosum, plus minusve 2 mm. in crassitudine, vaginis scariosis, tubulatis tectum. Pseudobulbi plus minusve 5 mm. longi, usque ad 2.5 cm. inter se distantes, pyriformes, erecti, in sicco rugosi, monophylli, juniores vaginis inclusi. Folium usque ad 11 cm. longum, 3-6 mm. latum, line- are, apiculatum, apiculo brevi, nervo medio subtus prominenti. Scapus usque ad 8 cm. longus, bialatus, infra racemum nudus, 1 mm. per medium, folio brevior. Racemus plus minusve 1 cm. longus, multiflorus. Flores bicolores, perpauci eodem tempore aperti. Bracteae inflorescentiae distichae, usque ad 8 mm. longae, vix 2 mm. altae, persistentes. Pedunculus cum ovario 8-1 1 mm. longus, gracilis, quam bracteae inflorescentiae longior. Sepala lateralia dependentia, 6-7 mm. longa, vix 3 mm. lata trans me¬ dium, ovato-lanceolata, extus nervo medio prominenti, in apicu- lum producto. Sepalum dorsale 5-6 mm. longum, 2 mm. latum infra medium, lanceolatum, extus nervo medio prominenti in apiculum producto. Petala erecta, usque ad 5 mm. longa, 1 mm. lata, ligulato-oblonga, basi angustata, apice rotundata, margine revoluta. Labellum 5.5 mm. longum, usque ad 3.5 mm. latum
[ 109 ]
ORCHIDACEiE
trans basim, 2 mm. latum trans medium, vix 3 mm. latum trans apicem, supra basim gibbosum, abrupte deflexum, circuitu late panduratum, apice retuso-apiculato, prope basim callo lobato ornatum. Columna carnosa, 2.5 mm. alta, prope medium dilatata; alae terminales in dentes triangulares, porrectos reductae. An- thera antice obtusa.
This species differs very much from L. gibbosa Finet in the structure of the column, the wings being reduced to triangu¬ lar teeth which are somewhat deflexed at the acute apex, and directed forward. Among Philippine species L. propinqua Ames is most similar to L. prava in the structure of the flower, but the wings of the column in that species are blunt, and the la- bellum when spread out is not conspicuously constricted at the middle. In habit the plants of L. prava resemble those of L. Amesiana Schltr., but the floral differences are great
Philippines, Leyte, Jaro, Conpagal, C. A. Wenzel 0718 , November 24, 1914. Sepals and petals flesh color, labellum flesh color and orange. Epi¬ phyte. 800 meters altitude.
Liparis (§Distichae) propinqua A messp. nov. Radices fibra- tae, albidae, graciles. Rhizoma lignosum. Pseud obulbi approxi¬ mate lageniformes, monophylli, 2-2.5 cm. longi, vaginis folia- ceis scariosis subtenti. Folium 7-16 cm. longum, usque ad 1 cm. latum, lineari-oblongum, acutum, quam scapus longius, rigidum, subcoriaceum, basim scapi plus minusve vaginans. Sca¬ pus bialatus, cum racemo usque ad 12 cm. longus, gracilis. Ra- chis brevis, 1-2 cm. longa. Bracteae inflorescentiae distichae, ap- proximatae, conduplicatae, plus minusve 4 mm. longae, acutae. Flores succedanei,parvi, aurantiaci. Pedicellus cum ovario usque ad 1 cm. longus, gracilis, glaber, leviter bialatus. Sepala latera- lia 4 mm. longa, plus minusve 2 mm. lata, oblongo-elliptica, acuta, apice incrassata, extus carinata, trinervia, dependentia.
[ no ]
ORCHID ACEiE
Sepalum dorsale elliptico-lanceolatum, 4 mm. longum, plus mi- nusve 1.5 mm. latum, apice leviter incrassatum, extus leviter cari- natum. Petala 4 mm. longa, 1 mm. lata prope apicem, spathu- lata, valde obtusa vel obscure apiculata, apiculis obtusis. Label- lum sessile, 3 mm. longum, plus minusve 2 mm. latum prope basim, in circuitu late ovatum, mucronatum, glabrum; callus pulvinatus, transversus, antice bilobus, ad basim columnae co- haerens. Columna suberecta, lamina vel ala longitudinalis utrin- que columnae latera percurrens. Anthera ovata, antice subros- trata, obtusa. Pollinia rotundato-triangularia.
A close ally of Liparis gibbosa (Bl.) Finet, from which it is in part distinguished by the smooth labellum, by the pseudobulbs being less remote from one another, and by subtle differences in the structure and consistency of the flowers which are quite apparent when the two species are compared side by side. In the Javan L. gibbosa the labellum in specimens that I have studied is minutely, although conspicuously, papillose. In both species the labellum is similar in outline and in having the characteris¬ tic constriction at the sides which gives a pandurate aspect to the lamina when in natural position. It would seem that the Philip¬ pine specimens referred to L . disticha belong here.
Philippines, Bancalan Island, C. M. Weber Oil, October 14, 1916. Epi¬ phyte. Flower orange. At sea level.
MALAXIS Soland. ex Sw.
Malaxis bracteosa Ames sp. nov. Aff. M. Junghuhnio (J. J. Sm.) Ames. Herba epiphytica, cum racemo usque ad 30 cm. alta. Radices fibratae, elongatae. Pseudobulbi vel caules abbreviati, usque ad 3-4 cm. alti, basi tumida, plus minusve 1 cm. per me¬ dium, vaginis scariosis tecti, paucifoliati. Folia lanceolata, plus minusve 10 cm. longa, usque at 4 cm. lata infra medium, conferta,
[ HI ]
ORCHIDACEiE
acuminatissima, acuta, in petiolum latum, vaginantem transeun- tia, valde chartacea, in sicco pellucida. Scapus elongatus, usque ad 26 cm. altus, gracilis, fere usque ad basim bracteatus, multi- florus, floribus parvis, flavidis. Bracteae inflorescentiae lineares, 3-8 mm. longae, vix 1 mm. latae, patentes, margine scabridius- cula. Racemus plus minusve 20 cm. longus, cylindraceus, circa 1.5 cm. per medium. Pedicellus cum ovario circa 5 mm. longus, gracilis, quam bractea subtendens paulo brevior. Sepala lateralia circa 2 mm. longa, plus minusve 1 mm. lata, oblongo-elliptica, manifeste uninervia, deflexa. Sepalum dorsale simile, 2 mm. longum, erectum. Petala 2 mm. longa, linearia, obtusa, patentia. Labellum trilobatum, e basi columnae ad apicem lobi medii 2 mm. longum, inter apices loborum lateralium 2 mm. latum; lobi laterales antice rotundati, post columnam in auriculas 2 mm. longas, triangulas producti, auriculis obtusis. Columna 1 mm. longa, auriculis obtusis, erectis.
Malaxis bracteosa is a near ally of M. Junghuhnii (J. J. Sm.) Ames which it resembles very closely in the shape of the label¬ lum.
Philippines, Mindanao, Bukidnon Subprovince, Mahilucot River, M. Ramos &f G. Edaho Bur. Sci. 38658, July 15, 1920. Along streams in forest on tree. 4200 feet altitude.
NOTYLIA Lindl.
