AR AO T D CAR e m Kä ra S ee Ee EEEN cios es) "m 1 3 H jou 7 o de Ween? od PET DE B. cd 7 = Hp A RR edu Sy SR MISSOURI BOTANICAL GARDEN Annals of the Missouri Botanical Garden “EN SA INES rra EAT m 4, Volume XXIII 1936 With Thirty-four Plates and Twenty-five Figures Published quarterly at Fulton, Missouri, by the Board of Trustees of the Missouri Botanical Garden, St. Louis, Mo Entered as second-class matter at the post-office at Fulton, Missouri, under the Act of March 3, 1879. 54399 Annals of the Missouri Botanical Garden A Quarterly Journal containing Scientifie Contributions from the Missouri Botanical Garden and the Graduate Laboratory of the Henry Shaw School of Botany of a University in affiliation with the Missouri Botanical Garden olli The ANNALS OF THE MISSOURI BOTANICAL GARDEN appears four times during the ealendar year: February, resch ehe. and November. Four numbers constitute a volume. open ra NN ORE ad d $6.00 = volume Single Numbers.......... 1.50 each* Contents of previous issues of the ANNALS OF THE MISSOURI BOTANICAL GARDEN are listed in the Agricultural Index, published by the H. W. Wilson Company. * Except No. 3 of Vol. 22, containing 384 pages, price $5.00. STAFF OF THE MISSOURI BOTANICAL GARDEN trector, GEORGE T. MOORE. Assistant to the Director, KATHERINE H. LEIGH. HERMANN VON SCHRENK, EDGAR ANDERSON, Pathologist. Geneticist. JESSE M. GREENMAN ROBERT E. Woopson, JR., Curator of the Steeg Research Assistant. ERNEST S. REYNOLDS, Davip C. FAIRBURN, Physiologist. Research Assistant. CARROLL W. DODGE, NELL C. HORNE Mycologist. Librarian E. Editor of Publications. BOARD OF TRUSTEES OF THE MISSOURI BOTANICAL GARDEN President, *. GEORGE C. HITCHCOCK. Vice-President, DANIEL K. CATLIN. L. Ray CARTER. GEORGE T. MOORE. SAMUEL C. DAVIS EUGENE PETTUS. Tuomas S. MAFFITT. ETHAN A. H. SHEPLEY. EX-OFFICIO MEMBERS: GEORGE R. THRO paar rel ik DIc cr e CTT Mayor of the ay " St. Louis. ROBERT J. TERRY, WILLIAM SCARLE TM of "The Academy of Sci- Bishop of omg D n of Missouri. f St. Louis. Francis C. SULLIVAN, President of the Board of Education of St. Louis. GERALD E. ULRICI, Secretary. TABLE OF CONTENTS A Comparative Study of Conidial Formation in Cephalosporium and Some Related Hyphomycetes.......... M. Elizabeth Pinkerton A Study of the Genus Helicogloea. . Gladys E. Baker Morphology, Physiology and Cytology of Syringospora inexorabilis (Monilia inez- orabilis)....Carroll W. Dodge and Morris Moore Transitional Pitting in Tracheids of FILUM 21 oan rss is Sem 8 John Adam Moore and Henry N. Andrews, Jr. An Experimental Study of Hybridization in the Genus Apocynum.......... Edgar Anderson Studies in the Apocynaceae. IV. The Amer- ican Genera of Echitoideae (Concluded) E ete desk Ot E Robert E. Woodson, Jr. A New Sequoioxylon from Florissant, Colo- jv. c e Ee Field and Herbarium Studies, IV.......... LONE MI IUS: y 2d E en, Louis O. Williams The Species Problem in Iris...... Edgar Anderson Hybridization in American Tradescantias ee res). A Edgar Anderson The Effect of Environment upon the Pro- duction of Sporangia and Sporangiola in Blakeslea trispora Thaxter...... David Goldring New Organisms of Chromomycosis........ ....Morris Moore and Floriano Paulo de Almeida Henry N. Andrews, Jr. PAGE 1- 68 69-128 129-150 151-158 159-168 169-438 439-446 447-456 457-509 911-525 921-542 043-052 DLE eee he EE E George Zebrowski 553-564 Hydnangium and Related Genera......... lycos Carroll W. Dodge and Sanford M. Zeller 565-598 Elasmomyces, Arcangeliella, and Macowan- med... Sanford M. Zeller and Carroll W. Dodge 599-638 NS «uiri chao er eCXRERO S09 2n NND Sanford M. Zeller and Carroll W. Dodge 639-655 General Index to Volume XXIII.................. 657-662 Annals of the Missouri Botanical Garden Vol. 23 FEBRUARY, 1936 No. 1 A COMPARATIVE STUDY OF CONIDIAL FORMATION IN CEPHALOSPORIUM AND SOME RELATED HYPHOMYCETES! M. ELIZABETH PINKERTON Formerly Graduate Scholar in the Henry Shaw School of Botany of Washington Uniersity I. INTRODUCTION Among the organisms isolated by the author from dermato- mycoses at the Barnard Free Skin and Cancer Hospital, St. Louis, Missouri, a red yeast, identified as Torulopsis mucilagi- nosa, appeared frequently. On ordinary media it formed a rose-red, pasty, and superficial colony; in liquid, a basal de- posit and some pseudomycelial proliferation throughout the solution. Upon nitrate agar of Stellung-Dekker (’31) the growth was homogenous and scanty on the surface and cloudy- white within the medium. Microscopic examination revealed a pseudomycelium bearing numerous lateral spore balls remi- niscent of Cephalosporium (pl. 2, fig. 43). This suggested a further study of the cause of the deviation and the relationship of the budding and cephalosporial states. A cursory investi- gation of the pertinent literature disclosed much confusion regarding the types of Hyphomycetes which form cephalo- sporia (spore balls). Hence, the problem has been extended to a comparative morphological study involving principally 1 An investigation carried out in the Graduate Laboratory of the Henry Shaw School of Botany of Washington University, and submitted as a thesis in partial fulfillment of the requirements for the degree of doctor of philosophy in the Henry Shaw School of Botany of Washington University. Issued March 20, 1936. ANN. Mo. Bor. GARD., VOL. 23, 1936. (1) [Vor. 23 2 ANNALS OF THE MISSOURI BOTANICAL GARDEN forms which may grow parasitically or saprophytically upon the human skin. It is the primary aim of this dissertation to explain the mechanism of cephalosporial produetion and to correlate such findings with data on other types of spore-producing organs. Observations of the pathogenicity of these organisms for man are recorded. Also, on the basis of methods of production and kinds of spores, as well as upon other physiological and mor- phological observations, the group is discussed systematically. Present treatments are criticized and emended, at least in con- sideration of the few forms studied. II. MarERIALS AND METHODS Investigations have been made upon the following species. For further discussion of systematies see p. 37. ORGANISM PLACE OBTAINED Allescheria Boydii Shear Es voor Schimmel- eultures, Baa Cephalosporium kiliense (Gruetz) Hartmann sadi voor Schimmel- eultures, Baa Cephalosporium niveolanosum Benedek "macies ja voor Sehimmel- eultures, Baarn Cephalosporium rubrobrunneum Nannizzi Centraalbureau voor Schimmel- eultures, Baar Cephalosporium Serrae Maffei Centraalbureau voor Schimmel- eultures, Baarn Cephalosporium Stuehmeri Schmidt & Beyma Centraalbureau voor Schimmel- cultures, Baarn Cephalosporium symbioticum Pinkerton Barnard Free Skin and Cancer Hospital, St. Louis Clonostachys Araucaria Corda Centraalbureau voor Schimmel- cultures, Baarn Clonostachys Araucaria var. rosea Preuss Centraalbureau voor Schimmel- eultures, rn Clonostachys Araucaria var. confusa Pinkerton C. Thom Clonostachys spectabilis (Harz) Oudem. & Sace. Centraalbureau voor Schimmel- eultures, Baarn Gliocladiwm deliquescens Sopp C. Thom Gliocladium fimbriatum Gilman & Abbott C. Thom Gliocladium penicilloides Corda C. Thom Gliocladium Vermoeseni (Biourge) Thom C. Thom Scopulariopsis brevicaulis (Saec.) Bainier C. Thom Torulopsis mucilaginosa (Jórgensen) Ciferri & Barnard Free Skin and Cancer Redaelli Hospital, St. Louis 1936] PIN KERTON—CONIDIAL FORMATION 3 The stock medium was glycerine agar (nutrient agar plus 6 per cent glycerine, pH 6.9). The following agars were also used: maltose (pH 5.4), Sabouraud’s dextrose (pH 5.6), wort (pH 4.8), Raulin’s (pH 4.15), Czapek’s (pH 4.43), Richards’ (pH 4.36), Endo’s (pH 7.52), Gorodkova-Maneval modified (pH 6.8), nitrate (pH 5.5), nutrient (pH 6.0), corn meal (pH 6.0), potato dextrose (pH 5.7); these liquids—nutrient (pH 6.0), lactose (pH 6.8), litmus milk; also sterilized carrot plugs, gypsum blocks, Bacto-beef blood serum (pH 7.3), and glycerine agar double strength plus 50 per cent Bacto-blood (pH 6.0). The latter gave the most luxuriant and rapid growth. Cultures were made chiefly in test-tubes and bits of them were examined at intervals in semi-permanent mounts of Amann’s (’96) lacto-phenol plus a 1 per cent aqueous solution of anilin blue. Such served for gross morphology and even showed a fair degree of the more minute structure of the phialide. The proper relationships of the various parts of the organism and their comparative colonial characteristics are shown in the photographs of Petri plate cultures containing Czapek’s agar (pl. 6). Cultures were also examined directly under high power after placing a cover slip over a sparsely grown section of the agar. Camera-lucida sketches were made of relevant structures and photomicrographs were taken where possible, but due to the minuteness of the conidial fructifica- tions the latter procedure was not very practical. Hanging drop and agar preparations made in van Tieghem cells were of little use, since the fruiting bodies were produced pendant in the air and hence not discernible under oil. Agar slide cultures, produced by placing a few drops of warm agar upon a sterile slide, inoculating, and applying an aseptic cover slip (No. 1, 24 x 40) when the agar was cool enough to barely spread, were found to be most helpful, giving a comparatively small circular agar matrix from which aerial fructifications radiated. The narrow ends of the cover slip were sealed with paraffin to make the mount rigid, and the microcultures incu- bated at room temperatures in sterile staining jars with moist filter-paper at the bottom. On the second day the formation of spores could sometimes be observed under oil immersion. [Vor. 23 4 ANNALS OF THE MISSOURI BOTANICAL GARDEN Material for eytologieal study was grown on Raulin's agar and on glycerine agar. Hermann's fluid! was employed for killing and Jeffrey's celloidin technique was followed for the earlier work. The fixing agent was applied directly to the slant eulture which was evacuated for about fifteen minutes to re- move air from the agar containing submerged growth. After twelve hours the material was washed for an equal time in slowly running water, following which the agar was eut into small blocks about 5 mm. square and run through the alcohols. Since it was impossible to get celloidin sections thinner than 10», for studying structures of 2 y average diameter the paraf- fin method was tried. After the agar culture had been cut into small bits, good infiltration was secured and sections were ob- tained 1, 3, 5, 8, and 10 y in thickness. The very thin ribbons had the disadvantage that so few aerial phialides were en- countered in a slice. Nevertheless, nuclear orientation was clearer in the paraffin material, whereas accessory cytoplasmic inclusions and external mucoid strueture were more noticeable in the celloidin matrix. ` Haidenhain's iron-alum haematoxylin was used exclusively for staining, applied according to the fol- lowing schedule: hydration, one hour in 20 per cent USP hy- drogen peroxide, washing, 4 per cent iron alum two hours, Yo per cent haematoxylin twelve hours, slight destaining with 2 per cent iron-alum while watching through the microscope, dehydration, clearing in xylol, and mounting in Canada bal- sam. The species cytologically investigated include: Alles- cheria Boydüi, Clonostachys Araucaria, C. Araucaria confusa, Cephalosporium Serrae, Gliocladium deliquescens, and G. penicilloides. An attempt was made to compare the wall composition of the ordinary mycelium and of the phialide by use of the polarizing microscope. The thinness of the cell walls and the difficulty of obtaining sufficient light with the higher magnifications prevented satisfactory observation. Hanging drop cultures, where the fructifications could be brought into focus, served best. ' The formula is: platinie chloride 3 ee, in 12 ee. of distilled water; osmie acid 2 ec. ; glacial acetic acid 1 ee. A fresh solution should be used for each application. 1936] PIN KERTON—CONIDIAL FORMATION 5 All of the organisms were grown at 30° C. The cultures were usually incubated in the light, since darkness often caused loss of color without other apparent changes. In an attempt to in- duce the perfect stage, glycerine agar was employed with H-ion concentrations from 3.8 to 7.6; and the following special media were utilized: gypsum blocks, Gorodkova-Maneval medium, earrot plugs, and corn meal agar. Attempts at cross fertiliza- tion were made, and while no perfect states were induced some interesting subsidiary observations were obtained. IIT. THE CONIDIAL APPARATUS A. THE CONIDIOPHORE The forms here investigated are believed to be closely related to Penicillium of the sub-family Aspergilleae in the Hypho- mycetes. The form-genus finds its perfect stage in Carpenteles Langeron (’22) of the ascomycetous order Plectascales, although the latter name is by no means in common usage. Its conidial apparatus consists of a basal stalk cell from which branches arise variously but more or less regularly for each species. The end cell of the penicillus, called the phialide, cuts off spores terminally in chains. The spores are temporarily adherent, due to connectives which later become disjunctors and probably are formed from the old wall of the phialide. In the material of the present study there are no disjunctors, and the spores mutually adhere in a more or less globular mucilagi- nous mass around the phialide. The spore ball is termed the cephalosporium in contrast to the entire fructification, or cephalophore. In most cases the cephalosporia are character- istic of the aerial mycelia, although submerged ones have been noted in Cephalosporium Serrae. The complete conidiophore varies from a short branch no longer than the spores produced (e. y. Allescheria Boydu) to a complex structure equivalent to the penicillus (e. g. Gliocladium); but in all cases where cephalosporia are produced the mechanism seems identical. 1. HISTORICAL a. On conidial production in the Aspergilleae and related forms.—Brefeld (’74) was one of the earliest to observe criti- [Vor. 23 6 ANNALS OF THE MISSOURI BOTANICAL GARDEN eally conidiophore formation although his methods were neces- sarily erude. He saw no need for the assumption of a special membrane outside that of the conidium, and thought that the process was one of budding. Referring to the forms with dis- junetors, Seynes (786) stated that the phialide wall formed the connective and the conidia differentiated their own walls within. The following statement of de Bary’s (’84) exhibited keen insight although his interpretation was incorrect: Bei reichlich inner halb enger und mit enger Miindung versehener Behiilter abgeschniirten Sporen wird durch die Gallert- oder Gummiabscheidung die Entleerung aus dem Behälter bewirkt, indem die durch Wasseraufnahme quellende Masse aus der Miindung hervortritt. [p. 75]. He believed that there was a gelatinous substance around the spore (a product of its outer lamella) which became deliques- cent by addition of moisture and so caused coherence. A number of investigators attempted to interpret the method of spore production in Thielavia basicola of the Plectascales, following Zopf’s original description in 1890. He reported an acropetal succession of spores whose lateral walls were dif- ferentiated into two layers, the outer being a sheath through which the conidia emerged successively due to the swelling of the mucilaginous middle lamella upon the addition of water. Gilbert (’09) thought that endoconidia were formed in the phialides and emitted consecutively by the force of the growing protoplasm in the basal part of the cell. Brierley (715) was the first to make a definite attempt to determine the method of conidium production, and he concluded that the method pre- viously called endoconidial was in reality merely acrogenous, being peculiar only in the mechanism of liberation. The co- nidium put down an additional wall within the parental one which was then ruptured and the new spores were constantly formed from below. The illustrations of the mode of new wall differentiation are quite anomalous. It was Duggar's view (709) that the endoconidia were ejected by osmotic pressure. Guegen ('05) cited Gliomastia chartarum as an example of endoconidial production. Buchanan (’11) thought that the spores of the cephalo- sporium were formed by the abstrietion of the tip of the 1936] PINKERTON—CONIDIAL FORMATION T sporophore, the amount of moisture in the air apparently con- ditioning the amount of adherence. He regarded as unneces- sary the separation of Hyalopus and Cephalosporium upon the basis of more slime in the former. Studies of Cephalo- sporium spp. on insects led Petch (728) to agree with Buchanan that the spores were really cut off in succession but became aggregated by the secreted mucilage. In a damp atmosphere some were found in loose chains due to the dissolution of the mucilage. He too disregarded the separation of Hyalopus and Cephalosporium. In Penicillium digitatum Scaramella (’28) illustrated the rupture of the parent sheath anew below each successive spore. Thom (710) thought that the spores were merely acrogenous disjunctions which produced their own internal walls and were held together by the phialide wall which did not necessarily rupture until late. This would seem to indicate continued elasticity of the conidiophore wall below the tip. In 1930 he added the following hypothesis regarding submerged cephalosporia: ... that the impulse to produce conidia when acting in a submerged hypha in which all parts are in constant contact with the nutrient, produces a trans- formed branch bearing spores which tend to enlarge, round up as if for germina- tion, and to lose the catenulate arrangement. This may produce a very peculiar spore attached as a single terminal cell or a cluster of cells not recognizable as conidia and without definite arrangement around the tip of the sterigmatic eell. [p. 73]. Wakyama (731), in cytological investigations upon Asper- gillus spp., indieated that the daughter nucleus migrated through a sterigma very similar to the condition in the basidio- mycetes, but he did not further investigate the mechanism. Finally Dodge (735) suggested that the mechanism of conidium formation in the genus Cephalosporium is probably close to that of Aspergillus and Penicillium although the conidiophore is not flask-shaped. If the spores are essentially in chains, they slip past one another very soon to form spore balls. b. Previous interpretations of the mechanics of spore pro- duction in general.—There are probably more common factors in all types of spore production than is generally realized, be- [Vor. 23 8 ANNALS OF THE MISSOURI BOTANICAL GARDEN cause the cutting out of similar small masses of protoplasm would not have many physical possibilities of variation. There have been numerous papers regarding this process in the sporangia of the Phycomycetes, a few on ascospore delimita- tion, and a number on the basidiospore release ; whereas, except for Brierley's work, the conidia have generally been neglected. There are a number of forces which have been championed for spore formation: vacuolation associated with exudation of water or some other liquid; centrosomes or similar bodies which are centers of force and direct the action of protoplasmic streaming; surface tension; Golgi apparatus or the like as- sociated with secretion; and electric forces. VacuoLEs AND ExupaTIONS.—Harper (’99) saw in the spor- angium of Synchytrium decipiens an early shrinking of proto- plasm probably associated with a throwing off of water which accumulated in vacuoles wherein cleavage furrows arose simul- taneously throughout the sporangium. In certain cases uni- nucleate masses were separated, in others multinucleate. He explained the inclusion of a nucleus by the characteristic chemical contents of nucleo-proteins and nucleic acids which were centers of moisture retention or at least had a greater af- finity for water than the surrounding cytoplasm. DeBary ('84) attributed the bursting of the aseus to in- creased vacuolization due to the absorption of water. In the cutting out of the sporangiospore segments of the Phycomy- cetes, he mentioned delimiting granular plates which somehow became gelified. Exuded droplets were shown by Buller (’22) to be significant in the disjunetion of basidiospores in all Basidiomycetes except the Gasteromycetes and the Ustilaginaceae. He found drop- lets also associated with the conidia of the Hymenomycetes and with the aeciospores of the rusts, but thought that such were merely passive. It is possible in these cases too, that, although not effecting violent spore disjunction, exudates may somehow bring about their release. In Coprinus sterquilinus he noted that a drop of water 5» in diameter appeared at the base of the basidiospore within ten seconds before its abjunction, and that 1936] PINKERTON—CONIDIAL FORMATION 9 the liquid was shot off with the spore and doubtless aided in attaching it to the substratum. It was supposed that the liquid came from the hilum, and was largely water, rather than local mucilagination from the sterigmata (often so narrow as 0.5 +). The latter, however, must have been ruptured, for it soon col- lapsed. Buller regarded the production of similar drops in Sporobolomyces as suggestive evidence that the genus be- longed to the Basidiomycetes in spite of the reduced vegetative growth and completely uninucleate state (Guilliermond, ’27). However, Stempell (’35) believed that the water-drop mech- anism was merely an ecological response and not of phyloge- netic importance. He reported its occurrence in association with both the sporidia and secondary conidia of Entyloma of the Tilletiaceae, and also with the sprout cells of Taphrina, usually included in the Ascomycetes. Buller thinks that the hydrostatic pressure of the basidial vacuole is very important in forcing the protoplasm and nu- cleus through the narrow sterigmata and in the final spore discharge, since the basidium collapsed immediately after the final release of one set of spores: **Now my experience, based upon extensive observations on the living hymenium, is that a basidium never produces more than one generation of spores and that after producing a single erop of spores, its sterigmata and body quickly collapse." [Buller, l. c. 2: 28.] CENTROSOMES, SURFACE TENSION, AND Goor APPARATUS.— Maire’s (’02) illustration of basidial development in Psathy- rella disseminata first suggested a centrosomal interpretation of the terminal granules in the imperfects studied. The four centrosomes in the former, produced simultaneously with the four nuclei, situated themselves at the apex and followed the protruding sterigmata with tails of streaming protoplasm (**eordons kinoplasmiques’’). eventually drawing the nuclei into the spores. These structures, formerly believed to be confined to fungi, algae, and animal cells, are now being dis- covered frequently in the higher plants. They are apparently centers of energy associated with the action of the spindle in the separation of chromosomes, as well as with the movement [VoL. 23 10 ANNALS OF THE MISSOURI BOTANICAL GARDEN of flagella and cilia, and are generally believed to be absent in the resting cells. There has been controversy over their origin —cytoplasmic or nuclear—and confusion of them with other structures. Faull (’04) concluded that the centrosomes were of extra- nuclear origin and occurred after reforming of the membrane. Guilliermond, Mangenot, and Plantefol (’33) stated that the centrosome was still an enigma and if it aided in ascospore de- limitation, its influence was purely physical, perhaps by the superficial tension on the surface of the asteral parasol which conditioned the formation of new walls. Harper (’99) firmly believed that the aster rays (fibrous kinoplasmic elements) cut out the ascospores and formed the limiting walls. They might well direct the spore delimitation but were themselves included within a special wall, according to Faull ('04). Surface tension and perhaps a different chemical composition kept the endo- spores discrete from the epiplasm. Buller (’09), in discussing surface tension and the physics of the jet theory of the ascus, stated that a free cylinder of any liquid under the influence of surface tension became unstable as soon as the length was greater than aD. It is a necessary consequence that if once realized the diameter of such a cyl- inder will divide into as many equidistant spore lengths as 1D is contained in one length. Some such physical interpretation must be necessary for the regulation of the size of the endo- conidial segments in the forms studied by the author. Scarth (’27) thought that a definite structural basis of proto- plasm was necessary for life movements. He observed a reci- procity of substance in the increase of mitochondrial elements correlated with the disappearance of the active kinoplasm as a result of vital staining; and concluded that the characteristic component of the chondriome, viz. lecithin, entered into the composition of the kinoplasm, and that the mitochondria merely served as reserve substances for the formation of the more labile elements. Vacuoles were thought to arise from the kinoplasm by the rounding up of fibrillae or as blisters or bub- bles from the covering films of the chloroplasts or of the central 1936] PINKERTON—CONIDIAL FORMATION 11 vacuole. The wandering streams of protoplasm in the cyto- plasm condensed into mucilaginous-looking globules which distended as vacuoles; and an essentially similar process ef- fected the cleavage of coenocytic sporangia. In some cases, however, fissuring arose from fibers that radiated from a cen- tral body attached to the nucleus, in others by invaginations of the ectoplasm or by outgrowth of vacuoles. The limiting film of these extensions of the kinoplasm became the plasma mem- brane of the spores, while the fluid which was secreted between them was presumably analogous to the sap which formed in the vacuole. Sharp (’34) regarded the divisions by constriction, seen in unicellular algae, in the budding of yeast cells, and in the ab- striction of conidia and basidiospores, as special cases of cytokinesis by furrowing. The centriole, or inside of the centrosome, has been generally believed to be a point of high surface tension, and the aster, according to Sharp, might be primarily an expression of streaming movements in the cyto- plasm which is somewhat more solid than the rest of the more gelified granular protoplasm. He further said that in the plant cell the furrows probably separated regions of relatively high viscosity even though no aster, in the ordinary sense, was present: ‘‘Alterations in surface tension and viscosity, to- gether with protoplasmic streaming, are obviously important factors in cytokinesis of certain types, but comparatively little is known about the initial causes of these phenomena.”” Wilson (’25) mentioned the appearance in epithelial mus- cular and glandular cells of a chromidial apparatus assumed to be extruded from the nucleus and of particular function in trophic actions of the cell. He also thought that there was a definite karyoplasmic ratio which regulated cleavage to pro- duce a fixed or typical cell size at a given stage rather than a fixed number of cells. It was the surface of the nucleus and not its volume that was directly proportional to the number of chromosomes. The Golgi apparatus is known only in animal cells and is con- nected with secretion. In goblet cells of the colon, secretory [Vor. 23 12 ANNALS OF THE MISSOURI BOTANICAL GARDEN granules have been said to have originated in mitochondrial (chondriosomal) regions and to have migrated to the Golgi region where they were transformed into mucin granules (Duthie, '53). The granules were then liberated into the cyto- plasm, finally dissolved, and passed as liquid secretion into the lumen. This apparatus has also been suggested as a center for the formation of enzymes. Whether a similar process occurs in plants with liberation of mucin has not been investigated. The function and limitations of mitochondria are also disputed questions. Guilliermond (’29) homologized the plant vacuome —which he thought arose de novo from the cytoplasm—with the Golgi apparatus and Holmgren canals of animals. Nas- sonov (723) believed, on the basis of staining experiments, that the vacuolar membrane had a secretory function and was homologous to the Golgi apparatus. Weier (732) thought that the young plastid in sperma- togenous, archesporial, and epidermal cells of Polytrichum commune and young androcytes of Pellia epiphylla showed staining similarities to the animal Golgi apparatus and believed that such might be associated with enzyme secretion, in the former case linked with starch production and in the latter with secretory granules. He did not think, however, that the two should as yet be considered homologous. Sass (734) offered further evidence of a Golgi apparatus in the Basidiomycetes, although he failed to associate it with any property of motility as in animals or with the plastidome re- ported by Weier. The ‘‘nebenkern’’ in Coprinus sterquilinus was found to divide into four **Golgi crests’’ which eventually became situated at the tip and presumably cooperated in the formation of sterigmata and in nuclear migrations. These ele- ments resemble superficially the centrosomes of Maire ('02). His results were especially interesting in that he used various new fixatives absolutely free of osmic acid, which eliminated possible errors due to diverse osmophilie substances. 2. INVESTIGATIONAL RESULTS Because in all cases investigated the mechanism of spore formation has appeared to be similar, the results shall be out- 1936] PINKERTON—CONIDIAL FORMATION 13 lined in general for the whole group. Plate 1 includes exem- plary types selected at random and the figures here mentioned refer to that plate. The incipient phialide shows a blunt tip which is thicker than the lateral walls and stains darker, be it typically flask-shaped as in Gliocladium fimbriatum or essentially straight as in Cephalosporium niveolanosum. The dense granular contents do not extend quite to this tip since there is a hyaline area be- tween (fig. 2). A chain of small round granules collect longti- tudinally, and in fixed preparations show ‘‘tails,’’ suggesting rapidly streaming protoplasm (fig. 3). These granules appar- ently merge into several small rods (fig. 4). Presumably at about this stage a drop of liquid is exuded terminally (figs. 5 and 6) ; at least in living cultures on hanging drops and agar slides these drops are infrequently noted on blunt hyphae. In the latter material the globule sometimes disappears as it is viewed and runs down the side of the phialide, producing the obviously different appearance of this structure, so often noted, as compared to the rest of the mycelium. Inthe hanging drops, with low magnifications the downward-hanging fructifications portray eephalosporia with spores literally swimming in the slime which here has not been disseminated by touching the glass. Inthe next step (fig. 7) a cross-section of the tip shows three granular rods, horizontally aligned, from each of which is a narrow short strip of cytoplasm leading to the protoplast proper. Figure 8 shows a split between four rods which **hold open the sac." The blunt tips seem to be especially resistant at the corners, and the adherence of rods to the sides probably increases this opposition so that the tip swells in the center slightly before splitting. In the meantime, presumably associated with the exudation of liquid, the contents of the phialide have cleaved into a vary- ing number of segments which at first are noted as regularly occurring cross-lines of granular dark-staining material, indi- eating regions of high surface tension. Later, narrow hyaline areas appear between the segments which round up at the edges. At the very base of the phialide are a varying number of small vacuoles. The spore, clad only in a fine membrane, be- [Vor. 23 14 ANNALS OF THE MISSOURI BOTANICAL GARDEN gins to protrude from the rigid collar of the phialide whose narrowness considerably retards the process (figs. 9 and 16). As if forced out by pressure from the inereasing vacuoles at the base, the spore becomes somewhat constricted, as indicated by its tendency to round up on emergence, and causes the in- cluded end to become increasingly convex. Finally, after the greater part has emerged, the rest of the spore snaps off, and is immediately followed by a successor. This process is rela- tively slow, under optimum conditions only three spores being emitted from a phialide during seventy minutes (fig. 17). The exuded slime and the moist surface of the spore hold it near the phialide. In the spore ball it cannot escape until moisture dis- seminates the mucilage. Very thin sections of paraffin material seem to indicate that the spore is abstricted as is the yeast bud (fig. 15). This appearance may be explained by assuming that one of the granules at the tip really stays in the center of the phialide and draws the cytoplasm as a narrow stream into the spore; and in some thin sections it is impossible to see any lateral granules. Moreover, young phialides show blackish regions at the tips which probably represent central granules. Since the greatest diameter of the phialides averages only 2» and the contents are granular, it is impossible to make out nuclei in the celloidin material, but in the thinner paraffin sec- tions each cell and endoconidial segment appear to be uni- nucleate. The energid is but slightly larger than other dark- staining metachromatie granules throughout the cytoplasm. No definite mitotic stages were espied. The fine stream of protoplasm passing into the spore stains very darkly, perhaps due to a passing nucleus or to compression and rapid stream- ing. Sometimes mature spores show numerous refractile granules, at other times a more definite accumulation which is regarded as a true nucleus. After emergence, the spores form a thin wall around the membrane, swell doubly or triply, and germinate after several days. In some cases they appear to be germinating while still within the ball (pl. 5, fig. 16). Since the mucilage swells upon the addition of water, when favorable moisture conditions oceur the spores are released both for dispersal and germination. In the meantime, they have been 1936] PINKERTON—CONIDIAL FORMATION 15 protected by the slime. Thus cytologically the conidia in these species are haploid representatives of haploid individuals, ap- parently capable of indefinite existence without fertilization. While the spores are being released the vacuolar area within the phialide multiplies conspicuously. The latter is increas- ingly contorted and devoid of protoplasm (especially notice- able in the larger single conidiophores of Clonostachys Arau- caria). Doubtless, the vacuolar force, as well as the mechanical one of twisting, aids in the liberation of the spores. Neither by differential staining nor polarized light were any essential dif- ferences in the chemical nature of the phialide and ordinary mycelium observed. Young tips of ordinary hyphae were more clearly observed in hanging drops than on agar. They are always more or less pointed and of the same consistency as the rest of the wall. Vacuoles extend quite close to the apex and contain one or sev- eral very active and large dancing bodies (pl. 3, fig. 21). Buller (733) mentioned highly refractive particles (Woronin bodies) in Ascobolus pulcherrimus, on both sides of the septa and in terminal cells in a dispersed condition. In older cells they were said to settle down in the walls of the vacuoles, never in the protoplasm proper, and were associated always with the sap cavity, even passing through septa with it. These may be identical with the large dancing bodies, or even the larger septal granules (pl. 3, fig. 20), although the author does not think the latter two are the same. In fact, there seem to be dancing bodies which appear dark even in vivo, and cytoplasmic granules which show only in fixed preparations. The crowded vacuoles with small metachromatic granules visible between them soon occupy approximately the whole cell except for a narrow peripheral layer and a cytoplasmic region in the center where the nucleus is hung (pl. 3, fig. 19). Around all aerial hyphae on agar slides was noted a fine layer of mucilage, which indicates that apparently such exudation is normal in the life of the organism, and differs only in quan- tity in the phialide. For comparison, a cursory examination was made of liv- ing material of Scopulariopsis brevicaulis (Sacc.) Bainier, a [Vor. 23 16 ANNALS OF THE MISSOURI BOTANICAL GARDEN species whose spores are held together in chains by disjunetors. The process agrees very well with that shown by Scaramella (728) in Penicillium digitatum Sace. The terminal uninucleate segment puts down an internal wall as a chlamydospore does, and by its swelling ruptures the external wall at the shortest diameter of the ellipsoidal spore which remains adherent at the base and connected with the rest of the phialide. This leaves empty a lens-shaped connective between the chain of spores abjointed similarly and successively. Meanwhile, there has been a slight cleavage of the protoplasm into a number of seg- ments within the phialide, so that although only one spore be- comes encysted at a time the potential ones are already present as in the phialide of the cephalospore. The main differences in the latter are the lack of resistant spore walls and method of rupture of the phialide, whose functions, as well as that of the disjunctor, are taken over by the mucilage. Comparative cyto- logical data would be interesting. 