Notylia panamensis Ames sp. nov. Herba epiphytica, rhi- zomate valde abbreviato, radicibus fibratis, elongatis, flexuosis, glabris. Pseudobulbi plus minusve 2 cm. longi, complanati, com- pressi, monophylli, vaginis foliatis duabus obtecti. Folia plus minusve 1.5 dm. longa, 31-36 mm. lata, coriacea, nervo medio conspicuo, oblonga, utrinque vix attenuata, apice inaequaliter et obtuse bilobulata, basi in petiolum plus minusve 1 cm. longum conduplicatum contracta. Pedunculus infra racemum plus mi-
[ ]
ORCHIDACEiE
nusve 7 cm. longus, bracteis nervosis, setaceis obsessus. Racemus densiflorus, 12.5 cm. longus, in sicco usque ad 22 mm. per me¬ dium. Bracteae inflorescentiae in sicco scariosae, triangulari-line- ares, valde acuminatae, setaceo-acutae, trinerviae, plus minusve 4.5 mm. longae. Pedicellus cum ovario 5-6 mm. longus, gracilis. Flores albidi. Sepala lateralia 8 mm. longa, usque ad apicem in laminam lineari-lanceolatam connata, in sicco conspicue biner- via. Sepalum dorsale 7 mm. longum, 5 mm. latum, valde conca- vum, ellipticum, obtusum. Petala elliptico-lanceolata, acuta, 7 mm. longa, usque ad 2.5 mm. lata, basi leviter cuneata. Labellum sagittatum, basi carinato-callosum, cum ungue 6 mm. longum, 3 mm. latum trans basim ; unguis 1 mm. longus. Columna tere- tiuscula, glabra, 3 mm. longa, rostello in gibbum incrassato. Pol- linia generis.
The large flowers in a crowded drooping raceme give to this species a very distinct aspect from that of the other Central American species.
Panama, Marraganti and vicinity, R. S. Williams 977, April 3-9, 1908.
10-200 feet altitude. (Type in Herb. N. Y. Bot. Gard.)
OBERONIA Lindl.
Oberonia linearifolia Amessp. nov. AfT. O. neglectae Schltr. Herba parvula, radicibus fibratis, albidis, flexuosis. Caules abbre¬ viate plus minusve 1 cm. longi, vaginis foliorum omnino obtecti, complanati. Folia quinque ad decern, anguste ensiformia, linearia a latere visa, patentia, acuta, disticha, prope basim vaginantia, 2.5-6 cm. longa, 1.3 mm. lata, quam racemus multo breviora, oblique erecta. Racemus floribus in genere inter minimos, 5-7 cm. longus, gracilis, 2 mm. per medium, densiflorus. Bracteae infra racemum scariosae, lineari-setaceae, arete appressae Brac¬ teae inflorescentiae triangulares, acuminatae, valde acutae. Se-
[ US ]
ORCHIDACEJS
pala lateralia vix 1 mm. longa, subacuta, ovata, usque ad medium connata, membranacea, pellucida. Sepalum dorsale simile. Petala anguste elliptica vel oblonga, obtusa vel subacuta, obscure den- ticulata, .5 mm. longa. Labellum trilobatum, apice bilobulatum, mucrone obtuso interjecto, lobulis angustatis, setaceis, e basi la- belli usque ad lobulum terminalem 1 mm. longum, inter apices loborum lateralium 1 mm. latum ; lobi laterales basilares, trian¬ gulares vel subquadrati, margine plerumque inaequali. Columna minuta.
Closely allied to Oberonia neglecta Schltr. from which it dif¬ fers in the apical lobe of the labellum, in the proportionately shorter lateral lobes and in the slightly larger flowers. In habit O. linearifolia is very similar to O. neglecta Schltr. from Borneo and to O. potamophila Schltr. from Sumatra.
The terminal portion of the labellum resembles very closely the labellum of O. Mannii Hook. f. as illustrated in leones Plan- tarum t. 2008, f. 2. When one bears in mind the great range of variation that is characteristic of the very delicate lip-lobes in the microscopically small flowers of the alliance to which O. lineari¬ folia and O. neglecta belong, it may well be asked if they do not represent the same species. I originally identified my material as conspecific with O. neglecta. Material was submitted to Dr. Schlechter for comparison with his type specimen. On his assur¬ ance that my material did not agree with O. neglecta and rely¬ ing on a sketch and analysis of the flower of the type, it was decided to propose a new species. (Plate 114.)
Borneo, Sarawak, Native Collector 921. (Collected through the Sarawak
Museum for the Bureau of Science, Manila, Philippine Islands.)
Oberonia lipensis Ames sp. nov. Planta epiphytica, minima, vix caulescens, cum racemo usque ad 8.5 cm. alta. Radices fibra- tae, glabrae. Caules abbreviati, vaginis foliorum tecti, valde com-
[ HI ]
QRCHIDACEiE
planati. Folia disticha, anguste ensiformia, 16.5-42 mm. longa, usque ad 4 mm. lata a latere visa, acuta, infra medium haud attenuata, coriacea, oblique ascendentia, scapo multo breviora. Scapus cum racemo usque ad 7 cm. longus, gracilis, multiflorus, sublaxiflorus, racemo 4.5 mm. per medium, rachide glabra. Brac- teae inflorescentiae 2 mm. longae,ovato-lanceolatae,acuminatae, acutae, margine erosa. Flores flavidi, parvi. Pedicellus cum ova- rio 1.5 mm. longus. Sepala lateralia 1 mm. lata, valde concava, late ovata, acuta, reflexa, in sicco pellucida. Sepalum dorsale sim¬ ile. Petala 1 mm. longa, anguste lanceolata, obtusissima, margine erosa. Labellum 1.5 mm. longum, valde concavum vel leviter sac- catum, basi cordata in lobulos producta, lobulis suberectis obtu- sis; pars terminalis labelli bilobata, lobis quadratis, mucrone obtuso, lato interjecto. Columna minuta.
Oberonia lipensis is allied to O. benguetensis Ames which has a very different labellum, the terminal lobes being very short. It is also allied to O. minima Ames, which is a much smaller spe¬ cies with acute lobes at the tip of the labellum. In habit not un¬ like O. linearifolia Ames and O. neglecta Schltr., but with a very different labellum.
The base of the labellum is distinctly cordate, the rounded basal lobes or rather auricles directed backward. In front of the column the lamina of the lip is deeply concave or subsaccate, at the middle it is constricted and terminates in two divergent quadrangular lobes which are truncate at the tip.
Philippines, Mindanao, Bukidnon Subprovince, Mount Lipa, M. Ramos <Sf G. Edano Bur. Sci. 38535, July 13, 1920. On tree in mossy forest. 6600 feet altitude.
Oberonia minutissima Ames sp. nov. Herba epiphytica, 6- 10 cm. alta, gracilis, acaulescens. Radices fibratae. Folia basi im- bricata, linearia a latere visa, usque ad 10 cm. longa, plus mi-
[ US ]
ORCHID ACEJH
nusve 5 mm. lata, oblique erecta, anguste ensiformia, equitantia, folia exteriora quam folia interiora multo breviora, 2-4.5 cm. longa. Pedunculus quam folia brevior, 6-8 cm. longus, 2-4 mm. per medium, flexuosus vel arcuatus, longitudinaliter plurialatus. Bracteae infra inflorescentiam lineares, acuminatissimae, conges- tae, numerosae. Bracteae inflorescentiae plus minus ve 2 mm. longae, integrae, lineares, apice longe acuminatae, acutissimae, flores excedentes. Flores verticillati, verticillis plus minusve 2 mm. inter se distantibus. Pedicellus cum ovario 1 mm. longus, subgracilis, glaber. Sepala lateralia .75 mm. longa, .25 mm. lata trans basim, late triangulari-ovata, acuta, pellucida, reflexa. Se- palum dorsale simile. Petala .5 mm. longa, vix .25 mm. lata, el- liptica, apice rotundata, margine crenato-dentata. Labellum .5 mm. longum, .25 mm. latum trans lobos laterales, conspicue tri- lobatum ; lobi laterales alte et inaequaliter quadri- vel quinque- dentati, patentes, .25 mm. longi; lobus medius .25 mm. longus, e basi oblonga leviter dilatatus, in lobulos duos, divaricatos productus. Columna minuta, apice valde dilatata, pellucida.
Although much more slender in all its parts and with much smaller flowers, this species is closely allied to O. cylindrica Lindl.
Philippines, Mindanao, Province of Surigao, Placer, C. A. Wenzel 1030, July 16, 1916. Epiphyte in forest. Flowers light green. 150 meters altitude.
PLEUROTH ALLIS R. Br.
Pleurothallis consimilis Ames sp. nov. Herba epiphytica. Rhizoma repens, in sicco curvatum et angulatum, vaginis in fibras solutis vestitum, radicibus flexuosis, glabris, maxima parte sub caulibus secundariis. Caules secundarii 1-1.5 cm. inter se distantes, plus minusve 2 cm. alti, valde compressi, prominen- ter canaliculati, uniarticulati, folio breviores, dimidio basali va-
[ 116 ]
ORCHID ACEiE
gina in fibras soluta vestiti. Folia elliptica vel oblongo-elliptica, 2. 5-3. 5 cm. longa, 8-13 mm. lata, coriacea, apice minute biden- tato et apiculato, basi cuneata, in sicco multinervosa, nervo me¬ dio subtus carinato. Pedicelli uniflori, singuli vel bini, basi va- ginis in fibras solutis vestiti, in parte superiore vagina tubulari donati. Flos a sepali dorsalis apice ad labelli apicem 1.3 cm. longus, membranaceus. Sepala lateralia oblique semiovalia, 5 mm. longa, 2.5 mm. lata trans medium, obtusa, trinervia. Se- palum dorsale lineari-oblongum, 8 mm.longum, 1.45 mm. latum sub apice, acutum, trinervium. Petala tenuiora, anguste rhom- boideo-lanceolata, 3.8 mm. longa, .8 mm. lata, acuta vel obtusa, uninervia, marginibus superioribus serrulatis. Labellum triloba- tum, hastatum, unguiculatum ; unguis basi callo convexo omnino obtectus ; lamina 3.8 mm. longa, 2.5 mm. lata trans lobos latera¬ ls, basi callo magno, convexo; lobi laterales basales, minuti, patentes, ovato-triangulares; lobus medius deltoideus 2.9 mm. longus, 1.8 mm. latus, apice late obtusus, margine ciliata, dimi- dio basali callis duobus remotis. Gynostemium gracile, 3.2 mm. longum, apex alis tribus dentatis praeditus.
Near allies of this species are Pleurothallis jimbriata Lindl., which has smaller flowers and secondary stems that are much longer than the leaf, and P. papillosa Lindl. in which the label¬ lum and petals are entire. P. josephensis Rodr. is similar in habit, but is more robust in all its parts.
Trinidad, B. W. I., Valencia, N. L. Britton, E. G. Britton T. E. Hazen
10%6, March 1920. On forest trees.
Pleurothallis Hitchcockii Ames sp. nov. Herba epiphytica. Rhizoma abbreviatum, radicibus numerosis, albidis, filiformibus. Caules caespitosi, approximati, prope basim uniarticulati et ni- gromaculati et vaginis duabus tubularibus arete cincti. Folium ellipticum, apice obtusum, basi cuneato-rotundatum, coriaceum,
[ in ]
ORCHIDACEiE
nervo medio subtus carinato. Petiolus strictus, canaliculatus. Pedicelli nonnulli, aggregati, uniflori, bractea scariosa nervosa in fibras soluta inclusi. Sepala lateralia tertia parte inferiore connata, anguste lanceolata, sensim longe acuminata, apicibus condupli- catis, basi concava. Sepalum dorsale lanceolatum, longe acumi¬ natum, basi valde concavum. Petala multo minora, lanceolato- linearia, prope basim lobo brevi porrecto. Labellum subhasta- tum, apice obtusum, basi utrinque lobulo anguste falcato, ob- tuso, valde incurvo et disco incumbenti praeditum ; discus inter lobos basales callo parvo ornatus. Columna minuta, basi val- dissime alata, ala subito angustata, apice triloba, lobo medio den¬ ticulate.
Plant epiphytic, up to 16.6 cm. high, the rhizome abbreviated and conspicuously supplied with whitish, flexuose fibrous roots. Secondary stems about 10 cm. high, monophyllous, erect, slen¬ der, terete, with a nigro-annulate articulation near the base and with two closely appressed scarious tubular bracts. Leaf about 6.35 cm. long by 2.4 cm. wide, elliptical, coriaceous, obtuse at the tip, cuneate at base, middle nerve prominent beneath. Peti¬ ole about 12 mm. long, rigid, canaliculate. Pedicels much re¬ duced, aggregated, one-flowered, partly concealed by the sheath¬ ing bract that soon breaks down into numerous fibres at the base of the petiole of the leaf. Flower about 18 mm. long, brownish and membranaceous in dried specimens. Lateral sepals about 17 mm. long, connate for about one third of their length, strongly concave at base, each about 2 mm. wide at point of separation, narrowly lanceolate, acuminate, acute, thickened at the apex and somewhat conduplicate. Dorsal sepal 1.8 cm. long, 4 mm. wide near the base, lanceolate, long acuminate, strongly concave, sub¬ acute. Petals about 6 mm. long, .8 mm. wide below the middle, linear-lanceolate, provided — about 2 mm. above the base —
[ L8 1
ORCHIDACEiE
with a minute lobe that is approximately .5 mm. wide and sit¬ uated on the posterior margin. Labellum 7 mm. long, about 4 mm. wide at the broadest part (2.5 mm. above the base), sub- hastate in outline from a cuneate claw, three-nerved, near the base on each side provided with a falcate, obtuse lobe that is about 3 mm. long and shortly decurrent along the lateral nerve, slightly angled on the exterior margin and cucullate at the tip. Between the lateral lobes, on the disc, there is a small oblong callus which is truncate at both ends. Below the middle of the lip, on either side, there is apparent in dried specimens a thick¬ ening of the surface tissues of the disc. Column with broad wings, minutely three-lobed at the apex behind the anther, the middle lobe minutely denticulate. (Plate 114.)
British Guiana, Potaro, ten miles south of Potaro Landing, lat. 5° 10' N., long. 59° W., A. S. Hitchcock 1739 If,, January 7-8, 1920. Distributed by the United States Department of Agriculture, Gray Herbarium of Harvard University and the New York Botanical Garden.