3. INTERPRETATION It seems probable that the incipient phialide eventually forms a rigid tip which is no longer capable of stretching and of intussuscepting material. The contents of the cell are con- tinually swelling due to vacuolar as well as osmotic pressure and to surface tension, indicated by accumulation of granular material which eventually causes a rupture. In the meantime the granules have strengthened the corners so that the middle of the tip is stretched and broken and the size of the opening is considerably narrowed. The exudation of the mucilage, whether effected by the granules or not (though such seems likely, and it must come from the living protoplasm itself), doubtless makes possible the cleaving of the protoplasm as has been suggested in the basidium and sporangium. Whether the terminal granules represent centrosomes or Golgi bodies, the author is not prepared to state; but at any rate they must rep- resent kinoplasmic elements similar to those of Scarth (’27). They may be an aggregation of chondriosomes whose signifi- cance is not clear. 1936] PINKERTON—CONIDIAL FORMATION P Further, the mucilage production may be compared with guttation in the angiosperms (Wilson, ’23), where there were found 220-1030 ppm. of organic material, mostly nitrates and nitrites, in the exudate. This indicates that occasionally col- loidal material can be passed through intact walls. The ter- minal wall was not broken in the phialide at first, since it was sometimes visible within the droplet. The theories that the slime arises from between the membranes, as suggested by Knoll (712) in eystidia, or by deliquescence of the middle lamellae (Brierley, ’15) or of spore walls themselves (deBary, '84), are plainly inapplicable here. Granules such as those occurring terminally in phialides are not noted in the ordinary mycelial tips, although similar (?) ones (pl. 3, fig. 20) are discernible in ordinary cross septa, per- haps exercising a function associated with the protoplasmic connectives. The cleavage itself is largely a surface tension phenomenon and is doubtless often only partial, i. e., the cleaved parts occur only in small numbers towards the tip of the phialide and this process may be repeated ; hence all the nuclear divisions do not have to take place at onee. Such is exemplified in embryo- logical studies of cleavage of animal eggs. Possibly, due to the water-holding capacity of the nucleus, liquid is exuded equi- distant between the energids where increased surface tension is noticeable in fixed material by a layer of basophilic granules. Buller’s physical theory would explain the size of the segments (ef. p. 10 of the present paper). The spores are released by the vacuolar pressure in the lower part of the phialide. Cohesion and surface tension factors, associated with a narrow rigid opening and the tendency of an elastic portion of a spore to round up when the pressure is released, assist the process. The size and shape of the spore are doubtless dependent upon the extent of the phialide and its opening as well as upon the amount of cytoplasm included in the endoconidium, all being explainable upon a purely physical basis. That the species of Cephalosporium have the longest narrowest spores is due to a comparatively straighter phialide which makes the endoco- [Vor. 23 18 ANNALS OF THE MISSOURI BOTANICAL GARDEN nidial column more extensive. Extrusion thus takes longer, and the membrane or new cell wall becomes more or less rigid before the spore is completely released. The purely physical force of pressure must also be assisted by that of twisting caused perhaps by spiral streaming of the protoplasm. Although the author has seen no previous men- tion that this force assists in spore dispersal, Castle's ('34) reference to it in growth is suggestive. He states that the twisted or helical forms of growth are generally due to the reso- lution of two growth vectors: one longitudinal and one rota- tional. Oort ('31) had measured simultaneous growth and twisting along the axis in the coenocytie spore-bearing cells of Phycomyces and found at 17.5? C. an average rate of elonga- tion of 39 » a minute and an average rate of rotation of 3.7 de- grees a minute. The angle at which micelles were incorporated in the wall was about 6 degrees. The steepness of the growth spiral was not structurally fixed but was reversibly altered (re- versed or abolished) by change of temperature. The majority of the cells grew in the form of a right-handed helix, i. e., left- handed thread on a screw. The rapidity with which the angle of coiling changed seemed to refute the interpretation of altered proportions of different types of isomeric molecules. It may be that spiral streaming of protoplasm helps in the phialide twisting, but this is purely conjectural since no proof has been obtained. The eventual collapse of the phialide results from the lack of compensating turgor pressure from the end wall which has become open (also an aid in the spiral twisting). Such a condi- tion is less noticeable in the typical penicillate fructifications which have the most minute openings, especially in comparison with those of the rest of the phialide, than in the broader, straighter, and simple ones of Clonostachys Araucaria. If the spores were merely produced acrogenously there would be no exhausted remnant; moreover the endogenous condition here is more similar to that in the basidium and less like that in the ascus because there is no residual epiplasm. The slime forms a protective covering around the delicate 1936] PINKERTON—CONIDIAL FORMATION 19 thin-walled spores until optimum moisture conditions arise, when it not only makes the spores available for germination but also causes considerable dispersal due to the colloidal swelling accompanying its solution in water. Because of this miscible property the amount persistent around the spores will depend upon the humidity of the atmosphere. In very dry conditions the external layers tend to become almost rubbery, as ascer- tained by Cooke and Ellis (’78) in the solid slime balls on her- barium specimens of Penicillium repens. The submerged cephalosporia of Cephalosporium Serrae ap- pear to be homologous with the aerial ones except possibly for a greater tendency to irregularity in shape due to the confining action of the medium. In Torulopsis mucilaginosa, on the other hand, the whole growth is obviously a budding pseudo- mycelium and actual production of endoconidia was not ob- served. In Cephalosporium symbioticum, the cephalosporia attached to the short laterals more nearly simulate the fructifi- cation of Allescheria Boyd than that of Cephalosporium. Since this condition occurs in T'orulopsis mucilaginosa only in submerged growth, it is an abnormality, and whether it may in- dicate a foreshadowing of the conditions in the regular forms or a reversion from an advanced to a reduced type (the yeast) is still a question. The latter view is the more logical. In agreement with Wakayama (31), the present data indi- eate that the conidiophore is an asexual organ of the haplont producing uninucleate spores which repeat the life cycle with- out variation except for secondary asexual spore forms. The occasional conjugation of conidia is without significance so far as is known. B. OTHER CONIDIAL FORMS ENCOUNTERED Omnipresent in this group are terminal or intercalary chlamydospores whose new cell walls are put down inside the parent cells. The production of chlamydospores is ordinarily associated with adverse conditions, and dissemination occurs when the adjacent cells die. Besides these, in Gliocladium fimbriatum and G. deliquescens large terminal cells rich in oil [Vor. 23 20 ANNALS OF THE MISSOURI BOTANICAL GARDEN droplets are formed, which differ from chlamydospores in their relatively thinner walls and capability of immediate germina- tion as evidenced in hanging drop cultures. For such, the designation akinete has been adopted from algological termi- nology. Moreover, there are noted within the medium enlarged round to ovoid terminal cells beyond which the hypha may grow again (pl. 3, fig. 30). Their subsequent development was not followed, but when observed they were thin-walled. They may possibly be aborted ascogonia, and Thom has referred to them as due to the same impulse as that causing submerged cephalo- sporia. If they function as chlamydospores, at least their method of formation is somewhat irregular. The peculiar short-stalked spores produced promiscuously over the terminal clubs in Clonostachys spectabilis cannot be discussed in detail since the culture died early, but they are obviously not comparable to the other forms studied. A super- ficial resemblance between spores and the warty structures of the phialides of Clonostachys Araucaria may have caused Oudemans (’86) to include that species in Clonostachys. How- ever, the warts are not spores but thickenings on the wall, not uncommonly seen in fungi and are incapable of germination. Infrequently oidia were noted in Gliocladium V ermoeseni (pl. 4, fig. 11). The significance of these secondary spore forms is obscure. If such occur in Penicillium and Aspergillus, Thom (726, ’30) did not mention them, unless the ‘‘hiille’’ cells first cited by Eidam (783) in some species of Aspergillus could be considered such. It is obvious that hypnosporie types of spores are produced in response to unfavorable conditions, especially where the colony has exhausted the medium and an indefinitely resting spore is needed. IV. PuvsroLoav A. COMPARATIVE CULTURAL CHARACTERISTICS In table 1 are listed most of the organisms studied on a repre- sentative number of media, with brief notes regarding color, surface, and microscopic characteristics. Some correlations COMPARATIVE CULTURAL CHARACTERISTICS IN VARIOUS MEDIA : Nitrate Czapek Potato dextrose Corn meal Gorodkova-Maneval Sabouraud’s dextrose ong broth [Sabouraud’s Conserv i sC ation| Glyeerine Organism pH 5.5 pH 4.43 pH 5.7 pH 6.0 pH 6.8 pH 5.6 68 ` pH 7.0 pH 6.9 Carrot qu Allescheria Gull-gray, powdery, e or flat aerial. |Gray-green, black below, Gray, powdery; single|Tan-gray, flat, matted,|Pale-gray, 1 Me som "T y, long, fuzzy;|Gray pellicle Biz small ra-|Gray-violet, black, flat,[Dark gray, corrugated;|Gray-green, sho fuzzy. Boydii merged; single spore wrinkled; coremia; sin-| spores. filamentous; single} many single spores. i dia = colonies, cu Ze ; Bac y ZS es. " idm ae gle spores. spores, Cephalosporium |Apricot-buff,cerebriform;|Orange, rough, heaped,|Deep orange be oed Pale pink; submerged|Pale pink, flat, moist,|Pink, corrugated; cor-|Pink ring, liquid turbi Pink i wrin- 3 ili chlamydospores; sub-| edges submerged. cephalo — T y chlamy dospores and wil + : s emia; sual ohio dan ` Er"? TA nl. Pee ae ER gien me merged cephalosporia. Spot cephalosporia. : emia. C.niveolanosum |Shrimp-pink, flat, wrin-|Bluish-white, flat, with ipio Pl heaped in|White, wrinkled; many|Pale tan, heaped, hairy|Few cephalosporia. White yeasty ring, slight| White, round elevated cen- | Seashell- p Loan ines bent like; few kled ; chlamydospores; radiate striations. enter, hairy projec- cephalosporia pro otrusions; chlamydo- turbidity. ters and spreading| with core submerged cephalo ag sterile. spores. airs. sporia; long coremia. C. rubrobrunneum |Orange-pink, flat, corru- White, edges submerged,|Coral-pink, corrugated ;|Deep pink, flat, moist;|Pale pink, wrinkled, much|Pale orange, flat, moist;|White ring, slight turbid-|White-silvery, hard cere-|Frosted seashell-pink, White, tallowy. ated ; submerged eeph-| later cottony. few cephalosporia. ee ia submerged, sterile. few cephalosporia. ity. briform colon cerebriform; alospor cephalosporia, C. Serrae Gull-gray, short-fibrous; White, heaped, rough,|Steel-gray; many chlamy- E E powdery;|Gray-powdery, flat, wrin- Downy white pelliele. Yellow, powdery tops,|Pale pink, cottony, White, tallowy, some submerged ehlamydo-| edges submerged. dospores; few cephalo- uch submerged chlam-| kled; chlamydospores. wrinkled. slightly wrinkled. fuzz; some co spores and cephalospo- sporia. ydospores; few ceph ria. alospor C. Stuehmeri White, short, rough sur-|White, flat, granular, fim-|Pale EM ad eor-|Pale pink, slightly ele-|Opaque, round elevated, AN moist; some cor-|White yeasty ring, tur- White, hard, wrinkled nr wrinkled below ; eor-| White, tallowy. face, submerg ed; many] briate. mia ; cephalo-| vated, dry; many ceph-| radial furrows, sterile. mia; few cephalospo-| bidity. sheli, orange below emia. cephalospor sporia alosporia. N remia. C. symbioticum ale p mostly sub- d maed with pink iori mde, Arie A sub-|Pink, flat, wrinkled; sub- Yellow- -green p ;|Pink then black, mostly|Orange coremia and pink-|White powdery. merged; few coremia; medium grass-[merged cephalosporia. merged Y aH pink submerged hyphae.| submerged; cephalospo-| yeasty then black; many cephalosporia. and chlamydospores. ria and ehlamydospores.| chlamydospores. Clonostachys Citron, in ium too,| White, Wée? raised,|Pink, cottony; many|Pale pink, scatter 1¡Cephalosp Pale pink, dry, powdery; roe white pellicle,|Pale pink, cottony; ceph-|Shrimp-pink, downy, um-|Pinkish- Tu short- Araucaria tem Ti pad: granular center, orange chlamydospores; è few dry heaps; Gi many cephalosporia. me hyaline bas al alosporia. bonate, wrinkled below.| haired fuz few aerial cephalospo-| below; ps wee ria. cephalosporia. ria; chlamydosp h. ria. Clonostachys Primuline-yellow, in me-|White-opaque, diserete|Chalky white, heavy, flat- Opaque, flat; cephalospo-|Chalky-white, matted, White yeasty ring, slight| White, yellow below, wrin-|Pink center, rough, hairy, White, small cottony, raucari Wes too, powdery, sub-| tufted colonies; few) matted growth, sterile.| ria. coarse; hlamydo-| turbidity. kled. wrinkled below; ceph-| sparse growth. var. rosea erged; few aerial Weg ia. spores; cephalosporia. alosporia. tee, C. Araucaria Orange- Sa nen ;| Pink, raised, fuzzy. Pale pink, flat with pow-|Pink, scattered tufts|White, rough, matted; Pale pink and white,|Pink-white, powdery pel-| White, matted, yellow be-|Pink-orange, thick-mat-|Pink, fuzzy, good growth. var. confusa da rid tufts; eephalospo-| dry; cephalosporia. cephalosporia. downy. liele, liquid clear. low, wrinkled. ted, wrinkled below; al Gliocladium deliquescens G. fimbriatum G. Vermoeseni cephalosporia White, green dots, mostly b submerged scattered threads. e. ée green, medium yel- w-brown, felted; cephalosporin and aki- Pink EN many| cephalosporia White, fuzzy; A — cephalospor Dark green, white edges; cephalosporia. White, sparse mycelia; pink cephalosporia. Blackish-green, medium brown; aerial cephalo- sporia. Deep green, medium tan, matted, moist; many cephalosporia. White, fibrous; o. spores; erged e Bei Green dots scattered, -—— growth; loose spor Green, Dan many ceph- alospor Pink, powdery, elevated; | few small cephalospo- ria. Sterile. Yellow-green, powdery, moist; cephalosporia. White, powdery, sterile. White; many cephalospo- ria. White, medium yellow- green, cottony. Transparent, scattered. Gray-green surface pel- below, li- Pale green, feathery pel- licle, h in liquid too. Pink ring, liquid clear. Pink spore balls, white eottony hyphae. len thin-walled cells. White cottony center; cephalosporia at edge. Meadow-green, acess white, rapid ene growth, medium yellow; many ones sporia. White mycelia; ro spores; rapid gro Delicate green fuzzy rowth. White hairy growth. No growth. 