Pleurothallis Schaferi Ames sp. nov. Aff. P. elegantulae et P. rhomboglossae Lindl. Planta inconspicua, caespitosa, vix 1 cm. alta. Radices carnosae, glabrae. Caules secundarii abbrevi¬ ate inconspicui, plus minusve 2 mm. longi, monophylli. Folia cum petiolo 4-6 mm. longa, usque ad 3 mm. lata, coriacea, el- liptico-oblonga, marginata, margine scabridiuscula, apice retuso- apiculata, in petiolum plus minusve 1.5 mm. longum sensim attenuata. Scapus gracilis, usque ad 8 mm. longus, glaber, infra racemum paucibracteatus, vulgo bibracteatus. Racemus pauci- florus, vulgo triflorus, vix 3 mm. longus. Bracteae inflorescentiae subinfundibuliformes, 1 mm. longae, scariosae. Pedicellus cum ovario 1 mm. longus, leviter arcuatus, glaber. Flores minuti, in sicco albidi et semipellucidi, plus minusve 2 mm. longi. Sepala lateralia 1.75 mm. longa, plus minusve 1 mm. lata, anguste lan-
[ ]
ORCHID ACEdS
ceolata, acuta, uninervia, extus leviter carinata, mentum .5-75 mm. longum valde obtusum formantia. Sepalum dorsale 2 mm. longum, valde concavum, oblongo-ellipticum, obtusum, apice ro- tundatum. Petala spathulata, plus minusve 1 mm. longa, valde obtusa, apice rotundato, enervia ut videtur. Labellum 1.25 mm. longum, ecallosum, .5 mm. latum infra apicem, spathulatum, ob- tusum. Columna 1 mm. longa, gracilis, supra medium utrinque alata, ala subquadrata, obscure et irregulariter crenulata.
Pleurothallis Schaferi is similar in habit to P. Grobyi Lindl., but is a much smaller plant with a very different lip and column. In the aspect of the foliage it resembles very closely P. licheni- cola Griseb., from which it is clearly distinguished by its smaller flowers and ecarinate lip.
Cuba, Oriente, Sierra Nipe, near Woodfred, J. A. Schafer 3J/J/.1, Jan¬ uary 5, 1910. Bases of trees in rich woods. 450-550 meters altitude.
Pleurothallis Williamsii A mes sp. nov. Herba epiphytica ut videtur, usque ad apicem inflorescentiae plus minusve 1 dm. alta, radicibus numerosis, albidis. Caules secundarii caespitosi, monophylli, usque ad 7 cm. longi, bracteis infundibuliformibus ornati. Bracteae quinque ad duodecim, marginibus ciliolatis, in sicco diaphanae, apiculatae, nervis prominentibus, brunneis, plus minusve 5 mm. longae, plus minusve 5 mm. inter se distantes. Folium cum petiolo 15-19 mm. longum, usque ad 8 mm. latum trans medium, coriaceum, in sicco conspicue marginatum, apice minute bidentatum, basi cuneatum, nervo medio subtus carinato. Petiolus 2 mm. longus. Pedunculi terminates et laterales, gra- ciles, plus minusve 5 cm. longi, infra racemum bracteis tubulari- bus paucis, vulgo quinqueflori. Bracteae flores subtendentes par- vae, infundibuliformes,quam pedicellus cum ovario multo brevi- ores. Sepala lateralia vix 4 mm. longa, usque ad 1 mm. lata infra medium, triangulari-lanceolata, acuta, conspicue uninervia. Se-
[ 120 ]
0RCHIDACEA5
palum dorsale ovato-lanceolatum, acuminatum, 4 mm. longum, 1-1.5 mm. latum, apice valde incrassatum, abbreviato-caudatum, conspicue trinervium, valde concavum ut videtur. Petala 2 mm. longa, 1 mm. lata supra medium, spathulata, valde obtusa vel ro- tundata, uninervia, in sicco pellucida. Labellum 2 mm. longum, 1 mm. latum, prope basim obscure rotundato-lobatum vel dila- tatum, breviter unguiculatum, prope apicem leviter incrassatum, obtusum, disco bicarinato, carinis usque ad 1 mm. longis, nervo medio incrassato. Columna prope apicem valde bialata, alis quadratis.
Pleurothallis Williavisii is similar to P. Broadwciyi Ames, but is larger both in floral and in vegetative parts, and with an ob¬ long rather than an elliptical labellum. From P. lepanthoides Schltr. it differs in its smaller leaves, fewer flowered racemes, and smaller sepals.
Panama, Cana and vicinity, R. S. Williams 976, April 17-June 8, 1908.
2000-6500 feet altitude.
PLOCOGLOTTIS Bl.
Plocoglottis McGregorii Ames sp. nov. Rhizoma repens, lignosum, validum. Pseudobulbi vel caules monophylli, erecti, sensim in petiolum folii transeuntes, cum folio plus minusve 40 cm. alti, pauciarticulati, statu juvenili vaginis vestiti. Bracteae vel vaginae mox in fibras persistentes solutae. Petiolus gracilis, plus minusve 1 dm. longus, in sicco 2 mm. in crassitudine, pur- pureo-tinctus. Folium plus minusve 3 dm. longum, utrinque at- tenuatum, plus minusve 3.5 cm. latum, anguste elliptico-lanceo- latum, acuminatum, acutum, prominenter trinervium, in sicco viridi-flavum et albo-punctatum. Scapus lateralis, plus minusve 40 cm. altus, purpureo-tinctus, setulosus vel pubescens, infra racemum paucibracteatus ; bracteae infimae vaginantes, elon-
[ 121 ]
ORCHIDACEffS
gatae, acuminatae. Racemus 8-10 cm. longus, plus minusve 3.5 cm. trans medium. Bracteae inflorescentiae plus minusve 9 mm. longae, triangulari-lanceolatae, acuminatae, acutae, pubescen- tes. Flores flavi, purpureo-punctati. Pedicellus cum ovario plus minusve 1.7 cm. longus, setoso-pubescens. Sepala lateralia 1.6 cm. longa, 6 mm. lata trans medium, inaequaliter lanceolata, acuta, extus pubescentia. Sepalum dorsale 1.5 cm. longum, prope basim 6 mm. latum, oblongum, acuminatum, acutum, extus pu- bescens. Petala 1.4 cm. longa, 2 mm. lata, lineari-oblonga, car- nosa, subacuta, glabra. Labellum 9 mm. longum, 8 mm. latum trans apicem, quadrato-cuneatum, apiculatum, apiculo 1.5 mm. longo, triangulo, prope basim utrinque callosum, callis breviter carinatis. Columna crassa, 6 mm. longa, dense pubescens.
Plocoglottis McGregorii is most nearly related to P. Cope- landii Ames among the Philippine species, but is easily distin¬ guished by its larger flowers and differently proportioned leaves. From P. bicallosum Ames it differs markedly in the proportions of the leaf.
Philippines, Panay, Antigue Province, Culasi, R. C. McGregor Bur. Sci. 33189, June 20, 1918. Flowers yellow speckled with wine red. Mossy for¬ est, hills east of Culasi. 1000 meters altitude.
PONTHIEVA R. Br.
Ponthieva parvilabris (LzW/^JS^'cA&./lXen.Orch. 3 (1878) 18; Benth. ex Griseb. in Goetting. Abhandl. 24 (1879) 337. Cra- nichis parvilabris Lindl. Orch. Linden. (1846) 27.
Schlechter in his enumeration of Colombian orchids1 referred Cranichis parvilabris Lindl. to the genus Cranichis, although it would seem that Lindley based his description on a specimen that is referable to Ponthieva. On the same sheet with the type of Cranichis parvilabris there is a stouter plant of similar aspect,
1 In Fedde Repertorium, Beihefte 7 (1920) 215.
[ 122 ]
ORCHIDACEJS
but as it was collected in 1857, it was identified after the Linden plant was described. Beneath the type specimen Lindley made a sketch of the flower which leaves no room for doubt as to the generic position of his species. It exhibits the characteristic la- bell um of Ponthieva and agrees with the very brief description published in Orchidaceae Lindenianae. The generic position of this species was noted by Bentham in his paper published in the Journal of the Linnean Society 18 (1881) 342. Under Cranichis he wrote “ Cranichis, Swartz, has nearly twenty species, to the exclusion of C. parvilabris Lindl., which according to our speci¬ mens, is a species of Ponthieva.'1'1 Grisebach apparently took up this suggestion made by Bentham, but for quite another plant, ignorant of the fact that Reichenbach f. had made the combina¬ tion Ponthieva parvilabris to cover the species described under Cranichis by Lindley. The nomenclatorial history of the species, as I understand it, is given above.
ROBIQUETIA Gaudich.
Robiquetia Merrillii Ames comb. nov. Malleola Merrillii Ames Orch. 5 (1915) 236.
Robiquetia Ramosii^Tftfr? sp. nov. Herba epiphytica, plus minusve 4 dm. alta, robusta, floribus in paniculam laxam dispo- sitis. Caules validi, complanati, vaginis foliorum tecti. Vaginae foliorum plus minusve 5 cm. longae, in sicco nervosae, persisten- tes, rigidae. Folia ligulata, disticha, plus minusve 3 cm. inter se distantia, oblique ascendentia, plus minusve 2 dm. longa, usque ad 3.8 cm. lata, apice inaequaliter bilobata, lobis obtusis. Pedun- culus cum racemo paniculato usque ad 2.8 dm. longus, prope basim paucibracteatus, bracteis vaginantibus. Bracteae inflores- centiae squamiformes, 2 mm. longae, acutae, rigidae. Rami pa- niculae plus minusve 1 dm. longi, laxiflori. Pedicellus cum ova-
[ 123 ]
ORCHIDACEiE
rio plus minusve 12 mm. longus, gracilis. Flores albidi et pur- purei, in racemis pyramidatis dispositi. Rachis incrassata. Sepala lateralia 4 mm. longa, 3 mm. lata prope apicem, obovata, carnosa, basi cuneata, vix patentia. Sepalum dorsale 4.5 mm. longum, plus minusve 2 mm. latum, valde concavum, valde acutum, apic- ulatum. Petala 3.5 mm. longa, supra medium 2.75 mm. lata, late obovata, obtuse apiculata, basi breviter cuneata, textura carnosa. Labellum trilobatum, longe calcaratum; lobi laterales erecti, sub- quadrati, truncati, 1 mm. longi, vix 1.25 mm. lati, carnosi, quam columna multo breviores ; lobus medius vix 1 mm. longus, valde carnosus, tumidus, minute pubescens, infra apicem protuberans; calcar usque ad 9 mm. longum, obtusum, leviter curvatum, plus minusve 1.5 mm. per medium. Columna erecta, 3 mm. longa, carnosa, utrinque prope basim ala rectangula. Rostellum exser- tum, alte bifid um, divisionibus flaccidis, valde acutis. Anthera operculata, antice longe rostrata, rostro 1.5 mm. longo, acuto; caudicula pergracilis, vix 3 mm. longa, superne dilatata, in disco lineari inserta.
Robiquetia Ramosii is closely allied to R. Vcinoverberghii Ames from which it is to be distinguished by the longer spur, promi¬ nent lateral lobes of the labellum and by the elongated rostel¬ lum. In R. Vanoverberghii the lateral lobes of the labellum are suppressed, and the rostellum is short, and simply forked. From R. Merrillii Ames it is separated in part by the spur being curved forward at the tip, not backward, and by the divisions of the rostellum being flexible rather than rigid. Robiquetia com- pressa (Lindl.) Schltr. differs from R. Ramosii in its broad obo- vate petals with a longer cuneate base and in the simple, not paniculate, inflorescence.
Philippines, Catanduanes, M. Ramos Bur. ScL 30396, December 2,1917.
On trees along the Santo Domingo River. 30 meters altitude. (Type);
[ 124 ]
ORCHIDACEiE
Panay, Antigue Province, R. C. McGregor Bur. Sci. 88598, July 18, 1918. In mossy forest. Sepals and petals spotted with red. Labellum lobes yel¬ low, spur white. About 900 feet altitude.
SARCOCHILUS R. Br.
Sarcochilus tripercus Ames sp. non. Herba epiphytica. Radi¬ ces elongatae, glabrae, albidae. Caules abbreviati, plus minusve 1 cm. longi, vaginis foliorum obtecti. Folia disticha, articulata, 4-11 cm. longa, usque ad 9 mm. lata, lingulata, coriacea, apice inaequaliter bilobata, subacuta, oblique erecta. Scapus plus mi¬ nusve 7 cm. longus, setosus, gracilis, quam folia brevior. Rachis leviter incrassata, plus minusve 5 mm. longa. Bracteae inflores- centiae rigidae, obtusae, squamiformes, 1 mm. longae, quam pedi- cellus cum ovario multo breviores. Pedicellus cum ovario 5 mm. longus, glaber. Flores succedanei, subflavi. Sepala lateralia 5 mm. longa, plus minusve 3 mm. lata trans medium, inaequaliter ovata, obtusa, apice incrassato minute appendiculata, quinque- nervia. Sepalum dorsale 6 mm. longum, 2 mm. latum, valde con- cavum. Petala plus minusve 5 mm. longa, 1.5 mm. lata prope apicem, oblongo-spathulata, apice rotundata. Labellum compli- catum, unguiculatum, cum sacco 1 cm. longo, trilobatum ; lobi laterales asciiformes, ad medium valde constricti, apice 1 mm. lato ; lobus medius trilobulatus, lobulis lateralibus lineari-oblon- gis, 2 mm. longis, .5 mm. latis,patentibus, membranaceis, lobulo medio triangulo, acuto, 1 mm. longo, concavo; unguis gracilis, .75 mm. longus. Saccus elongatus, cylindratus, 7 mm. longus, apice curvatus, obtusus, prope apertionem inflatus, ad medium attenu¬ ate. Columna curvata, in pedem 2 mm. longum producta, infra apicem gracilis, rostello producto, triangulo, acuto, deflexo. An- thera ovata, antice subrostrata, acuta. Pollinia globosa, glandula lineari-triangula.
In habit not unlike Sarcochilus mindanaensis Ames, but with
[ 1^5 ]
ORCHIDACEAC
very dissimilar flowers. The labellum suggests S. appencLiculatus J. J. Sm., but the elongated sac is a differentiating character. The labellum is a very interesting structure and might well be described as five-lobed ; the basal lateral lobes being erect, more or less membranaceous and suddenly dilated above the middle ; the apical lateral lobes being membranaceous and linear-oblong ; the terminal or fifth lobe being more or less rigid and apparently thickened at the tip, when fresh.
Philippines, Leyte, Tacloban, C. A. Wenzel 0968, February 11, 1916.
Epiphyte in forest. Flower yellow and white. At sea level.
SPIRANTHES, L. C. Rich.