1936] PINKERTON—CONIDIAL FORMATION 21 may be drawn here, indicating as favorable media the glycer- ine, Sabouraud’s dextrose, and Czapek’s agars; and as unfa- vorable, nitrate and Gorodkova-Maneval agars; as less so, corn meal and potato-dextrose agars, carrot slants, and liquid media. Growth on the first group is typified by abundant large, aerial mycelia with comparatively few types or numbers of spore forms, deeper internal coloring with little if any exuded pigment, much guttational water, and a convoluted colony. Un- favorable conditions are recognized in sparse surface growth of long slender filaments, a tendency to submerged mycelium, relative increase in reproductive bodies such as chlamydo- spores, conidia, and coremia, dry powdery growth, paler or tending toward yellow rather than orange pigmentation endog- enously, with frequent exogenous coloration of the medium. In the next section some of the prominent factors and theories for cultural deviations which seem especially ap- plicable in the case at hand will be more fully discussed. B. FACTORS INFLUENCING MORPHOLOGY AND PHYSIOLOGY Of the numerous variables possibly relevant, the present dis- cussion is limited to hydrogen-ion concentration, surface and oxygen tensions, and kinds of nutrients. Temperature, doubt- less influential, will not be considered, since all were grown at 30° C. 1. Hydrogen-ion concentration—It must be remembered that the hydrogen-ion concentration refers to the relative acid- ity or percentage of dissociation as contrasted with the total acidity ; although the latter factor, as well as the chemical na- ture of the acid itself, also influences growth. Moreover, micro- organisms contain natural buffers and have to a limited extent the power to adjust the pH of a medium toward the optimum as a result of their activity. The media utilized covered a wide natural range of H-ion concentration, from Raulin’s (pH 4.15) to Endo’s (pH 7.5), with growth upon all. An artificial series from pH 3.8 to 7.6, maintained upon glycerine agar (table 11), gave the most strik- ing results in the tendency towards submerged growth at the [Vor. 23 22 ANNALS OF THE MISSOURI BOTANICAL GARDEN higher concentrations of pH. At pH 4.4 and below and at pH 7.6 the media did not solidify. It is interesting that Glio- cladium deliquescens grew in the liquid at the higher acidities, but formed a pellicle at the higher alkalinities. On the other hand, Torulopsis mucilaginosa grew extensively and normally upon wort agar (pH 4.8) but with mainly submerged cephalo- sporia upon nitrate agar (pH 5.5); whereas the true molds showed an inundated state upon the latter and abnormal aerial conidiophores even upon wort. Organisms listed as Monilia albicans have been extensively used in physiological studies. Marantonio (’93) stated that at the higher acidities there was a greater quantity of mycelium; and that there was a tendency to filamentous growth on liquid, with sprouting almost exclusively upon solid media, or hyphae only on old cultures. According to Fineman (’21), mycelial growth was better under low surface and oxygen tensions, but the yeast form predominated upon solid media, simple carbo- hydrates, and a low acidity. Mallinkrodt-Haupt (732) stated that the yeast form predominated in a strongly acid medium, whereas the filamentous occurred in a very alkaline one; and he also mentioned that H-ion concentration influenced the metab- olism, enzyme actions, morphology, toxin production, and im- munological characteristics. Buchanan and Fulmer (’28) found that some of the molds could tolerate a range of pH from 1.6 to 11.2, but that there was an optimal section with atypical growth deviations on either side. With fermenting yeasts alcohol was formed under acid conditions, and glycerol as an intermediary under alkaline ones. In Endomyces capsulatus, Moore (’33) said that in acid media the tendency was towards abundant budding, numerous conidia, and longer narrower cells; contrarily, on alkaline sub- strata there was an inclination towards shorter thicker cells with yeast-like formations and fewer conidia. Hewitt (’33) said that the pH had an effect upon the oxida- tion reduction system, which resulted in an alteration of the ionic equilibria. Clark (722) thought that the H-ion concentra- tion functioned chiefly as a conditioning agent and was only in- 1936] 23 PINKERTON—CONIDIAL FORMATION "wpodsoreqdeo ou sjuoguoo '8juoquoo snou “et1ods -o[peudoo ouros íf1ie[nueid |'e10dsopeudoo ou “S[[8M uo[[OMS FO SUIL 9117) "pegiour ‘Seq 9UI[S ‘sjop us913 ['19B3uo, pue UA *po3103u09 S[[£^ [[99 Leen |-qns j3eqAouros ‘ous uee13 |uoo18 qy” iiam 931A |po310guoo SSe[ ba *re[nue 13 Sue y Zurjerpez |pue qg34013 Iuta uyj4O1d (|q34012 099g [y3mol 3 pa |s? f prnbr ut “porra uly} su? |susosanbyop pue oppd Laeopjovezans [[np 9814/9414 M 3uo[nooopq|-zns SPA eurjp|-Z1eumqns W)4}4018 SS9[IO[O) 3 pediowqng) = wpe) ‘sorods -OpLUuUtI]yo E “grua 3uruo»oq 2DAA2S -100 {03300 'Sonuo[oo juoos “yymo13 wni1ods "ymor ON| 'poe3jnj NYM l'pepruria “9314 M|yoeduoo pedeogresi1e09 qurd "34018 ON "j34018 ON -0101d2;) "u1401d "jSrour "grua oo podiouqns ‘umop ‘uow j|xurd pue "981800 amb məş “£iopmod DEN{U0D "IBA jo 3unoure [[gurg|-[es uay} 931]M|£uo3309 LI9A\‘Ku0}400 ‘quIg|‘asuBIO -yurd "j34018 ON :qym013 3qZti[g| DMIDINDLY '9 *Urmnruio ‘kip “£iepmod "jsrour. ‘pory -109 euo Cap ‘quid Zuruoo |-urim ‘umop ‘kiopmod |-mod qurd MADINDAF "j34018 ON|-0q UMOP 9319 M|9yrya pue xurq|/£ggnjg ‘yurg|‘wnyootu oug *"¡m013 ON "(34018 oN| s/i90250u0];) “PITT "jstour *1903uo9 -urim ‘poom yovtq UMmOp "(34013 "[24012 |-qns 3e Meuros uphog "jj401d on|4e13 quermmxnTi439A194 -Lerig y 3rs £19 AlZ3oA[oA ‘Leip "ymor ON *ym013 ON DILIYISINMP 9'; Hd 04 Hd +9 Hd 0'9 Hd os Hd rr Hd 8'e Hd usrue210 UVOV HNIHSSOA'I2—NOLLVLLNWHONOO NOI-NHDOUGAH HO LOI II d'IS ViL [Vor. 23 24 ANNALS OF THE MISSOURI BOTANICAL GARDEN directly concerned with chemical transformations closely re- lated to life processes. 2. Surface tension.—Surface tension is the cohesional pull of the internal and adjacent molecules upon those of the sur- face layer. It is particularly important in the life of the organ- ism in relation to adsorption and absorption. This tension produces surface energy, which, unhindered, effects the as- sumption of minimal surface for a given volume, i. e., a sphere. At the interfaces which are sites of chemical and physical ac- tivity, surface tension is constantly being opposed. Anything causing a lowering of surface tension will permit a larger sur- face; for example, mycelial growth will be favored, but under the other condition the cell size will be limited and yeast-like proliferation result. The appearance of filamentous forms in liquids is probably a reaction due partly to surface ten- sion, and represents unfavorable adaptation to an unfamiliar substratum. Artificial reduction of surface tension has been secured by Fineman (’21) with the addition of sodium ricinoleate to a medium with dextrin solutions, but not those containing galac- tose. Glycerine has also been found to lower surface tension and increase mycelial growth. Upon the basis of similar ob- servations, Czapek (’11) was led to stipulate the necessity of lipoids in the surface membrane. Surface tension is greatly affected by the polar orientation of interfacial substances, the amount of ionization, and the pH (which determines the basic or acidic activity of amphoteric proteins). Larson (’21) attributed his success in growing normal pel- licle-forming bacteria on the bottom and vice versa by lowering or raising surface tension, to the non-wetting properties of the lipoids in the surface pellicle. Buchanan and Fulmer (728) re- ported similar results in making Clostridium tetani grow su- perfieially, but they suggested that the surface-active solutes concentrated in the surface boundary might also cut down the solubility of oxygen. They also mentioned the tendency of Bacillus subtilis to sporulate in media of low surface tension. Oxygen tension, availability of nutrients, etc., may overbalance or at least modify the effects of surface tension. 1936] PINKERTON—CONIDIAL FORMATION 25 3. Oxygen relationships.—The aerobic and anaerobic states are dependent upon the oxygen relations. Topley and Wilson (729) stated that anaerobes were not supersensitive to oxygen itself, i. e., that oxygen was not toxic, but that hydrogen perox- ide was produced abundantly in the presence of free oxygen for whose destruction no catalase or equivalent system was pres- ent. This theory also explains why anaerobes can sometimes be cultivated with an aerobic form in air, the latter destroying the peroxide. Fineman (721) found mycelium formed in Monilia albicans on surface agar slants cultured in carbon dioxide. The usual correlation of budding and high oxygen tension and of myce- lium and low oxygen tension was noted, but there were excep- tions. According to Fuchs (726), reduced oxygen tension fa- vored sporulation in yeasts. Dodge (735) stated that lowered oxygen tension favored hyphal production to some extent. The metabolic relation to oxidation is also intimately associ- ated with the substratum. Stephenson (Buchanan and Fulmer, '8) found the normally aerobic Bacterium coli (Escherichia coli) developing anaerobically during the first stages of glucose breakdown, and thereafter aerobieally. However, if grown in pure oxygen it utilized such from the beginning, nor was there any increase in the ratios of glucose decomposition, but the ac- tive metabolism persisted longer than in the former ease and the acidity increased more slowly. Thus the apparent effect of the inereased oxygen was to induce further breakdown of some acid product of the fermentation of glucose. An in- creased oxygen tension accelerated the respirational rate when dextrose was replaced by ammonium-lactate, and made pos- sible the derivation of carbon from salts of acetic or succinic acids but under aerobic conditions only. Ina synthetic medium of simple organic acids (such as lactic, succinic, fumaric, or pyruvic) to which nitrate was added, the organism could grow anaerobically by utilizing the reduced nitrate as a hydrogen acceptor or oxygen donator. Buchanan and Fulmer (’28) further noted that in Bacterium coli (Escherichia coli) and Clostridium Welchii smaller amounts of carbon dioxide were produced aerobically than anaerobically. [Vor. 23 26 ANNALS OF THE MISSOURI BOTANICAL GARDEN Hewitt (’33) offered an explanation of aerobisis and anaero- bisis on the basis of oxygen-reduction potentials in place of the view that anaerobes have no catalase. He showed that aerobes (such as staphylococci, streptococci and pneumococci) reduced the broth medium whereon they were cultivated, which had an initial potential of almost —0.3 volts, to as low as -0.15 to —0.20 volts. Anaerobes, on the other hand, could not begin growth on such a medium, but with pieces of meat in the broth the potential became reduced to —0.2 volts, due to a reversible oxidation-reduction system therein. Such potential was not changed by staphylococci which had reached their minimum, but was lowered by anaerobes to a limit of —0.4 volts. The pre- liminary reduction of the medium might be effected artificially by four methods: (1) hydrogen in an anaerobic jar, (2) pure nitrogen, (3) various chemical-reducing agents, (4) growth of aerobic organisms. 4. Kinds of nutrients.—Besides the above-mentioned inter- relations of oxidation and nutrient necessities, there have been some definite observations regarding the nature of the latter alone. Talice (730), in his extensive study of media and conditions favoring the formation of hyphae, found that mycelium was formed the first few days and then mainly yeasts, with the fila- mentous form only in contact with the agar surface. Surface tension may enter here and oxygen phenomena, since the inside of a colony is under reduced oxygen pressure. He noted that normal yeast forms produced hyphae with dextrin peptone media or glucose, and to a lesser extent with protein. Inciden- tally, he regarded the yeast state as a senescent one. Plaut (713) found yeasts in sugar-rich media and mycelia in sugar-free nitrogenous substrata. Buchanan and Fulmer (^28) spoke of the ‘‘protein-sparing”’ effect of dextrose. Top- ley and Wilson (’29) thought that there was no such effect, but rather an inhibition of bacterial growth, due to the rapid in- crease in hydrogen-ion concentration of those media contain- ing a fermentable carbohydrate. Further conclusions have been drawn from work with Mo- nilia albicans. Roux and Linossier (790) stated that the com- 1936] PINKERTON—CONIDIAL FORMATION 27 plexity of morphological structure increased with the molecu- lar weight of the food elements, i. e., yeasts in simple sugars, and filaments in complex carbohydrates. Such statements have not been fully confirmed. According to Mallinkrodt- Haupt (’32), the yeast form appeared in the tissues of the host, but in culture the mycelial one. Discordantly, Fineman (’21) reported both conidia (yeasts?) and mycelium in the host, but only the one or the other in culture. No explanation was of- fered but if such be true, the heterogenous nature of the inter- nal environment may be responsible. 5. Discussion.—It soon becomes obvious, when trying to check physiological causes and effects, that there are many var- iables concerned, making it practically impossible to attribute a definite reaction to one alone. Nevertheless, some likely in- fluences can be designated. There is an optimum range of hydrogen-ion concentration between pH 5 and pH 7, and the best is around pH 6.4 for the majority of these forms. Below and above the range the growth is restricted and the relative number of reproductive bodies, especially chlamydospores, augmented. Moreover, the maladjustment is exemplified by atypical colony characters— i. e., submerged contorted growth at the higher acidities and aerial if scanty growth with increased alkalinities. The forms studied show strong aerobiosis, tending to form pellicles on liquids, but are occasionally submerged with modi- fied hyphal growth due to other more influential causes. If other conditions are unfavorable, however, all of the organisms may grow immersed even on a solid substratum. These molds seem to require media rich in both carbohydrates and proteins. The organic nitrogen foods are more easily assimilated than the mineral salts; and sugars better than starches. Glycerine is readily utilized. Some of the present results may be explained as follows: the relative acidity may occasion the submerged growth in Czapek's medium; low carbohydrate supply in nitrate and Gorodkova-Maneval agar necessitates the utilization in the first case of nitrogen of a mineral nature and in the other of an organie nature, with comparable results; utilization of com- [Vor. 23 28 ANNALS OF THE MISSOURI BOTANICAL GARDEN plex plant carbohydrates in situ in corn meal, potato, and car- rot, with increased production especially of coremia; strong aerobic tendencies evidenced by heavy pellicle, with the species of Cephalosporium best able to grow anaerobically if atypi- cally submerged. Although some have regarded the yeast type as a senescent primitive form (Talice, '30), present evidence would seem rather to indicate a reduced type. The yeast has become a highly specialized saprophyte, i. e., adapted to growth in sugar solutions. Such a habitat has effected extreme morphological changes which caused sexual degeneration, and the almost ex- clusive assumption of the budding state due probably to chem- ical and physical causes associated especially with surface ten- sion and the nature of the substratum. Perhaps the status of oxidation-reduction potential in the sugar medium may eluci- date upon the morphological effect. Hewitt (’33) lists carbo- hydrates as having an unknown reversible oxidation-reduction system. The insertion of the yeast stage normally in a filamentous form is interesting too. For example, the smut sporidia propa- gate rather indefinitely by budding. Conidia of the Plecta- scales have also been found upon occasions to bud for various time intervals. Fuchs (’26) gave a review of the early con- cepts which regarded the sprouting interval as a part of the life cycle of filamentous forms. He found that Aspergillus Oryzae, especially on wort and wort gelatine, remained perma- nently in the yeast stage. Such data have not been duplicated to the knowledge of the author. The latter media have been found unfavorable to the molds, probably because of the large amount of sugar, reduced proteins, and also relatively high acidity. In pseudomycelial forms—Monilia in the general sense of the name—the yeast state is prevalent under favorable conditions, and the mycelial one under unfavorable situations according to Henrici (730). C. COLOR The two dominant colors found in this group are green and red in various shades, although one organism was gray and 1936] PINKERTON—CONIDIAL FORMATION 29 three were sporadically yellow. The first pigment seems quite constant under all conditions and is apparently in the mycelial protoplasm as well as the spores. In Gliocladium deliquescens optimal growth shows almost complete covering with a green slime whose color is doubtless due to the numerous spores in- cluded. Gliocladium penicilloides generally is chalky-white, but sometimes becomes pale yellow and on glycerine agar ex- udes the same pigment into the medium. A similar condition is present in Clonostachys Araucaria and its variety rosea on nitrate medium. The pigment is water soluble. It is the various red and orange shades which attract par- ticular attention. The pink pigment of Gliocladium V ermoes- eniis obviously contained within the spores since the mycelium is a white flocculent growth. In the species of Cephalosporium the pigment is dispersed within the mycelium, but its intensity or mere presence is conditioned by varying factors of light, temperature, substratum, oxygen relations, associated prob- ably with general metabolism. A few observations are to be mentioned here, although little experimental work was done upon their nature. Cephalosporium kiliense when growing most luxuriantly in the light is a rich apricot-orange; but as the culture stales, or on media low in accessible nutrients, or in the dark, it fades to a pale yellow-orange or even almost white. C. niveolanosum is a paler pink at best and is white if kept in the dark (at the same temperature) for even a few days. The color of the Torulopsis mucilaginosa is more persistent, but in one case when a culture was placed in the refrigerator (13° C.) for two months it be- came white and remained so on all subsequent transfers. An attempt was made to extract the pigment of Cephalo- sporium kiliense with various solvents. With concentrated sulphuric acid the specimen became blue-green immediately, and the liquid later had a cherry-wine color due perhaps to the dissolution of agar. Concentrated hydrochloric acid turned the solution a blue-green color which faded to yellow. Fuming nitric acid, as well as 20 per cent sodium hydroxide, dissolved the fungus and medium with no color reactions. There was slight solubility in chloroform as well as in hot 95 per cent [Vor. 23 30 ANNALS OF THE MISSOURI BOTANICAL GARDEN alcohol and in lactic acid. Entirely ineffective were petroleum ether, benzol, oxalic acid, salicylic acid, acetone, formaldehyde, and ethyl-ether. The pigment was especially soluble in acetic acid and upon evaporation yielded a yellow-orange mass which appeared amorphous under the microscope. This observation is insufficient to identify the pigment, although it seems related to xanthophyll. The reaction with hydrosulphuric acid was not the typical one of lipocyanin. Color variations have been occasionally noted in fungi, and Zopf (’90) worked out some rather complicated methods of ex- traction and determination. Palmer (’22) stated that in some cases the reds and yellows were doubtless due to carotinoids, whereas in other cases they were not, as in chrysophanic acid. Pigments have also been cursorily noted in the bacteria, and it has been possible to separate rather constant strains, and at other times conspicuous color variants have been noted in a colorless colony or vice versa. Little or nothing is known of the chemistry or physiological purpose of pigments. Some theories have been promulgated: reserve of oily nature in the rust spores (Zopf, '90), a protective function to cellular en- zymes (Went, '04), and an oxidation pigment. It may be that the pigmentation is largely associated with conditions of acid- ity and alkalinity in the medium, as Thom (’26) has suggested, yellow occurring in acid conditions and red in alkaline. In the cases at hand—Cephalosporium niveolanosum and C. kiliense—it is possible that the pigment is a reserve and in- dicates excessive anabolism, but chemical investigation would be necessary to substantiate the view. A protective function seems quite logical whether for enzymes or the cytoplasm itself. The association with respiration seems unnecessary, since existence is possible without pigment in these species and in closely allied ones (e. g., C. rubrobrunneum and C. Stueh- mert, separated almost entirely upon the presence or absence of pigment). At any rate the problem is still an open one and possibly full of significance. According to Henrici ('30), the pigment production in Ac- tinomyces was more constant and also more striking in media of rather low nutrient value, which is the reverse of the present 1936] PINKERTON—CONIDIAL FORMATION 31 findings except for Clonostachys Araucaria, where abnormal pigments are released in the medium. Colley (’31) enumer- ated changes in color of Serratia marcescens and Azotobacter chroococcum upon the addition of certain salts to Bacto-agar. In the latter case the color was supposedly due to the oxidation of tyrosin to melanin, the tyrosin being formed in the breaking down of proteins when the carbohydrate supply was insufficient and unsuitable; but on the addition of zinc salts the color pro- duction was due probably not to the exhaustion of carbohy- drate but to unfavorable conditions preventing its use. The intense yellow water-soluble pigments diffusing into the medium is obviously a metabolic product, perhaps the chrys- ophanic acid of Zopf. In the case of the nitrate medium, at least, it might indicate unfavorable growth, although it doesn’t especially appear so for all media, certainly not with glycerine agar. D. POLARIZED LIGHT Since, as has been stated, the walls of reproductive cells are sometimes of different constitution from vegetative cells and since they might possibly have peculiar physical character- istics, they were studied with a polarizing microscope. Agar- slide cultures and hanging drops were employed under high power and oil immersion. No structural differences were de- tectible by this method, although the twisted nature of the wall in the phialide was most conspicuously brought out. Some- times, in ordinary mycelia, adjacent cells were noted to refract differently. This is presumably due to the different orienta- tion of the cells caused by twisting, but it may have been that some of them were beginning to form chlamydospores and hence were of different structure. V. FUNGI Imperrecti AND SEXUALTY Lindau's (707—710) treatment of the Hyphomycetes has been followed. The subfamily Cephalosporieae is separated from the Aspergilleae in that the spores are in balls in the former and in chains in the latter. The Aspergilleae include Penicil- lium and Gliocladium. Clonostachys belongs in the Verticil- lieae. Allescheria Boydit, also described in different develop- [Vor. 23 32 ANNALS OF THE MISSOURI BOTANICAL GARDEN mental stages as Cephalosporium Boyd and Dendrostilbella Boydu, because of its perithecial fructification, is placed in the Eurotiaceae of the Perisporiales. Clements and Shear (’31) have made Allescheria a synonym of Eurotium. The fertile state of Gliocladium penicilloides placed it in Eurotium insigne according to Winter (787), but this disposition was later dis- puted by Matruchot (795). Thom and some later workers have segregated the perfect forms of Aspergillus, as well as imper- feet forms with similar conidia, in the Aspergillaceae, close to the Gymnoascaceae of the Plectascales. A. EXPERIMENTS TO INDUCE SEXUAL FRUCTIFICATIONS The author at first believed that the imperfect stages were merely labile and that perfect fructifications might possibly be produced by various special methods. In no ease were these successful. After elaborate experimentation with a number of algae and other forms, Czurda (733) came to the conclusion that copulation was caused only by pH and internal disposi- tion, and since the latter was intangible, pH was the only fac- tor subject to investigation. The present experiments did not indicate that there was a specific pH, at least one which was solely responsible. The usual media were employed to evoke fertile stages. Drying or maintaining at low temperatures merely showed the extreme longevity and resistance of these organisms. It was thought that by crossing nearly related spe- cies and varieties fertile stages might be attained. Therefore cross-inoculations were made of Clonostachys Araucaria and its varieties rosea and confusa, as well as of all of the species of Cephalosporium. The only noticeable result was an aversion of Clonostachys Araucaria and the variety confusa. The rest intermingled promiscously. Agar micro cultures, as well as Petri plate specimens, were observed, and although no particu- lar hyphal fusions were noted, neither were there ordinarily signs of repulsion. B. INTERPRETATIONS REGARDING LOSS OF SEXUALITY Ramsbottom (’33) believed that heterothallism and salta- tion were responsible for the origin of the Hyphomycetes. 1936] PINKERTON—CONIDIAL FORMATION 33 Thus theoretically we may regard one of the Fungi Imperfecti as able to carry on the influence of the environment. If this is so it is possible that a permanent change may be brought about. . . . If such changes occur naturally then we may get differences which might throw heterothallism out of gear. It was his suggestion that these should not be regarded as spe- cies but rather as form-species. Hansen and Smith (’32) held similar views on Botrytis cinerea which theoretically has a per- fect stage in Sclerotinia Fuckeliana. They discussed previous explanations as (a) impure material, i. e., a complex of estab- lished types on the host which must be separated (b) muta- tion, (c) hybridization, (d) microconidia—some possible sig- nificance but little evidence that they even have enough power of germination to play an important role, (e) anastomoses, hyphal fusions, heterocaryosis, i. e., condition of nucleus con- taining two or more genetically different nuclei, (f) mixo- chimaera (no cytological evidence of this yet). Their cytolog- ical studies showed that the mycelial cells and conidia were multinucleate. Further analyses of types were made by 128 single spore cultures, from which were selected 8 showing the greatest morphological variation. Twenty-five single spore cultures were made of these and a record kept for a number of generations. Hansen and Smith concluded that the regu- larity and completeness with which the homotypes separated from the heterotypes indicated that the character-determining elements were discrete units of limited number, suggestive that the basic unit of the individual was the nucleus and not the cell. Therefore a multinucleate spore, a colony really, can- not give rise to a genetically pure culture unless all of its nuclei are genetically identical. Gregory (735) suggested the follow- ing explanation for loss of sexuality in the dermatophytes, namely, environmental factors, elimination of sexuality (nega- tive adaptation to parasitism), and heterothallism. Although the species here studied are uninucleate, heter- othallism may well play a part. It is not unlikely that there are many races, new ones constantly being formed by the mech- anism of hyphal anastomoses, and some of these are actually antagonistic although morphologically very similar. Perhaps some compatible strains have died out, leaving imperfect [Vor. 23 34 ANNALS OF THE MISSOURI BOTANICAL GARDEN halves to vegetate ceaselessly unless by saltations new har- monious ones shall arise. VI. MEDICAL Aspects A. PATHOGENICITY Of the forms isolated from skin lesions—Allescheria Boydii and the species of Cephalosporium—there is some doubt as to real pathogenicity. Of the Cephalosporium species treated by Dodge (’35), only C. Serrae, of keratomycosis of the eye, and C. Doukourei, of a gummatous lesion, are said to be pathogenic to laboratory animals. Allescheria Boyd, isolated by Boyd and reported by Shear from a ease of granular mycetoma of man, was thought by them to be a contaminant and not an anaerobic form because of its peculiar morphology within the foot and its inability to in- fect laboratory animals. A number of other Hyphomycetes, including species of Indiella, Glenospora, Scedosporium, As- pergillus, and Penicillium, have been reported as etiological agents of maduromycosis. At present it would seem that such were purely accidental contaminations, which may be fatal but are not naturally so. The author secured species of Penicillium and Aspergillus frequently from eczematoid infections at the Barnard Free Skin and Cancer Hospital, and the Cephalosporium symbio- ticum, with the intimately associated Torulopsis mucilaginosa, from a particularly severe case of dermatomycosis. She also isolated the latter with a similar co-form which was eliminated in subsequent dilution cultures from a case of pemphigus vul- garis. Apparently the same red yeast was found by Engel- hardt (727) in a case of pemphigus. He illustrated the organ- ism as it grew in the host, i. e., filamentously, similar to the condition here found on nitrate agar, although no cephalo- sporia were shown. Whether such organisms are purely sap- rophytie secondary or primary invaders, or necessary sym- bionts of forms better believed to be pathogenie (Hartmann, '26), in the present case of a Cephalosporium and Torulopsis mucilaginosa, is as yet unknown; but immediate evidence at 1936] PINKERTON—CONIDIAL FORMATION 35 least indicates that they have rather specialized habitats, and constitute a group which should be investigated from a patho- genic viewpoint. The various species of Cephalosporium have been regarded as accessory perhaps symbiotic saprophytes. Hartmann (’26) found C. niveolanosum associated with Trichophyton gypseum and left the question open. So great a number have been se- cured from superficial lesions that although animal inocula- tions may be negative (perhaps due to methods of inoculation) a parasitic nature is certainly suggested. The author has found species of Cephalosporium from superficial lesions of the glabrous skin especially. Various species of Penicillium and Aspergillus are even more common. Alternaria, Sarcina lutea, and Torulopsis mucilaginosa are likewise frequently found. Some species of Cephalosporium are also reported as saprophytic in nature on soil or humus. None of these were studied; but an organism treated by Buchanan (’11) as Cepha- losporium Pammelii appears different from the skin group, at least in the possession of multiseptate conidia, whereas only uniseptate ones are found in the group here investigated. Spe- cies of Cephalosporium have also been found on insects (Petch, '28) and as causative agents in various plant diseases. In the Dutch elm disease it is only a stage of an organism, of which the perfect stage is Ceratostomella Ulmi of the Sphaeriales. Reddy and Holbert (’24) gave a good description of the effects of C. Acremonium Corda emended Fresenius, upon corn, caus- ing the black bundle disease. In all these cases it is the trans- porting system which is attacked. The habitat would certainly indicate parasitic tendencies which might as well be developed in animal tissues as in those of the plant. B. OCCURRENCE OF GRANULES IN CULTURE One strange cultural condition certainly deserves further mention. On blood serum C. rubrobrunneum, after several weeks, produced numerous small white spherical bodies several mm. in diameter. Under low power they appeared surrounded by a fine felt of white to pale pink mycelium (pl. 6, figs. 15-16). These proved to be very hard and when crushed revealed a [Vor. 23 36 ANNALS OF THE MISSOURI BOTANICAL GARDEN white chalky interior. Under higher power numerous linear crystals radially arranged were seen. These crystals did not stain with lacto-phenol and cotton blue, nor with glycerine- eosin, and there was no definite cellular mass visible, although there were some irregularly shaped stainable masses. The latter were suggestive of the granular bodies occurring some- times in the animal body, in fact quite definitely characteristic of certain diseases. Of late such have been noted more fre- quently upon artificial culture (Area Leão and Lobo, ’34). Two recent papers (Almeida, '34, and Weidman, ’32) gave rather fully the history of granular occurrences. These varied in the host from the capsules of bacteria, Saccharomyces hominis and Cryptococcus histolyticus, to the radiating struc- tures of purely membranal or cellular nature, as in Aspergil- lus, and the definite extra-cellular grains produced in Actino- myces, Madura foot, and the like. They have been interpreted as a reaction to the host comparable to distortions of haustoria, as living structures capable of absorbing food, ete., or as mere dead extensions of the membrane. They are compared to the ‘*Hiille’’ cells first mentioned by Eidam ('83) in species of Aspergillus—all strains of the A. nidulans group and some strains of A. terreus, A. flavipes, and A. ustus. These cells de- velop from older mycelia as smaller branches, and the terminal and subterminal cells become surrounded by a very thick muci- laginous wall, comparable to that of Saccharomyces hominis produced in the host. The other structures have been evoked upon organic media frequently—a fact which eliminates a vital cause and probably puts such formations on a chemical basis. Weidman (732) compared these incrustations to the similar mechanism in the ‘‘Dauercyst’’ of protozoa, i. e., that the cap- sule (especially) is protective against adversity and prepara- tory to reproduction, as likewise the asci in fungi are reproduc- tive cells whose development is stimulated by unfavorable con- ditions. He does soften this rather far-fetched homology of reproduction and adverse conditions (although such was prob- ably the primitive state) by saying that possibly reproduetion could be absolutely separated from environmental factors. He thought that adverse conditions might also be assumed for 1936] PINKERTON—CONIDIAL FORMATION 37 fungi in the comparatively anaerobic and warm environment of the mammalian tissue. Incrustations of the type found by the author are extra- matrical, of the ‘‘caleareous incrusting substance’’ type in Weidman’s classification. Their origin and purpose are poorly comprehended, but they are probably reminiscent of pathological conditions in the host. Various solvents (concen- trated acids of sulphuric, nitric, and hydrochloric) were tried on these crystalline masses which it was conjectured might be calcium oxalate. The most rapid action occurred in hydro- ehlorie acid, releasing numerous square-oblong flat crystals and some loose cubical arrangements of about sixteen small ones; and the remainder of the ball appeared segmented with fine long crystals which protruded radially. In nitric acid there was much bubbling, and the balls assumed a rougher appearance under the microscope but dissolution was very slow. The bubbles were colorless, of a peculiar hexagonal outline, and their heavy part contained small crystalline frag- ments. In sulphuric acid there were crystals similar to those in the hydrochloric acid, and some hollow ones. It could not be noted if the ball were actually diminishing in mass, but the liquid around it seemed of a deeper yellow than the normal acid, perhaps indicating solution. VII. Taxonomy AND MonPHOLOGY The general classification of this group has been given in a previous section, and only the individual members will be dis- cussed here. Due to the temporary nature of this group, much changing of nomenclature is both unnecessary and confusing. A couple of very obvious errors are corrected, however, i. e., in regard to the former Gliocladium roseum and Clonostachys spectabilis. The new organism isolated by the author is de- scribed as Cephalosporium symbioticum. | Descriptions are based mainly upon giant colonies grown upon glycerine agar. The names of the colors are according to Ridgway (712). ALLESCHERIA Sacc. & Syd. in Saec. Syll. Fung. 14: 464. 