Spiranthes bicaudata Ames sp. nov. Radices fasciculatae, fusiformes, carnosulae, plus minusve 4 cm. longae, 7 mm. cras- sae, foliis radicalibus ut videtur, paucis, laminis foliorum usque ad 7 cm. longis, 17-27 mm. latis, membranaceis, in sicco charta- ceis, flavidis, ellipticis, utrinque leviter attenuatis, acutis, quin- quenerviis, petiolis usque ad 3 cm. longis. Scapus erectus, gra¬ cilis, cum racemo usque ad 22 cm. longus, infra racemum brac- teis setaceis octo, illis prope basim scapi vaginantibus. Racemus elongatus, multiflorus, plus minusve 11 cm. longus, plus minusve 1.5 cm. crassus per medium. Bracteae inflorescentiae lineari-lan- ceolatae, plus minusve 7 mm. longae, scariosae, setaceae, quam pedicellus cum ovario longiores. Pedicellus cum ovario glandu- loso-pubescens, usque ad 8 mm. longus. Flores albidi,minuti,3-4 mm. longi. Sepala lateralia ligulata, 4 mm. longa, acuta, apice leviter incrassata, uninervia, glabra. Sepalum dorsale simile, 4 mm. longum, vix 1 mm. latum trans medium, uninervium, obtu- sum, prope apicem leviter carinatum vel incrassatum. Petala quam sepalum dorsale breviora, 2.5 mm. longa, leviter sigmoidea, uninervia. Labellum 3 mm. longum, 1 mm. latum trans medium,
[ 126 ]
ORCHIDACEiE
late unguiculatum, usque ad medium dilatatum nunc constric- tum demum prope apicem leviter dilatatum, trinervium, ad basim utrinque in caudam elongatam, gracilem, plus minusve 1 mm. longam productum. Columna mediocris, stigma simplex, vix bi- lobatum. Rostellum haud productum, breviter emarginatum. Anthera late ovata, 1 mm. longa.
Allied with S. Lucayana Cogn. The small flowers and bicau- date labellum are differentiating characters.
This species is apparently a member of the group recently described under the genus Mesadenus by Dr. Rudolf Schlechter in his monograph of the Spiranthinae.1 I have not taken up this name for the Trinidad species, because I am convinced that Schlechter’s treatment of the Spiranthinae needs to be carefully tested from beginning to end before it is adopted as a basis for progress. Schlechter’s monograph is very subtle in conception, and is based on characters which are too recondite for practical purposes. It is certainly perplexing and one may fairly question the advisability of accepting its innovations.
Dr. Schlechter has, it is true, revealed some clear lines of de¬ marcation in a group that is large and heterogeneous, yet sev¬ eral of his new genera seem hardly isolable from Spiranthes and rest on characters that are difficult to define. Whether or not the lines of demarcation are sufficiently clear for generic separation, it is patent, as one uses Schlechter’s system, that difficulties have been shifted rather than removed.
The conception of genera admittedly rests on personal judg¬ ments. The history of biological science indicates quite clearly that personal judgments in the erection of genera receive harsh treatment when they fail to elucidate relationships, when they give place to burdensome perplexity and when they emphasize
l Beihefte Bot. Centralbl. 37, Abt. 2 (1920).
[ 127 ]
ORCHIDACEiE
the magnitude of a difference rather than its constancy. Some thirty-two thousand personal judgments have characterized the history of the genus in systematic botany, and of these about eleven thousand have withstood conscientious scrutiny and the test of time. In other words, the generic synonymy that we en¬ counter in systematic work among plants records a surprising disregard of genera that have been too loosely defined or too hastily proposed or that rest on decisions which ignore the prac¬ tical application of taxonomy.
In genera comparable to Spiranthes — genera that have been retained in their broadest sense, although made up of very unlike elements — the limits imposed by an historical and traditional past seem almost sacrosanct. Of course, such groups should not be maintained forever as heterogeneous assemblages because they have been so maintained in the past. Such a procedure would effectually put an end to progress. But when dismemberment of such genera as Spiranthes results in difficulties of interpretation and gives rise to a situation in which a working botanist is shifted from one exasperating perplexity to another, his final state of mind being uncertainty as to the accuracy of his diagnosis, then dismemberment is inadvisable. If one of these large genera is to be dismembered, it is fair to expect that proposed changes will be published only after a reasonably long period of probation in the hands of their author.
It is not my intention to review Dr. Schlechter’s monograph of the Spiranthinae in these pages, but having criticized it, a single example to substantiate my position may suffice.
Spiranthes novaezelandiae Hook, f., a species in facies similar to S. sinensis (Pers.) Ames, is characterized in part by the appar¬ ent suppression of the rostellum. It would, I think, be described
as erostellate by the majority of botanists. The pollinia appear
[ 128 ]
ORCHIDACEdS
to rise above the stigma as a naked mass. In flowers that are ex¬ amined at the time of anthesis it is possible to detect a very thin membrane which is continuous with the body of the gynoste- mium. To this membrane the pollinia are in part agglutinated. That this membrane is deeply bifid, as is the case in related spe¬ cies, is by no means clear. In fact the use of this membrane in generic classification would, I am sure, be regarded as reprehen¬ sible by cautious investigators. Unless one were pretty well ac¬ quainted with the more delicate characters of the flower in Spi- ranthes, it would be natural to suppose that development of the rostellum in Spiranthes novaezelandiae had been suppressed. However interpreted, the gynostemium does not bring the spe¬ cies within the proper genus if Schlechter’s monograph of the Spiranthinae is depended on for identification. Indeed, the con¬ dition of the rostellum removes S. novaezelandiae from the group in which Schlechter placed it. Gattungsreihe I of Schlechter’s sys¬ tem, in which S. novaezelandiae is included under Spiranthes, is defined as follows : “ Gattungen, bei denen die Klebscheibe der Pollinien zwischen den Fortsatzen des verlangerten, zweispalti- gen Rostellum festgehalten, resp. eingeklemmt ist.” Under Spi¬ ranthes, the first genus in the group, the column is described as follows: “Columna brevis, pede brevi, apice incurvulo; rostello alte bifido, cruribus erectis, subulatis.” And yet in Schlechter’s notes under Spiranthes novaezelandiae he tells us that the column in the flowers he has examined had not developed a rostellum.
By the preceding remarks it is not my purpose to demonstrate that Schlechter’s system is open to general condemnation. I have found it a very helpful effort toward clarity in a highly technical group. One would regard it more favorably, however, if the new genera proposed and some of the genera reinstated had been kept
as sections or as tentative groups, on which to build up a more
[ 129 J
ORCHID ACE^E
profound knowledge than we now possess of relationships in the complex made up of Spiranthes, Cyclopogon and similar con¬ cepts. The multiplication of genera attempted by Schlechter has tended to conceal the larger affinity indicated by the more com¬ prehensive genus, and the difficulties of identification have been unduly intensified by reliance on characters which are hard to define. As modified by Dr. Schlechter the Spiranthinae has lost much even if it has gained a little.
In a survey of Schlechter’s proposals I find myself leaning strongly toward the point of view expressed by Dr. B. L. Rob¬ inson in his paper on “The Generic Concept in the Classifica¬ tion of the Flowering Plants,” a point of view clearly expressed as follows: “Different minds may work in unlike manner when confronted by the difficulties of identifying plants. Personally, I should very much prefer to have the difficulty at one point rather than at two ; that is to say, I should rather have generic lines drawn so widely that genera would be pretty definite and readily recognized, in the manner, let us say, of Cyperus, Astra¬ galus, or Euphorbia in the broader and long traditional sense. The recognition of such genera requires little or no mental ef¬ fort on the part of a botanist of any training. The attention is left free for the specific identification, and this may be undertaken with a happy confidence that all the species likely to come into question will be found in the same group and under the same generic name. These species may be inconveniently numerous, but at least one is not disturbed by any lurking doubt whether, after all, he has got the right genus.” (Proc. Am. Assoc. Adv. Sci. 55 (1906) 424.)
Trinidad, B. W. I., heights of Aripo, N. L. Britton § W. G. Free¬ man 2827, March 16, 1921. Flowers white, minute.