1899. It was thought that if this form were included with a known [Vor. 23 38 ANNALS OF THE MISSOURI BOTANICAL GARDEN fertile stage more light might be shed upon the significance of the eonidiophore. Unfortunately, the entire life cycle was not recalled and a complete cytological study was impossible. The conidiophores are poorly developed and resemble some- what those of Cephalosporium; but the dimorphism of conidia is a constant feature, and the soft downy colony is quite distinct from the smooth rope-like one of Cephalosporium; so that it is doubtful if even the spore-ball stage should be included in that genus. Type: Eurotiopsis Gayoni Costantin. 1, AuLEscHERIA Bovpmr Shear apud Boyd and Crutchfield, Am. Jour. Trop. Med. 1: 258-268. 1991; Mycologia 14: 239- 243. 1922. Pl. 1, figs. 18-28. Colony: pale gray velvety, of short filaments, sometimes be- coming greenish in spots, or with much longer pile and ap- proaehing white, cerebriform and growing into medium at edges; reverse slightly hardened and wrinkled; growth rela- tively slow. Conidiophores: simple; phialides on aerial mycelia, 3-20 p long, septate or not, cutting off oblong conidia which adhere irregularly ; spores with two oil droplets, ovoid, capable of im- mediate germination, 8 x 5.5 », heavy-walled spores occasion- ally preceding hyaline ones, the former characteristic of single- spored phialides of coremia, 15.5 x 4 »; terminal chlamydo- spores 8.5 x 11 ». Habitat: isolated from a case of granular mycetoma in a negro. Culture of Shear, from Baarn. CEPHALOSPORIUM Corda, Icon. Fung. 3: 11. 1839. Colony: smooth, much corrugated, rope-like, forming a con- sistent layer, slightly adherent. Conidiophores: simple or with few short lateral branches along cords of hyphae; hyphae repent, phialides long and slender, only slightly narrower at tips than at base, bearing spores in slime balls; coremia frequent on certain media; chlamydospores present. Type: C. Acremonium Corda. The six species here studied form quite a homogenous group. Cephalosporium symbioticum and C. Serrae vary most widely, 1936] PINKERTON—CONIDIAL FORMATION 39 the former including yeast-like growth and becoming fuscous with chlamydospores, the latter becoming pubescent and also dark with chlamydospores. KEY TO SPECIES OF CEPHALOSPORIUM 1. Producing some yeast-like growth on solid culture, apricot-buff, becoming eil er EE C. symbioticum 1. Not found to produce yeast-like growth in culture..........oo.o.ooooomo.m.... 2 2. Colony white with a rough dry surface, sane becoming fuscous due to chlamydospores.....o.oooocoommmoorosorsrrorsscrrrcrnnanan»s C. Serge 2. Colony not becoming ÍUScOUS..........oo<. «ooo oooocororcoscrossrscian.s 3 3. Colony moist, rope-like, turbinate.......... eee 4 3. Colony dry, often with a bloom, irregularly dissected............. eee 5 4. Colony ivory to pale pink; conidiophores simple; spores 6 x 3 uw. .C. niveolanosum 4. Colony soon becoming deep orange; eonidiophores simple or branched; &porea 2-4 x LB Rate EE d nda A ok EAR ee ES C. kiliense 5. Surface of colony smooth, cerebriform, seashell pink; spores 7 x 1.5 e Perr EEN EE e ae C. rubrobrunneum 5 . Surface of colony short-pubescent, slightly convolute, white to salmon- buff; spores 4—5.5 x bänn EE EE S C. Stuehmeri 1. CEPHALOSPORIUM RUBROBRUNNEUM Nannizzi, Tratt. Micopat. Umana [Pollaeci] 4: 455. 1934. Pl. 2, figs. 22-25. C. rubrobrunneum cerebriforme Hartmanni Benedek, Arch. f. Derm. & Syphilis 154: 166. 1928. Colony: at first an irregular small dry pink heap in center, spreading irregularly for about a cm., then flat, about 1 mm. thick, moderately dry, whitish pubescent at edges, center of coremia becoming frosted, seashell pink, very cerebriform, edges abrupt; little guttational water; reverse corrugated and consistent, pale orange, little color diffusing into medium; no coremial balls. Conidiophores: borne on ropes of numerous hyphae up to 11 „in diam., surrounded by thick mucus, spreading from both sides, non-septate, up to 30 x 2-3 », producing spores in balls 5-15 y in diam., or irregularly clinging along sides of phialides in packets; spores oblong-ellipsoid, 7 x 1.5 », or almost spher- ical, 2-2.5 », sometimes slipping past each other at only a slight angle and forming a column; terminal chlamydospores on wort agar, 17 x 14 y, globular crystalline bodies on serum (see pl. 6, fig. 9, and discussion p. 35). [Vor. 23 40 ANNALS OF THE MISSOURI BOTANICAL GARDEN Habitat: isolated from a case of dermatitis in Frankfort, a.M., Germany. Culture of Pollacci, from Baarn. 2. CEPHALOSPORIUM STUEHMERI Schmidt & Beyma in Beyma, Centralbl. f. Bakt. I, 130: 102-105, 3 figs. 1933. Pl. 2, figs. 17-21. Colony : white to a pale pink, slightly pubescent above, little aerial growth, small white balls 1 to several mm. in diam. aris- ing on elevated spots of colony in eultures 7-10 days old, about as thick as heavy paper; guttational water; reverse corrugated, pale orange, as is also adjacent medium; small coremia on Sabouraud's medium, and on carrot plugs a white pustular tallowy growth. Conidiophores: unbranched, usually one to several septate, borne in a row along both sides of a straight hypha ; sheaths of hyphae often with some strands protruding and bearing spores; phialides 35 x 1.5 »; cephalosporia 10-15 y; conidia ellipsoid, 4-5.5 x 2.5-3 y, swelling slightly on germination, sprouting at one point and giving a mycelium that becomes early septate ; vegetative cells 10 x 2 y, uninucleate ; submerged terminal chlamydospores 6.5-7 » in diam., intercalary ones 7 p. Habitat: eczematoid infection of skin, Germany. Culture of Stuehmer, Baarn. 3. CEPHALOSPORIUM KILIENSE? (Gruetz) Hartmann, Derm. Woch. 82: 569. 1926. Pl. 2, figs. 26-31. Acremonium kiliense Gruetz, Derm. Woch. 80: 765-774. 1925. C. asteroides griseum Gruetzii Benedek, Arch. f. Derm. & Syphilis 154: 166. 1998. C. Acremonium Pollaeci & Nannizzi, I Miceti Pat. Uomo Anim. 9: No. 81. 1930. Non Corda, 1839. Colony: a few mm. thick, bittersweet-orange, finely corru- gated, consisting of a number of small irregularly arranged strands, aerial growth slight, flat, and parallel; mycelium ex- tending up on glass and consisting mainly of relatively thick cords upon which tiny cephalosporia can be seen with a lens; * See Dodge, Med. Myc. p. 828. 1935. 1936] PINKERTON—CONIDIAL FORMATION 41 much guttational water; reverse slightly irregular, lighter- colored than upper surface; on Sabouraud’s dextrose agar a faint white surface powder; on carrot plug an orange hairy coremial growth. Conidiophores: along both sides of main stalk, especially clustered at ends and upon lateral branches where compound phialides are often borne in whorls of four or more, forming cephalosporia up to 14 » in diam. in the air, or irregular non- slimy packets ; phialides slightly tapering, 243 x 2.5 »; conidia oblong-ellipsoid, hyaline, 2-4 x 1.5 »; terminal chlamydospores 4-7 x 2-5 p», intercalary ones 5 y in diam. Habitat: gummata and ulcers in man. Culture of Pollacci (?), from Baarn. 4. CEPHALOSPORIUM SERRAE Maffei, Atti Ist. Bot. R. Univ. Pavia IV, 1: 183-198, 9 figs. 1930. Pl. 2, figs. 1-11. Colony : about 5 mm. thick, corrugated, at first white then sea- shell pink, in about ten days turning fuscous, aerial growth rough and eoarse wooly, of short projections several mm. high, heaped, dry and partially wrinkled; a bit of fine growth on glass; no guttational water; reverse wrinkled and black; no eolor diffusing into medium ; on earrot plug an olive-buff pus- tular glabrous growth with a few pink coremia. Conidiophores: borne laterally, simple or once-seriate, of 3-5 phialides ; main stalk 42 » long; phialides somewhat taper- ing, 29 x 2.5 »; cephalosporia 7.5 » in diam. ; conidia ellipsoid, 3-4 x 1.5 n, sometimes not in a ball; chlamydospores 7.5 x 5 p and 10 x 3.5 », terminally or on strands of upright coremia. Habitat: keratomycosis of the eye. Culture of Pollacci, from Baarn. 5. CEPHALOSPORIUM NIVEOLANOSUM Benedek, Arch. f. Derm. & Syphilis 154: 166. 1928. Pl. 2, figs. 12-16. Colony: at first elevated, ivory-white, pasty, 5 mm. in diam., becoming a slightly pink, glistening, dry heap, edges flat with radial striations arranged turbinately (in the dark remaining ivory-white but becoming pink upon exposure to daylight) ; short delicate coremia at times; reverse convex from medium. [Vor. 23 42 ANNALS OF THE MISSOURI BOTANICAL GARDEN Conidiophores: simple, rare, septate or not; spores ellipsoid with two oil droplets, 6 x 3 ; hyphal cells 20 x 2.5 un: intercalary submerged chlamydospores 6.5 » in diam., terminal thinner- walled ones, 15 y. Habitat: isolated from a case of dermatitis in Leipzig. Cul- ture of Benedek, from Baarn. 6. CEPHALOSPORIUM symbioticum‘ Pinkerton, n. sp. Pl. 2, figs. 32-38, 45-46. Colony: apricot-buff, on wort agar sometimes becoming drab-green, with smooth surface except for small coremia, con- voluted, growing into medium below; on liquid media produc- ing a ring or pellicle, decolorizing litmus, coagulating milk, liquefying gelatine, no gas in sugars, producing acid in dex- trose, d-levulose, raffinose, maltose, and sucrose, alkali in amygdalin, arabinose, d-galactose, lactose, and dextrin. Within medium colony forms pseudomycelium of cells 4.0 x 2.5 n, cephalosporia 8 » in diam. at the apex of phialides, the latter 5 » in length; aerial thallus a true mycelium, cells 8-10 x 2 1; phialides 2-5 » in diam. ; spores 4-5 x 2.5 », some germinat- ing in place. Mature colonies becoming fuscous with many terminal and intercalary chlamydospores 4-5 u in diam. Habitat: isolated from a severe case of dermatitis at the Barnard Free Skin and Cancer Hospital. It was intimately associated with Torulopsis mucilaginosa from which it was separated only by extreme dilution cultures. The subgenus Cephaloblaston of Ciferri (732), typified by C. pseudofer- mentwm, would also encompass this species. * CEPHALOSPORIUM symbioticum, sp. n Colonia: ‘‘apricot-buff’’ aliquando ‘‘ pem -green,’’ in medio musto, superficie aevi praeter coremia parvo, convoluta, in media subter succrescens; in mediis liquidis annulum pellieulumve efficiens, litmo decolorans, lactem coagulans, gela- tinam liquefaciens, in saccharis non effervescans; acidum in ''dextros e, d- cen raffinose, maltose, sucrose?" faciens, alcalinum in ‘‘amygdalin, arabinose, d-galac- tose, lactose, dextrin. 77 olonia mycelium falsum intra medium formans, cellulis 4.0 x 2.5 By BEE 8 w in diametro apice phialae 5 y in longitudine; thallus aerius mycelium j cellulis 8-10 x 2 y; phialae 2-5 y in diametro; sporae 4-5 x 2.5 u, aliquibus in in germinantibus. In coloniis maturis fuscescentibus ehlamydosporae multae ter- minales aut intercalares, 4-5 y in diametro. 1936] PINKERTON—CONIDIAL FORMATION 43 Case History.—The patient, a white female 15 years of age, first reported to the Barnard Free Skin and Cancer Hospital on April 2, 1934. She had a severe dermatitie eruption occurring in patches upon her hands, especially the fingers, and upon the forearms. The case was diagnosed as 77 dermatitis contact.”? Trichophytin tests were given and erude eoaltar salves advised. Various patent medicines previously used by the patient eomplieated the irritation. By August 31, 1934, there was a marked spa but on September 20, the condition was worse and X-ray treatments were give On Oetober 2, the hands showed many new vesicles and bullae and ruptured —— which left raw spots. The condition was then diagnosed as a combination of tinea and Staphylococcus infection. She also suffered from acne and reported a history of scabies. In addition to x. -ray treat- ments, a tincture of xeroform was recommended. The patient was last heard from on October 24, 1934. Skin cultures included numerous organisms of which Staphylococcus, Torulopsis mucilaginosa, Cephalosporiwm symbioticum, and a black filamentous form (un- identified) were dominant. The Cephalosporium was non-pathogenic to white mice. Cionostacuys® Corda, Pracht-Flora, p. 31. 1839. Creeping hyphae; conidiophores simple or branched; phialides four or more branched, verticillate; spores adhering in cephalosporia. Type: C. Araucaria Corda. 1. CLowosrAcHys Araucaria Corda, Pracht-Flora, p. 31. Pl. 5, figs. 14-26. eege white-cottony, soon turning pale flesh to seashell pink, powdery, rapidly growing; coarse white-pink coremia in older cultures; mycelium extending up glass for several mm., bearing cephalosporia; much guttational water; reverse wrin- kled. Species differs superficially from the former Gliocla- dium roseum in pinker color, slower growth, less extensive coremia formation, and producing acid in maltose. Conidiophores: polymorphic, some of the characteristic of four-branched type; asymmetrical penicillate fructifications 5 The investigation of the type species and varieties permits enlargement upon od s original limited deseription. Unfortunately, the author is unfamiliar with the other species usually included here, but because of the limited generic naa it is doubtful if they form a homogeneous group. The type, which has been inadequately characterized, does seem sufficiently different to justify its posi- tion in a genus intermediate between Cephalosporium and Gliocladium. Clono- stachys spectabilis, which was studied only briefly due to an early demise, peguen is far removed and certainly suggests Botrytis. Presumably Oudemans ('86) e pared the botryoid spores to the lateral protrusions sometimes found on C. prea and its varieties, which are not spores [Vor. 23 44 ANNALS OF THE MISSOURI BOTANICAL GARDEN formed on reversed wort plates with terminal clusters of six phialides and laterals of three to six, phialides 10 x 2 p, flask- shaped and bent outwards, conidia spherical to ellipsoid, 3 x 1-2 »; cephalosporia borne along the main cord, singly or in clusters of three to six, phialides slightly flask-shaped, 17-33 x 1.54 »; cephalosporia 13 y, conidia 5.5-6.5 x 2.5-3 »; hyphae warty, especially near cephalosporia; chlamydospores rare, on corn meal agar in terminal series, 13 x 10 »; or on lactose broth interealary, 10 x 12.5 »; hyphal cells 13 x 3 p. Habitat: **In cortice ramulorum arborum in Germania, Brittania”” (Corda). Culture of Wakefield, from Baarn. 2. CLONOSTACHYS ARAUCARIA Var. ROSEA Preuss, Linnaea 25: 121. 1852. Pl. 4, figs. 27-31. Colony: salmon-pink almost at first, mottled with white, floccose, of strands several mm. long, heaped in center and irregularly wrinkled, coarsely flocculent at edges; much gutta- tional water; reverse wrinkled. Variety differs from the for- mer Gliocladium roseum and the species in the salmon tint, intermediate speed of growth, and is more wrinkled than either. Conidiophores: mostly typical of four phialides as illus- trated by Corda, the spores from several phialides sometimes forming confluent slime balls; cephalophores up to 100 »; phialides non-septate, 17-40 x Dn: cephalosporia 21 » in diam. ; conidia 4 x 5 y; chlamydospores terminal and intercalary, 3.5 x 10 p. Habitat: ‘‘In ramulis, in piris et pomis immaturis dejectis, frequens, Hoyerswerda”” (Preuss). Culture of Shapovalov, from Baarn. 3. CLONOSTACHYS ÁRAUCARIA var. confusa Pinkerton, n. var. Pl. 4, figs. 15-26. Penicillium roseum Link (?), Ges. Naturforsch. Fr. Berlin, Mag. 7: 37. 1816. Gliocladium roseum Bainier, Bull. Soc. Myc. France 23: 111-112, pl. XV, figs. 1-6. 1907. Nec Clonostachys Araucaria var. rosea Preuss. : Colony: pale ochraceous salmon, slightly mottled with 1936] PIN KERTON—CONIDIAL FORMATION 45 white, short-cottony and mealy on top, growing up tube very little, slightly adherent; on inverted Petri plate of wort agar producing white feathery extensions measuring up to 1 cm. long; much guttational water; reverse much wrinkled. Conidiophores: polymorphic; penicillate conidiophores asymmetrical with branches terminating in whorls of three to four, penicillus 85 » high, main lateral segments 45 p», phialides 8-10 x 1.5-2 y, flask-shaped, mature conidia spherical to slightly falcate, 5 », swelling on germination to 8-10 », growing from one or both poles; spore balls from single phialide along main hyphae or rope, or usually in whorls of three, borne lat- erally or terminally, phialides 30-40 x 3-4 », warty, straight or slightly narrower at top, adjacent balls sometimes becoming confluent and reaching 10-30 » in diam., conidia ellipsoid, smooth, hyaline, pink in mass, 6-7 x 4-5.5 »; chlamydospores rarely formed but found terminally on submerged hyphae on Raulin’s agar, 4.5 » in diam., also on inverted wort plates, inter- calary ones 14 x 20 y. Habitat: *In caule solani tuberosi exsiecati. Misit Dr. Nees ab Esenbeck.”” (Link); ‘‘du carton pourri” (Bainier). Culture from Thom No. 454—4640.428, from Kral in Prague, Bohemia. Bainier stated that this form resembled Acrostalagmus roseus of the Verticillieae except that the precedence of penicil- late over verticillate series was reversed in Gliocladiwm. More probably the variations are dependent upon nutrient relations, especially moisture conditions. On lactose broth hanging drops, the spore balls are most luxuriant, as are also the warts. Complex penicillate fructifications are also present, however, and in fact precede the simpler cephalosporial stage both in time and on the main trunk. The spores on the penicillus cling together only loosely in irregular chains as seen on slide cul- tures, and are quickly disseminated in water. Conspicuous deliquescence of the penicillate heads, as occurs in G. deli- quescens, was not noted. The fact that in crossing this form is antagonistic to the species, whereas the rest are all indif- ferent, might indicate a close relationship of a lethal nature. [Vor. 23 46 ANNALS OF THE MISSOURI BOTANICAL GARDEN 4. CLONOSTACHYS SPECTABILIS (Harz) Oudem. € Sace. Ned. Kruidk. Arch. IT, 4: 539. 1886. Pl. 5, figs. 1-7. Botrytis spectabilis Harz, Bull. Soc. Imp. Nat. Moscow 44: 88-147. 