[ 130 ]
ORCHIDACEdS
Spiranthes Wrightii Ames nom. nov. Sauroglossum mono- phyllum Griseb. Cat. PL Cub. (1866) 269 as to Wright 1480, excl. synonymy and Wright 621. Spiranthes monophylla Cogn. in Urb. Symb. Antill. 6 (1909) 339 as to description, not as to synonymy, excl. Wright 621.
Spiranthes W rightii differs from its nearest allies in Cuba in being monophyllous. Apparently there is a wide range of varia¬ tion in the height of the plants and in the number of flowers borne by the racemes. The sheet in the Gray Herbarium bears two specimens that represent what may be taken as extremes, the tallest plant being 40 cm. high with nine flowers, the smaller one being 15 cm. high with only two flowers.
From the discussion under Cranichis it is evident that the plant referred by Cogniaux to S. monophylla under the number 1480 of Wright’s Cuban plants is without a valid name. As a new name is necessary, S. Wrightii is proposed.
Cuba, Pinal, Monte Verde, Wright 1^80, February 19, 1859. In dense
woods. Whole plant tinged with dull red. Petals and sepals one-nerved.
Lip five-nerved, nerves reddish green.
STELIS Sw.
Stelis parvibracteata Ames sp. nov. Aff. S. obscuratae Reich b. f. Radices fibratae, patentes. Caules oblique ascendentes, plus minusve 3 cm. longi, uniarticulati, vaginis duabus alte et arete amplectentibus obtecti, monophylli. Folium cum petiolo 5-11 cm. longum, usque ad 1.5 cm. latum, ligulatum, coriaceum, anguste oblanceolatum, obtusum, marginatum, in petiolum 2 cm. longum sensim productum. Pedunculus singulus, folio multo longior, cum racemo 10-18 cm. longus, infra racemum pauci- bracteatus. Racemus plus minusve 10 cm. longus, laxe multi- florus. Bracteae inflorescentiae minutae, vix 1 mm. longae, late ovatae, valde acutae, quam pedicellus cum ovario multo breviores.
[ 131 ]
ORCHIDACEdE
Pedicellus cum ovario plus minusve 2 mm. longus. Flores 5 mm. trans medium, purpurei. Sepala lateralia 2 mm. longa, vix 3 mm. lata, late ovata, vix obtusa, trinervia. Sepalum dorsale 2.5 mm. longum, 3 mm. latum, late rotundato-ovatum, trinervium. Pe- tala cuneata, 1 mm. longa, plus minusve 1 mm. lata prope apicem, trinervia, apice valde incrassata, glabra vel exiliter papillosa. Labellum 1 mm. longum, plus minusve 1 mm. latum, elliptico- reniforme, apiculatum, apiculo triangulo, acuto, erecto; discus callo permagno ornatus. Columna brevis, apicem versus con- spicue dilatata.
In habit similar to S. coiloglossa Schltr., but different in hav¬ ing shorter stems and a reniform-apiculate, not ovate obtuse lip.
Panama, Cana and vicinity, R. S. Williams 97%, April 27, 1908. On
trees, flowers purplish. 6000 feet altitude.
Stelis pleurothalloides Ames sp. nov. Herba epiphytica, usque ad apicem racemi plus minusve 22 cm. alta. Radices car- noso-fibratae, albidae, glabrae, in sicco nitidae, plus minusve 6 cm. longae, usque ad 1 mm. in crassitudine. Caules plus minusve 8 cm. longi, vaginis elongatis, laxe appressis omnino obtecti, mo- nophylli. Folium valde coriaceum, oblongo-ellipticum, 6-6.5 cm. longum, in sicco usque ad 1.5 cm. latum, apice breviter apicula¬ tum, nervo medio subtus prominenti, conspicue marginatum, in petiolum brevem contractum. Pedunculus singulus, cum racemo plus minusve 16 cm. longus, basi vagina 1.8 cm. longa inclusus. Racemus usque ad 12 cm. longus, multiflorus, 1 cm. in diametro. Flores plus minusve 4 mm. inter se distantes, textura carnosi, in sicco 4 mm. longi, straminei. Bracteae inflorescentiae plus minusve 3 mm. longae, scariosae, oblique infundibuliformes, rachidem amplectentes, quam pedicellus cum ovario breviores, acutae. Pedicellus cum ovario plus minusve 5 mm. longus, valde
ORCHIDACE^E
arcuatus. Sepala lateralia 4.75 mm. longa, infra medium 2 mm. lata, prope basim cohaerentia, fere libera, anguste triangula vel triangulari-lanceolata, obtusa, extus leviter carinata, trinervia. Sepalum dorsale 5 mm. longum, 2.25 mm. latum prope basim, ovato-lanceolatum, obtusum, trinervium. Petala 1.5 mm. longa, apice 1 mm. lata, valde incrassata, cuneata, truncata, extus levi¬ ter umbonata prope apicem, in sectione transversa triangula- ria, .5 mm. in crassitudine. Labellum 1 mm. longum, basi .75 mm. latum, supra medium dilatatum 1 mm. latum, orbiculari-apicu- latum, a basi breviter oblonga usque ad medium labelli callo bi- lobato, antice excavato extendit. Columna brevis, apicem versus dilatata, petala aequans, apice bilobulata.
In habit this species suggests S. micrantha Sw. The callus on the labellum resembles that of S. patula Schltr. The aspect of the flowers recalls some species of Pleurothallis.
Colombia, State of Cauca, Pdramo de Buena Vista, Huila group, Central Cordillera, H. Pittier 1156, January 1906. Upper zone of forest. 3000— 3600 meters altitude. (Type in U. S. Nat. Herb. No. 531353.)
Stelis Williamsii Ames sp. nov. Aff. S. confusae Schltr. Radices fibratae, copiosae, patentes. Caules caespitosi, abbrevi¬ ate vix 1.5 cm. longi, monophylli, oblique ascendentes, in sicco longitudinaliter sulcati. Folia coriacea, in petiolum abbreviatum sensim decrescentia, cum petiolo 4-10 cm. longa, usque ad 11 mm. lata, ligulata, marginata, apice retuso-apiculata, subtus nervo medio leviter carinato. Scapus singulus, folio longior, us¬ que ad 2 dm. longus, infra racemum paucibracteatus. Bracteae acutae, plus minusve 6 mm. longae, vaginantes. Racemus multi- florus, usque ad 9 cm. longus. Rachis gracilis, in sicco fragilis, floribus 3-5 mm. inter se distantibus. Bracteae inflorescentiae triangulari-ovatae, plus minusve 3 mm. longae, acutae, extus longitudinaliter carinatae. Pedicellus cum ovario 2 mm. longus.
[ 133 ]
ORCHIDACEiE
Flores purpurei. Sepala lateralia 3.5 mm. longa, 3.5 mm. lata, usque ad medium cohaerentia, late ovata, glabra, subacuta, tri- nervia. Sepalum dorsale simile, trinervium. Petala 1 mm. longa, plus minusve 1 mm. lata, cuneata, truncata,prope apicem minute verruculosa vel papillosa, margine apicali scabridula. Labellum plus minusve 1 mm. longum, prope basim 1 mm. latum, trulli- forme, obtusum, apice rotundatum, ecallosum. Columna minuta, utrinque in lobum spathulatum vel subrotundatum producta.
Stelis Williamsii is a near ally of S. confusa Schltr., a Mexican species which is dissimilar in the column-lobes, in the structure of the labellum and in the differently shaped petals. The racemes suggest S. ciliaris Lindl., but the sepals are eciliate.