1871. Original colony gray-white and fuzzy, turning black due to production of numerous conidia. Conidiophores compound, dichotomously branched and bear- ing spores on short stalks laterally along the phialides; spores spherical, 6. au » in diam. ; chlamydospores 6.5-10 » in diam. Habitat: ‘‘sur la RA dans les serres du jardin botanique d' a (Balen). Culture of Shear, from Baarn. GurocLADrUM* Corda, Icon. Fung. 4: 30-31. 1840. Conidiophores erect, septate, penicillately branching above, branches and branchlets septate, appressed, forming a solitary gelatinous head; conidia unicellular, borne upon the tips of branchlets and held together by mucilaginous substance in a dense mass, those on adjacent phialides fusing. Type: G. peni- cilloides Corda. l. GLIOCLADIUM FIMBRIATUM Gilman E Abbott, Iowa State Coll. Jour. Sci. 1: 304, fig. 38. 1927. Pl. 3, figs. 33-41. Colony : at first white, fuzzy, nearly circular, very thin, moist, translucent with parallel wrinkles, finally antique green, mot- tled with white, dry, floccose, faintly zonate and radially striate, broad masses of white hyphae growing up tube several em.; reverse radially corrugate. * The inclusion of the former Gliocladium rosewm in Clonostachys might possibly also justify lumping Gliocladiwm in that genus Whose publication preceded it by one year. However, the three sections of Gliocladium mentioned by Thom (730)— the rose, white, and green series—seem quite disti: eulturally and iis log- ically. There is one common character to the two genera, i. e., the lack of dis- junctors and presence of cephalosporia, which in view of the Goen mass zonate growth and exudation of yellow color into medium, but differs in the walled auxiliary cells (akinetes). G. Vermoeseni is quite distinct in its colonial characters and the spore type which approaches Penicilliwm in its rigidness, but still lacks disjunctors. G. deliquescens and G. fimbriatum are very different from eac other as well as the others in colonial characters but have akinetes. Therefore, due to insufficient knowledge of the Gliocladium and Clonostachys segregates as a whole, they will not be combined here 1936] PINKERTON—CONIDIAL FORMATION 47 Conidiophores : homogeneous, borne laterally and terminally in a wide range of heights up to 300 », more or less symmetrical in two series, basal cell 14.5 x 5.3 »; first series 6.5 x 4 nr phi- alides very turgid and asymmetrically flask-shaped, 8.5 x 4.5 u; spores ellipsoid, 4.5 x 3.3 », forming irregular spore balls up to 16.5 » in diameter, dry and only slightly mucoid; akinetes ter- minal and intercalary, frequent, 11-13 x 8.5—10 y, containing oil droplets up to 1.3 », these soluble in chloroform leaving stria- tions ; chlamydospores intercalary, 9-9.5 x 8 p. Habitat: in soil, Iowa and Louisiana. Culture of Gilman & Abbott, from Thom No. 459—4894.20. 2. GLIOCLADIUM DELIQUESCENS Sopp, Monogr. pp. 89-93, taf. I, fig. 1-6. 1912. Pl. 3, figs. 20-32. Colony: at first colorless, of mainly submerged hyphae, quickly becoming thin and white-floccose over whole surface, then meadow-green and slimy, progressing slowly centrip- etally from edge; much guttational water; very rapid growth of hyaline semi-submerged radiating hyphae. Conidiophores: of two types—penicillate and simple ceph- alosporial; penicilli at first white but spores become green and aggregate in a heavy slime, typically tri-seriate and each series trichotomous; brush 45 x 40 y; first series 17.5-24 x 4 p; sec- ond series 9-11 x 4 y; phialides 6-9 x 2-2.5 m, spores smooth, spherical to ellipsoid, 3.5 x 2.5 », germinating spores swelling up to 6.5-9 p, sprouting in one or more places; spore balls up to 30 »; cephalosporia up to 60 pn, circular with thick matrix of slime, cephalolides 40 x 6 y; conidia 2 » in diam. ; akinetes 7.5-10 x 6.5—8.5 », terminal or intercalary ; pseudoterminal ovoid cells 14 x 11.5 »; chlamydospores 6 x 7.5 y. Habitat: on Daedalea unicolor in Norway. Culture of Thom No. 457-4894.17. 3. GLIOCLADIUM VERMOESENI (Biourge) Thom, The Peni- cillia, p. 502. 1930. Pl. 4, figs. 1-14. Penicillium V ermoeseni Biourge, Là Cellule 33: 230. 1923. Colony: pale vinaceous, rapidly growing, irregularly pow- dery and with thin long hairs; much guttational water ; reverse homogeneous and when viewed through appears splotchy. [VoL. 23 48 ANNALS OF THE MISSOURI BOTANICAL GARDEN Conidiophores: usually penicillate, 45 » long from first branch, decidedly asymmetric, poly-seriate, final series trichot- omous; phialides up to 10 x 1.5-3.5 »; spores spherical, 2.5 a; cephalolides 40 y and more, cephalosporia 20-25 y», cephalo- spores ellipsoid, 5-6.5 x 4 »; intercalary chlamydospores 13 x 15.5 p, terminal 5 x 11 »; akinetes 15 » in diam.; arthrospores 10.5 x 5 »; oidia 3 x 5.5 m, rare. Habitat: parasitic or semi-parasitie on certain species of Areca. Culture of Westerdijk, from Thom No. 451-5090.4a. 4. GLIOCLADIUM PENICILLOIDES? Corda, Icon. Fung. 4: 31, taf. 7. figs. 88-89. 1840. Pl. 3, figs. 1-19. Colony: at first chalky-white, of short regularly centrif- ugally appressed hairs, faintly and scarcely striate, within a month very dry and tan tolight pink. On glycerine agar some- times producing a lemon-yellow pigment which diffuses into the medium ; growth relatively slow and consistent. Conidiophores: asymmetrie, often borne as more or less short laterals along main branch, up to 114 » long, septate; phialides 17 x 3 », slightly flask-shaped, borne in threes, occa- sionally fours or more, of which one is larger; lowest branches alternate, usually not more than three main divisions; spores ellipsoid, 5.5-6.5 x 3 », swelling before germination to 8 x Du: single branched phialides, septate or not, 35.5 x 2.5 y, bearing cephalosporia 14 » in diam.; spores 4 x 2.5 »; vegetative cells 3.5-5 » in diam., racquet-shaped, consisting of several large vacuoles ; SE 10 y in diam., full of oily reserves and ca- pable of almost immediate ee Habitat: on rotting Thelephora. Culture of Thom No. 453- 5113. ScoPuLaniorsis Bainier, Bull. Soc. Myc. France 23: 98. 1907. Type: Penicillium brevicaule Saccardo. "Winter (’87) suggested that probably Eurotium insigne Winter was the ascosporic form of this species. Matruchot (795) found perithecia of the Plectascales type, but through the unique method of ascospore germination dis- tinguished it from E. insigne; however, he offered no other nomenelatorial sugges- tion, Theoretically, the species should be removed to the perfect classification, but for the present such shall not be attempted. 1936] PINKERTON—CONIDIAL FORMATION 49 1. ScoPULARIOPSIS BREVICAULIS (Sacc.) Bainier, Bull. Soc. Myc. France 23: 99-103, pl. 11, figs. 1-6. 1907. Pl. 5, figs. 8-13. Penicillium brevicaule Saec. Fung. Ital. taf. 893. 1881. White flocculent growth upon glycerine agar. Conidiophores simple or few-branched; phialides 8 x 2-3 a; spores 5 „in diam. with disjunetors between, adhering in chains indefinitely ; no mucus ; no chlamydospores noted. Habitat: one strain reported as pathogenic by Dr. D. J. Davis of Chieago. Culture of Thom No. 4858. This species was studied only very superficially as living ma- terial, in an attempt to correlate the presence of disjunctors and no slime with the lack of disjunctors and mucus, also to note methods of phialide mechanism. ToruLorPsis Berlese, Giorn. Viticoltura ed Enologia p. 54. 1894. Type: T. rosea Berlese. TorULOPSIS MUCILAGINOSA (Jörgensen) Ciferri & Redaelli, Atti Bot. Ist. R. Univ. Pavia IIT, 2: 256. 1925. Pl. 2, figs. 39-44, 47. Torula mucilaginosa Jórgensen, Die Mikroorg. d. Gár- ungsindustrie. ed. 5, 402. 1909. Rhodotorula mucilaginosa Harrison, Trans. Roy. Soc. Canada 22: 187. 1928. Cryptococcus rubrorugosus Castellani, Arch. of Derm. & Syphilol. 16: 403. 1927. Rhodotorula mucilaginosa race rubrorugosa Lodder, An- askosporogenen Hefen 1: 104—105. 1934. Colony: Carnelian red to scarlet, moist and smooth, pasty, rapidly growing upon wort and maltose agar, with elliptical cells showing bipolar budding and a pair of oil droplets in ma- ture cells; on nitrate medium, following the sparse red surface growth, a submerged irregular hyaline one of pseudomycelial cells and small asymmetrical cephalosporia up to 15 » some- times occurring, spores 2.5 Dn: in malt broth a ring, some red basal proliferation and slight pseudomycelial growth through- out liquid; slight growth in ethyl alcohol; ring in grape juice but no fermentation, no liquefaction of gelatin, milk clotted ; no [Vor. 23 50 ANNALS OF THE MISSOURI BOTANICAL GARDEN gas in sugar solutions; acid produced in sucrose, d-levulose, maltose and dextrose, alkali in arabinose, amygdalin, d-galac- tose, lactose, raffinose and dextrin. Habitat: originally isolated as a laboratory contaminant; Castellani reported it from scrapings in the axilla; and it was secured by the author from skin scrapings in a severe dermatomycosis. VIII. Summary 1. The term cephalophore has been applied to a conidial fructification wherein the spores are held in a spherical mass by mucilage. These spores are found to be produced endog- enously by cleavage following exudation of liquid which forms the matrix of the spore ball. Terminal granules seem to be kinetic centers associated with exudation and spore discharge. The phialides are thought to be uninucleate as are also the spores. 2. Accessory spore forms include chlamydospores in all cases, associated both with the aerial and submerged mycelium, oidia infrequently, and thin-walled spores in some, which have been called akinetes after the algological terminology. 3. An attempt has been made to correlate comparative cul- tural data regarding color, macroscopic and microscopic char- acters. It is believed that such factors as hydrogen-ion con- centration, surface tension, oxygenation, and the nature of the nutrients are especially important in producing favorable or unfavorable conditions for the organism. Good growth is characterized by abundance of aerial mycelia with a consider- able pigmentation; and poor growth, by numerous spore- forms, attenuated aerial hyphae with proportionally more re- productive bodies and a tendency towards submerged growth and exudation of pigments into the medium, and a paler color of the organism. Torulopsis mucilaginosa develops a pseudo- mycelium in the submerged state and loses its red color. 4. Mycelial pigmentation is noted in some cases to be more intense under optimal conditions; in some to be light labile. The solubilities of the pigment of Cephalosporium kiliense are listed, although no identification is made. 1936] PINKERTON—CONIDIAL FORMATION 51 5. Polarized light indicates no structural difference between the walls of the phialides and of the mycelial cells. 6. Various unsuccessful attempts have been made to pro- duce sexual fructifications. Theories regarding loss of sexu- ality are discussed, but no further contributions are suggested. 7. The clinical findings of some of the organisms in dermato- mycoses are mentioned, although their pathogenicity has not been proven. The granular incrustations upon serum are de- scribed. These were found to be composed of fine needle-like crystals arranged radially and including no recognizable or- ganic matrix but surrounded by fine mycelial filaments. The crystals were slowly soluble in the concentrated heavy acids. The theoretical significances of these structures are discussed. 8. A brief morphological description is given of all the forms studied. Gliocladium roseum (Link?) Bainier is described as Clonostachys Araucaria var. confusa. The position of Gliocla- dium penicilloides is discussed. The new species isolated from the dermatomycosis case at the Barnard Free Skin and Cancer Hospital is designated as Cephalosporium symbioticum. IX. ACKNOWLEDGMENTS The author wishes to express her appreciation to Dr. C. W. Dodge for kindly direction in this study; to Dr. Morris Moore for technical instruction; to Dr. G. T. Moore for the use of the facilities of the Missouri Botanical Garden; to Dr. Martin F. Engmann, Sr., for permission to collect cultures at the clinic of the Barnard Free Skin and Cancer Hospital; to Dr. E. S. Reynolds for physiological advice and apparatus; to Dr. J. M. Greenman for suggestions regarding a nomenclatorial diffi- culty; to Dr. R. E. Woodson, Jr., Dr. E. E. Berkley, and Dr. J. A. Moore for photographic assistance. BIBLIOGRAPHY Almeida, F. de (34). Formacóes radiadas da membrana dos cogumelos parasitos. Ann. Fac. Med. Sao Paulo 10: 163-174. 1934. Amann, J. (796). Conservierungsfliissigkeiten und Einschlussmedien für Moose, Chloro-und Cyanophyceen. Zeitschr. f. Wiss. Mikr. 13: 18-21. 1896. [Vor. 23 52 ANNALS OF THE MISSOURI BOTANICAL GARDEN Area-Leáo, A.-E. de, et J. Lobo (734). Mycétome du pied à Cephalosporiwm recif ei n.sp. Myeétome à grains blanes. Compt. Rend. Soe. Biol. 117: 203-205, 1934, Bainier, G. (707). Gliocladium roseum. Bull. Soc. Myc. France 23: 111-114. 1907. Bary, A. de (784). Mi uio cepe E und Biologie der Pilze, Mycetozoen und Bacterien. 558 884, Biourge, P. (’23). Les moisissures du Bs Penicillium Link. Etude mono- graphique. La Cellule 33: 7—331. Brefeld, O. (774). Die Goeres von Penicillium. Bot. Unters. ti. chimmelpilze 2: 1-98, 1874. Brierley, W. B. (’15). The endoeonidia of Thielavia basicola. Ann. Bot. 29: 483- 493. 1915. m SEH R. E. ("11). Morphology of the genus eege with description of a new species and a variety. Mycologia 3: 170—174. and E. I. Fulmer (’28). Ph eid and ef of bacteria. 3 vols. Williams & Wilkins Co. Baltimor 8. Buller, A. H. R. (709, 722, ’33). Researches on Se 1: 1-274. 1909; 2: 1-492. 1922; 5: 1—416. 1933. Longmans, Green & Co. London. Castle, E. S . (234). The spiral Eer of single cells. Science 80: 362-363. 1934. ski R. . (32), Cephalospori udofermentum n. sp. isolato dalla bocca dell Arch, f. Protistk. ep ena 1932. E W. M. (722). The determination of hydrogen ions. 317 pp. Williams and Wilkins Co. Baltimore, 1922. Clements, F. E., and C. L. Shear E The genera of fungi. 496 pp. H. W. Wilson C termined by nephelometry. PEN Jour. Bot. 18: 266-287. 1931. Cooke, M. E., and J. B. Ellis (778). New Jersey fungi. Fei 7: 6. 1878. Corda, A. C. J. (1839). Pracht-Flora europaeischer Schimmelbildungen. 55 pp. Gerhard Fleischer. Leipzig, 1839. Czapek, F. (*11). Chemical phenomena in life. 148 pp. Harper € Bros. London, 1911. Czurda, V. (’33). Experimentelle Analyse der kopulationsauslósenden Bedingungen bei Mikro-organismen. I. Untersuchungen an Algen (Spirogyra, Zygnema und Hyalotheca). Beih. Bot. Centralbl. I. 51: 711-762. 1933 Dodge, C. W. (735). Medical mycology. 900 pp. C. V. Mosby Co. St. Louis, 1935. Duggar, B. M. (’09). Fungous diseases of plants. 508 pp. Ginn and Co. Boston, 1909 Duthie, E. S. (’33). Mucus formation in goblet cells. Proc. Roy Soc. London, B, 3. : 193 Eidam, E. (’83). Zur Kenntniss der Entwicklung bei den Ascomyceten. Cohn’s Beitr. Biol. d. Pflanzen. 3: 377. 1883. i W. (727). Über CR SUN: Befunde bei Pemphigusartigen Erkran- ungen. Derm. Woch. 85: 1337-1348. Pub J. H. (704). Development of ascus and nae formation in Ascomycetes. Proc. Boston Soc. Nat. Hist. 32: 77-114. Fineman, B. C. (’21). A study of the thrush GE Jour. Inf. Dis. 28: 185-200. 1921. Fuchs, J. Ge Schimmelpilze als Hefebildner. Zentralbl. f. Bakt. II, 66: 490— 500. 1 1936] PINKERTON—CONIDIAL FORMATION 53 Gáumann, E. A., and C. W. Dodge (’28). Comparative morphology of fungi. 701 pp. Me Due Hill Book Co. New York, 1928. Gilbert, W. W. (709). The root-rot of tobacco caused by Thielavia basicola. U. S. Dept. Agr. Bur. Pl. Ind. Bul. 158: 55 pp. 1909. Goss, R. W., and P. W. Frink (734). Cephalosporium wilt and die-baek of the white elm. Univ. Nebr. Agr. Exp. Sta. Res. Bull. 70: 24. 1934. Gregory, P. H. (735). The dermatophytes. Biol. Rev. 10: 208-233. 1935. Rb M. F. (’05). Gliomastiz chartarum n. gen., n. sp.; Se ee a formation endogéne des conidies. Ha Sos Mye. France 21: 230. 1905. Been A. (727). Etude eytologique et taxonomique sur we levüres du genre Sporobolomyces. Bull. Soe. Mye. Franee 43: 245-258. 1927. ————— ——, (239). The cae development of our idea of the vacuome of plant cells. "ma Jour. Bot. 16: 1-22. 1929. ———— — ——, G. Mangenot, et L. Plantefol qu" Traité de cytologie vegetale. 1195 pp. Librairie E. le Francois. Paris, 193 Hansen, H. N., and R. E. Smith (732). The Lou of variation in imperfect ungi, Botrytis cinerea, Phytopath. 22: 953—964. 1932. Harper, R. A. ('99). Cell division in sporangia and asci. Ann. Bot. 13: 467—525. 1899. Hartmann, E. (726). Über ein gemeinsames Vorkommen von Cephalosporiwm und Trichophyton gypseum. Derm. Woch, 82: 565. 1926 Henrici, A. T. (’30). Molds, yeasts and actinomycetes. 296 pp. John Wiley and Sons. New York, 1930. Hewitt, L. F. (733). Oxidation-reduction potentials in bacteriology and bio- chemistry. 2nd ed. London County Council. 1933. Jacobson, H. P. (’32). Fungous diseases. 317 pp. Chas. C. Thomas Co. Spring- field, Ill., 1932 Knoll, F. (12). Untersuchungen über den Bau und die virg der Cystiden und verwandter Organe. Jahrb. f. wiss. Bot. 50: 453—501. Langeron, M. (722). Utilité de deux nouvelles coupures nk dans les Peri- sporiacés: Diplostephanus n.g. et Carpenteles n.g. Compt. Rend. Soc. Biol. 87: 343-345. 1922. Larson, W. P. (721). The influence of the surface tension of the culture medium on bacterial growth. Soc. Exp. Biol. € Med., Droe, 19: 62-63. 1921. Lindau, G. (707-710). Fungi Imperfecti. Saber Te Krypt. Fl. Deutschl. 9: 1907-10. Maire, R. (’02). Recherches eytologiques et taxonomiques sur les Basidiomycetes. Bull. Soe. . 1902 ee e upt, A. St. von. (732). a Wert pm pH-Messung bei Pilzkulturen. tralbl. f. Bakt. 125: 368—374. 1932. eer R. (793). Contributo alla biologia del fungo del mughetto. Ann. Ist. 23. Ig. Roma 3: 199-2 1893. Matruchot, L. (795). E i E et forme parfaite des Gliocladiwm. Rev. Gén. Bot. 7: 321- Milochevitch, S. (729). ie wes und die Wachstumsformen des Soorpilzes uf verschiedenen Nàührbóden b "E auf solchen die mit Extrakten aus Körperorgane hergestellt wurden. Zentralbl. f. Bakt. 114: 174—185. 