Panama, Cana and vicinity, R. S. Williams 970, April 27, 1908. On trees, flowers purplish. 2000-6500 feet altitude.
THRIXSPERMUM Lour.
Thrixspermum subulatum (Bl.) Reichb. f Xen. Orch. 2 (1867) 122; J. J. Sm. in FI. Buitenz. 6 (1905) 578, f. 434. Den- drocolla subulata Bl. Bijdr. (1825) 291. Aerides subulatum Lindl. Gen. & Sp. Orch. (1833) 241. Sarcochilus subulatus Reichb. f. in Walp. Ann. 6 (1861) 500.
This species has not been recorded from the Philippines here¬ tofore, although it is known to be a native of Java, Sumatra and Ambon.
Philippines, Luzon, Bontoc Subprovince, Mount Masapilid, M. Ramos <$f G. Edafio Bur. Sci. 37905, March 17, 1920. Epiphyte. Flowers yellow. 3000 feet altitude.
Thrixspermum Weberi Ames sp. nov. Caules radicantes, plus minusve 15 dm. longi, usque ad 5 mm. in crassitudine, va- ginis foliorum tecti. Radices ramosae, fibratae. Vaginae foliorum usque ad 12 mm. longae, valde complanatae, coriaceae, rigidae,
[ 134 ]
ORCHIDACEdE
apice oblique truncatae, in sicco rugosae. Folia plus minusve 4.5 cm. longa, circa 11 mm. lata, 6 mm. inter se distantia, valde cori- acea, ligulata, apice inaequaliter bilobata, in petiolum brevem contracta, disticha, oblique ascendentia, in sicco rugosa. Pedun- culi elongati, usque ad 20 cm. longi, rigidi, vix 1 mm. per me¬ dium, paucibracteati, bracteis tubulatis, circa 3 mm. longis. Ra- cemus plus minusve 2 cm. longus. Bracteae inflorescentiae usque ad 6 mm. longae, lineari-triangulae, acutissimae, rigidae, quaqua- versae, quam pedicellus cum ovario breviores. Flores succedanei, flavescentes,perpauci aperti eodem tempore, membranacei. Pedi¬ cellus cum ovario vix 1 cm. longus, gracilis, glaber. Sepala later- alia plus minusve 7 mm. longa, supra basim 4.5 mm. lata, inae¬ qualiter triangula vel late falcata, acuminata, acuta, apice con- duplicata. Sepalum dorsale 7 mm. longum, lanceolatum, valde contractum, acutum. Petala 6 mm. longa, circa 1 mm. lata, ob- longo-lanceolata, acuta, valde membranacea. Labellum e basi columnae usque ad apicem lobi terminalis circa 5 mm. longum, trilobatum, membranaceum, in saccum 3 mm. longum produc- tum, unguiculatum, ungue 2 mm. longo; lobi laterales antice 2 mm. longi, late falcati, apice curvati, triangulari-acuti, membra¬ nacei, quam lobus medius multo longiores; lobus medius abbre¬ viate, vix 1 mm. longus, triangulus, porrectus, leviter concavus ; discus prope basim lobi medii callosus, callo bilobato, transverse complanato, prope apertionem sacci callo triangulo, carnoso in- structus. Gynostemium ad apicem antherae 2.5 mm. longum, carnosum. Capsula cylindrata, usque ad 6 cm. longa.
Among Philippine species T. Weberi is closely related to T. agusanense Ames and T. Wenzelii Ames from which it differs in the characters of the inflorescence. When spread out the lat¬ eral lobes of the labellum project far beyond the middle lobe
and converge. The lateral sepals are characterized by an excessive
[ 135 ]
ORCHIDACEiE
protuberance of the basal angle. This gives a bilobed appearance to each sepal.
Philippines, Mindanao, Province of Agusan, Cabadbaran, C. M. Weber 187 , August 8, 1911. Epiphyte on tree trunks. Fruit green-purple. At sea level.
TRICHOGLOTTIS Bl.
Trichoglottis brachiata Ames sp. nov. Aff. T. philippinensi Lindl. Caules plus minusve 30 cm. alti, rigidi, usque ad apicem foliati, vaginis foliorum obtecti, usque ad 8 mm. in crassitudine, internodiis 1.4 cm. longis. Folia plus minusve 4.5 cm. longa, 2.9 cm. lata, oblongo-elliptica, disticha, retusa, apiculata, apiculo rigido, acuto. Flores laterales, e nodis orientes, singuli. Pedicel- lus cum ovario usque ad 3.5 cm. longus. Sepala lateralia patentia, 21 mm. longa, 11 mm. lata, ovato-lanceolata, acuta, coriacea. Sepalum dorsale simile. Petala 2 cm. longa, 6 mm. lata, anguste lanceolata, utrinque attenuata, subacuta, valde coriacea, paten¬ tia. Labellum 19 mm. longum, quinquelobatum, usque ad me¬ dium columnae affixum, ecalcaratum, basi saccatum vel valde concavum ; lobi basilares erecti,trianguli, carnosi, circa 3 mm. alti, intus minute pubescentes; lobi laterales 8 mm. longi, usque ad 1.5 mm. lati, acinaciformes, apice inaequaliter dentati, utrinque minute pubescentes; lobus terminalis a latere complanatus, 9 mm. longus, 5 mm. altus, carnosus, cuneato-quadratus, angulo superiore valde elongato, utrinque pubescens, supra pilosus ; dis¬ cus carina valde carnosa ornatus, prope basim appendicula quad- rata, truncata, carnosa instructus. Columna brevis, 7 mm. longa, minute pubescens.
Trichoglottis brachiata is closely allied to T. phUippinensis Lindl. from which it is most readily distinguished by the elon¬ gated, linear, curved lateral lobes of the labellum which arise near the base of the terminal lobe, and by the nearly smooth
[ 136 ]
ORCHID ACEdE
appendage on the disc in front of the concave base. In habit these species are very similar and in the absence of flowers might easily be confused. In Lindley’s specimens of T philippinensis collected by Cuming the terminal pair of lobes is abbreviated and triangular in outline.
Philippines, Biliran Island, R. C. McGregor Bur. Sci. 18931, June 13, 1914. In mangrove swamp. Sepals and two upper petals light cadmium on the back, inside victoria lake except narrow edge and a central line on each which are light cadmium. Odd petal white with a few lines of true purple near base, large patch of cadmium yellow in centre with some vic¬ toria lake above — Ridgeway’s Color Standard.
Trichoglottis philippinensis Lindl. in Ann. & Mag. Nat. Hist. 15 (1845) 386. Stauropsis philippinensis Reichb. f. in Hamb. Gartenzeit. 16 (1860) 117.
In my previous lists of Philippine orchids the generic name Stauropsis was adopted for this species. Stauropsis is at present regarded as a monotypic genus, the only species S. undulata Benth., being a native of the Himalayan region.
[ 137 ]
.
PLATE 114
ORCHIDACE^E
Plate 114
Oberonia linearifolia. Plant, natural size, drawn from the type. 1, flower enlarged, lip removed, showing denticulate petals, and the sepals. 2, flower enlarged, with lip somewhat spread out to show the lobes.
Bulbophyllum nigroscapum. Plant, natural size, drawn from the type. 3, flower enlarged, show¬ ing sepals, petals, lip and column. 4, lip en¬ larged.
[ 140 ]
OBRRONIA &n eari/otfia JL
I
'■ * v * • • “*< 1
■
ADDITIONS TO THE ORCHID FLORA OF THE MOUNTAIN PROVINCE, LUZON
In late September