1929. Moore, M. (’33). A study of Endomyces pacer Rewbridge, Dodge a Ayres. Ann. Mo. Bot. Gard. 20: 471-568. 1933. [Vor. 23 54 ANNALS OF THE MISSOURI BOTANICAL GARDEN der D. (723). Das Golgische Binnennetz und seine Beziehungen zu der Sekre- tion. Arch. Mikr. Anat. 97: 136—186. 1923. Oort, A. J . P. (731). The spiral-growth of Phycomyces. Proc. Acad. Sci. Amster- am 34: 564—575. 1931. TO hp A. J. A. (786). Contributions à la flore mycologique des Pays-bas. 2-562. 1886. in E S. (722). Carotinoid and related pigments. 316 pp. Chem. Cat. Co. Peteh, T. (728). Studies on entomogenous fungi. VI. Cephalosporiwm and as- sociated fungi. Brit. Myc. Soc. Trans. 10: 152-182. 1928. Piney, A. (’31). Recent advances in microscopy. 260 pp. P. Blakiston’s Son and Co. Philadelphia, 1931. Plaut, H. C. (’13). Die Hyphenpilze oder Eumyceten. Kolle und Wassermann, Handbuch der path. Mikroorg. ed. 2. 5: 1-154. 1913. Ramsbottom, J. (733). The gege 's address: Fungi imperfecti. Jour. Quekett Mier. Club 16: 261-276. Reddy, c. S., and J. R. xe de m The black-bundle disease of corn. Jour. Agr. Res. 27: 177-205. 1924. Ridgway, R. ( pie Saag standards and color nomenclature. 43 pp., 53 pl. Wash- ington, D. C., jipes G., et G. breng (’90). Recherches morphologiques sur le champignon du muquet. Arch. Med. Exp. 2: 62-87. 1890. Saccardo, P. A. (1882-1931). Sylloge Fungorum. Publ. by author. Padua, 1882- 1931. Sass, J. E. (734). The presence of a Nebenkern and Golgi material in Coprinus sterquilinus Fr. La Cellule 43: 343—348. 1934. Searamella, P. (728). Ricerche preliminari sul molo di formazione dei conidi nil el picor digitatum.’’ Nuov. Giorn. Bot. Ital. N. S. 34: 1078-1084. 1928. Searth, G. W. (’27). The dope qi organization of plant protoplasm in the light of mierurgy. Protoplasma 2: 189—205. 1927. Seynes, J. de (786). Ree leches p pour servir à l'histoire naturelle des vegetaux inferieurs. III, lst part: ‘‘De la formation des corps reproducteurs appelés acrospores.?? Compt. Rend. Acad. Sei. [Paris] 102: 933-934. : Sharp, L. W. (’34). Introduction to cytology. 567 pp. MeGraw Hill Book Co. ed. 3. 1934. Stellung-Dekker, N. M. (731). Die Sporogenen Hefen. I Teil 547 pp. Amster- am, 1931. —— K. L. (735). Studien über die Entwicklungsgeschichte einiger Entyloma- n und über die pliss ai Stellung der Familie der Sporobolomycetes. Zeitsehr. f. Bot. 28: 225—259. 1935. Talice, R. V. (730). ¿de ge filamentisation des Monilia. Ann. Parasitol. Hum. Comp. 8: 394—410. 1930. Thom, C. (710). Cultural studies of species of Penicillium. U. S. Dept. Agr. Bur. Anim. Ind. Bull. 118: 1-109. 1910. ; (14). Conidium production in Penicillium. Mycologia 6: 211-215. 1914. — — —— ———-, (26). The Aspergilli. 272 pp. Williams and Wilkins Co. Balti- more, 1926. 1936] PINKERTON—CONIDIAL FORMATION 59 , (730). The Penicillia. 643 pp. Williams and Wilkins Co. Balti- more, 1930. Topley, W. W. C., and G. S. Wilson (’29). The principles of bacteriology and im- munity. Edward Arnold and Co. London. 1st imp., 192 Vuillemin, P. (731). Les champignons parasites et les mycoses de l’homme. Encycl. Myeol. 2: 290 pp. Paris, 1931. adas K. (731). Contributions to the cytology of fungi. III. Chromosome umber in Aspergillus. Cytologia 2: 291-301. 1931. Weidman, F. (732). dictó formations due to a hyphomycete (Aspergillus (?)). Arch. Path. 13: 725-744. 1932, Weier, T. E. (732). The structure of the bryophyte plastid with reference to the Golgi apparatus. Am. Jour. Bot. 19: 659-672. 1932. Went, F. A. F. C. (704). Uber den Einfluss des Lichtes auf die Entstehung des Carotins und auf die Zersetzung der Enzyme. Ree. trav. bot. Néerland. 1: 106— 119. 1904. Wilson, E. B. (25). The cell in development and inheritance. Macmillan Co. New York. 3rd ed., 1925. Wilson, J. K. See The nature and reaction of water from hydathodes. Cornell 923. Agr. Exp. Winter, G. (787). Die Pilze. Rabenhorst’s Krypt-Fl. 2: 61. Leipzig, 1887. Zopf, W. (790). Die Pilze. 500 pp. Breslau, 1890. [Vor. 23, 1936] 56 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 1 All drawings, except fig. 17, made with the aid of a camera lucida. Fig. 1. Mature compound cephalosporium in Clonostachys Araucaria. Nutrient broth. x 750. Fig. 2. Incipient phialide in Gliocladium penicilloides. Glycerine agar. x 750. Figs. 3-4. Basophilic granules in phialides of G. penicilloides. Glycerine agar. 1 Fig. 5. Exudation of mucilage in Clonostachys Araucaria. Nutrient broth hang- ing drop. x 1250. Fig. 6. Phialide enclosed in mucilage in C. Araucaria. Endo’s agar. x 750. Fig. 7. Orientation of granules at tips of phialides, and cleavage vacuoles in Gliocladium penicilloides. Glycerine agar. x 1250. Fig. 8. Splitting of phialide tip in G. aie Glyeerine agar. x 1250. Figs. 9-10. Emergence of conidia in G. penicilloides. Glycerine agar. x 1250. Fig. 11. Endoconidia in G. penicilloides. ane erine agar. x 1250. ig. 12. Twisted phialide with uninucleate endoconidia in G. penicilloides. ge agar. 1250. Fig. 13. Mature uninucleate conidium of G. penicilloides. Glycerine agar. x 1250. g. 14 Vorne de with endoconidia, sr 5d vaeuolate spores in Clomo- stachys Arauca: onfusa. Litmus milk. Fig. 15. Geier of Gliocladium um FAR paraffin sections at 3 y. Glyeerine agar. x 1250. ig. 16. Constriction of spore mass during exit in G. penicilloides. Litmus milk. Fig. 17. Diagram showing successive production of conidia, from observations at the end of 10, 20, 35, 65, and 70 minutes in Clonostachys Araucaria. Sabouraud’s dextrose agar. Figs. 18-28. Allescheria Boydii. Fig. 18. Submerged EE Nitrate agar. x 750. Fig. 19. Conidium. Raulin’s agar. 50. Figs. 20-21. Saco of Poa sed of submerged cephalosporia. Sabouraud’s dextrose agar. x 350. 2 Fig. Coremium stage. Sabouraud’s dextrose agar. x 1250. Fig. 23. Submerged lateral spore. Raulin’s lët x 1250. Fig. 24. Cephalosporium. Sabouraud’s dextrose agar. x 1250. Fig. 25 ature and young spore. Sabouraud » dextrose agar. x 350. Fig. 26. Lateral ACC Sabouraud’s dextrose agar. x 1250. Fig. 27. Young and mature spores in cephalosporium. Raulin’s agar. x 350. Fig. 28. Conidiophore. Raulin’s agar. x 1250. oo > ; z E Bit 3 E M 3 f Ge * ` X e ANN. Mo. Bor. GARD., VOL. 23, Die m A Si: d 36 M PINKERTON — CONIDIAL FORMATION [Vor. 23, 1936] 58 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 2 Unless otherwise stated, all drawings were made with the aid of a camera lucida. Figs. 1-11. Cephalosporium Serrae Fig. 1. Terminal chlamydospore. Potato agar. x 1250. Fig. 2. Series of chlamydospores. Potato agar. x 750. rn me 5 250. ig. 5. Twisted phialide with endoconidia. Sabouraud’s dextrose agar. x 1250. Fig. 6. Phialide with vacuoles. Sabouraud’s dextrose agar. x 1250. Fig. 7. Dimorphie um Maltose agar. Not camera lucida. Fig. 8. Diagram of fruc tion, Fig. 9. Germinating eiii Nutrient broth. x 750. Fig. 10. Mucilage drop. Nutrient broth. x 750. Fig. 11. Packet type of PE Nutrient broth. x 750. Figs. 12-16. Cephalosporium niveolano Fig. 12. Terminal ehlamydospore. x n Fig. 13. Chlamydospores. Corn mealagar. x 750. Fig. 14. Phialide with endoconidia. Nutrient agar. x 750. Fig. 15. Phialide producing spore. Nutrient agar. x 750. Fig. 16. Young eephalosporium. Nutrient agar. x 750. Figs. 17-21. Cephalosporium Stuehmeri. Fig. 1 Conidiophore with endoconidium. Sabouraud’s dextrose agar. x 750. Fig. 18. Diagram of fructification. Fig. 19. Terminal chlamydospore. Corn meal agar. x 750. Fig. 20. Chlamydospores. Sabouraud’s dextrose agar. x 350. ig. 21. Peculiar aerial spore. Sabouraud’s dextrose agar. x 350. Figs. 22-25. pa osporium rubrobrunneum. Fig. 22. Origin of phialides from ropes of hyphae. aoe agar. x 350. Fig. 23. Slanted spore cluster. Sabouraud’s dextrose agar 50. Fig. 24. Young cephalosporium. Sabouraud’s dextrose agar. x 350. Fig. 25. Uninucleate spore. Sabouraud’s dextrose agar. x 350. Cephal GI iliense. Fig. 26. Diagram of fructification. Endo’s agar. Fig. 27. Cephalosporium. x 350 Fig. 28. Packet cephalosporium. Sabouraud’s dextrose agar. Not camera Fig. 29. Cephalosporium. Sabouraud’s dextrose agar. x 350. Fig. 30. Chlamydospores. Corn meal agar. x 1250. Fig. 31. Chlamydospores. Malt extract broth. x 1250. symbioticum. Fig. 35. Conidia. Nutrient agar. x 1250. Fig. 36. Young submerged growth. Sabouraud’s dextrose agar. x 1250. Fig. 37. Submerged growth on wort agar. x 1250. [Vor. 23, 1936] 60 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 2 (Continued) Fig. 38. Chlamydospores. Sabouraud’s dextrose agar. x 1250. Figs. 39-44, Torulopsis mucilaginosa. Fig. 39. Resting cell. Malt extract broth. x 750. in : Fig. 41. Submerged pseudomycelium. Nitrate agar. x 750. Fig. 42. Pseudomycelium and irregular spore clusters. Endo's agar. x 600. Fig. 43. Submerged cephalosporium. Nitrate agar. x 750 Fig. 44. Submerged sporiferous tip. Nitrate agar. x 750 Figs. 45-46. Cephalosporium symbioticum. Fig. 45. Cephalosporium with spore germinating in situ. Wort agar. x 1250. Fig. 46. Cephalosporium. Wort agar. 250. Fig. 47. Submerged myeelium of Torulopsis mucilaginosa, forming new cell by constriction. x 1250. Kë k em e Aww. Mo. Bor. Gap, VOL. 23, 1936. E PINKERTON — CONIDIAL FORMATION [Vor. 23, 1936] 62 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 3 Drawings made with the aid of a camera lucida. Figs. 1-19. Gliocladium penicilloides. Fig. 1. Conidiophore. Glycerine agar. x 350. Fig. 2. Diagram of fructification. Fig. 3. Cephalosporium. Nutrient broth. x 350. Fig. 4. Cephalosporia. Nutrient broth. x 600 Fig. 5. Phialides with basophilie granules. Glycerine e agar. 1250. Fig. 6. ide with mature spore and endoconidia. Glycerine agar. x 1250. Fig. 7. Phialide with broken tip. Glycerine agar. 1250. Fig. 8. Phialide with basophilie granules. Glycerine agar. x 1250. Fig. 9. Phialide with basal vacuole. Glycerine agar. 1250. Fig. 10. Conidium. Glycerine agar. x i x 1250 x Fig. 16. Germinating akinete. A agar. x 750. Fig. 17. Hypha. Glycerine agar. x 1250. Fig. 18. Hyphal fusions. Nutrient broth. x 1250. ig. 19. Sub-phialide cell. Glycerine agar. x 1250. Figs. 20-32. Gliocladiwm deliquescens. Fig. 20. Submerged mycelial cell showing granular bodies at septa. Glycerine x 1250. Fig. 21. Tip of mycelial cell with large dancing bodies and small terminal x 1250. granules. Nutrient broth. Fig. 2 oung eonidiophore. Glycerine agar. x 1000. Fig. 23. Cephalosporium. x 350. Fig. 24. Diagram of cephalosporium branching structure. Fig. 25. Conidia with chromatic material. Glycerine agar. Fig. 26. Conidia from smear stained with gentian violet. Tl Fig. 28. Chlamydospore (?). Sabouraud’s dextrose agar. x 800. h Fig. 29. Chlamydospores. Glycerine agar. x 1000. Fig. 30. Aborted ascogonium (1). Glycerine agar pH 3.8. x 350. Fig. 31. Germinating akinete. Sabouraud's dextrose agar. 32. Phialide with been granules. Glycerine agar. Fig. Figs. 33-41. Gliocladiwm fimb Fig. 33. Germinating pa Endo’s agar. x 600 Fig. 34. Chlamydospores. Sabouraud's dextrose agar. Fig. 35. Conidiophore. Corn meal agar. x 75 0 Fig. 36. Dimorphie fruetifieation, Sabouraud’s dextrose agar. 50. Fig. 37. Conidia. Sabouraud's dextrose agar. Fig. 38. Akinetes. Glycerine agar. Fig. 39. Conidiophore. Sabouraud’s APPS agar. x 600. x 750. x 1250. x 350. x 350. Fig. 40. inete upon addition of alcohol and chloroform, showing dissolution 750. of oil x Fig. 41. Intercalary chlamydospores. Corn meal agar. x 750. i LUN 4 : 3. 25 a : T "RK i AS Kä as S Ka er Ee a l Ue jr i dE Es 2 ANN. Mo. Bor. GARD., VOL. 23, 1936 Libo cp p | PINKERTON — CONIDIAL FORMATION [Vor. 23, 1936] 64 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 4 Drawings made with the aid of a camera lucida. Figs. 1-14. Gliocladium Vermoeseni. Fig. 1. Submerged chlamydospores. Sahourend 's dextrose agar. x 350. Fig. 2. Aerial ehlamydospores. Raulin's agar. x 750. Fig. 3. Endoconidial method of Prats of chlamydospores. Sabouraud’s dextrose agar. 750. Fig. 4. Arthrospores. Corn meal agar. x 750. Fig. 5. Conidiophore. Sabouraud’s dextrose agar. x 750. Fig. 6. Mature conidium. Sabouraud’s dextrose agar. x 750. i Part of cephalosporium showing exterior thiek granular layer of mucus and spores ee by pressure. Sabouraud’s dextrose agar. x : Fig. 8. anch of penicillate fructifieation. Czapek’s agar. x 350. Fig. 9. Ser Corn meal agar. 50. Fig. 10. Unusual eonidiophore. Raulin’s agar. x 750. Fig. 11. Oidia (?). Raulin's agar. x 750. Fig. 12. Anomalous structure. Raulin’s agar. x 1250. Fig. 13. Diagram of fructification. Fig. 14. Submerged akinete. Raulin's agar. x 750. confus ig. 15. Conidiophore. Czapek's agar. x 350. Fig. 16. Conidiophore with warts, Czapek’s agar. x 350 Fig. 17. Short lateral cephalosporium. Sabouraud's dextrose agar. x 350. Fig. 18. Phialides with endoconidia. Endo's agar. x 1000. Fig. 19. Chlamydospore (?). Sabouraud's dextrose agar. x 600. i Dimorphie fructification. Lactose broth. x 600. Figs. 21-22. Copulating conidia. Litmus milk. x 1250. Fig. 23. Mature conidium 750. Fig. 24. Germinating conidium. Sabouraud’s dextrose agar. x 750, Fig. 25. Hyphal fusions. Endo’s agar. 50. Fig. 26. Submerged ehlamydospore. Raulin’s agar. x 750. Figs. 27-31. Clonostachys Araucaria var. rosea. Fig. 28. Chlamydospore. Corn meal agar. x 750. Fig. 29. Submerged chlamydospores (?). Wort agar. x 350. Fig. 30. Phialides with endoconidia. Sabouraud's dextrose agar. x 750. Fig. 31. Hypha with warts. Sabouraud's dextrose agar. x 750. A —— P». "m lo eso Zoe p a A^ y Ve a LE Sr 2 PINKERTON — CONIDIAL FORMATION [Vor. 23, 1936] 66 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 5 Drawings made with the aid of a camera lucida. Figs. 1-7. Clonostachys spectabilis. Fig. 1. Terminal chlamydospores. Cherry agar. ig. 2. Interealary chlamydospore. Malt extract broth. x 750. Fig. 3. Submerged terminal cell (?). Malt extract broth. x 750. Fig. 4. Botrytis type of spore. Sabouraud’s dextrose agar. x 350. Fig. 5. Same. Corn meal agar. Fig. 6. Mycelium associated with Bose type of fructification. Czapek’s agar. x 350. Fig. 7. Spores. Corn mealagar. x 750. Figs. 8-13. Scopulariopsis brevicaulis. Fig. 8. Young conidium before putting down endogenous wall. Nutrient agar. x 1250. Fig. 9. Mature conidium and phialide showing segmentation. Nutrient agar. 0 Fig. 10. Conidium. Nutrient agar. 50. Fig. 11. Branched eonidiophore. Nutrient broth. x 1250. Fig. 12. Phialide with conidia. Nutrient agar. x 1250. Fig. 13. Conidia. Nutrient agar. x 1250. Figs. 14-26. Clonostachys Araucaria. Fig. 14. Chlamydospores. Corn meal agar. Fig. 15. Myeelium with warts. Sabouraud's e H agar. x 750. Fig. 16. zi edd conidiophore. x 350. Fig. 17. Cephalosporium and phialide with endoeonidia. x 1250. Fig. 18. Phiali de pen twisted walls. Sabouraud's dextrose agar. x 600. Fig. 19. Fructification like Corda’s type. Sabouraud's dextrose agar. x 1250. Fig. 20. Diagram showing a type of branehing. Fig. 21. Another type of branching. Fig. 22. Phialide with warts. Czapek’s agar. x 350. Fig. 23. Endoconidia. Sabouraud’s dextrose agar. x 1250. Fig. 24. Intercalary ehlamydospore. neis: broth. x 600. Fig. 25. Germinating conidium. Lactose broth. x 600. Fig. 26. Conidiophore with six e branches. Wort agar. x 600. oe ee BEEN er ee ee em! ee Ee ANN. Mo. Bor. GARD., VOL. 23, 1936 PLATE 5 PINKERTON — CONIDIAL FORMATION :OCKAYNE, BOSTON. [Vor. 23, 1936] 68 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXXPLANATION OF PLATE PLATE 6 Figs. 1-14, giant colonies on Czapek's agar. Fig. 1. Cephalosporium niveolanosum Fig. 2. Cephalosporium symbioticum Fig. 3. Gliocl de ens Fig. 4. Gliocladium fimbriatum Fig. 5. Gliocladium penicilloides Fig. 6. Gliocladium V ermoeseni Fig. 7. Allescheria Boydú Fig. 8 lonostachys Araucaria var. confusa Fig. 9. Cephalosporium rubrobrunneum Fig. 10. Clonostachys Araucaria var. rosea Fig. 11 lono ys Araucaria Fig. 12. Cephalosporium kiliens Fig og Cep ium Serrae Fi ig. eph m Stuehmeri 7. Branched conidiophore of Clonostachys Araucaria with warts and cephalosporia. x 1000. Fig. 18. Agar micro-culture showing cephalosporia in Cephalosporiwm symbi- oticum. x 1000. ANN. Mo. Bor. GARD., VOL. 23, 1936 PLATE 6 17 PINKERTON — CONIDIAL FORMATION A STUDY OF THE GENUS HELICOGLOEA* GLADYS E. BAKER Henry Shaw School of Botany of Washington University Formerly Jessie R. Barr Research Fellow in the Henry Shaw School of Botany of Washington Uniwersity A paper published by Patouillard in 1892 included amongst the heterobasidiomycetes the new genus Helicogloea, repre- sented by a single species—H. Lagerheimi Pat. Later Patouil- lard seemed confused and uncertain as to its true identity and eventually, in a letter to Coker (720), stated that Helicogloea was not distinet from Platygloea. Inthe literature subsequent references are based on Patouillard's work. No one seems to have examined the type. Just as this work was going to press under the name of Saccoblastia, Dr. Donald Rogers called at- tention to the type which is now in the Farlow herbarium. Patouillard’s description fails to mention the ‘‘sac’’ or hypo- basidium, easily the most conspicuous part of the fructification. From his illustrations one is forced to conclude that although Patouillard was aware of the structure he was unable to inter- pretit. There can be no doubt as to the real nature of the speci- men. Not only is it a good genus but the type of what three years later Moller (795) called Saccoblastia. Since Patouil- lard’s formal description does not include a statement con- cerning the hypobasidia the genus stands in need of emenda- tion. Accordingly, the recognized species of Saccoblastia must be transferred to the older genus, Helicogloea. Heticociora Pat., emend. Fructification resupinate, indeterminate, ranging from a very thin film to a layer several mm. thick, mucous-gelatinous or 1 An investigation carried out in the graduate laboratory of the Henry Shaw Sehool of Botany of Washington University, and submitted as a thesis in partial fulfillment of the requirements for the degree of doctor of philosophy in the Henry Shaw Sehool of Botany of Washington University. Issued March 20, 1936. (69) [Vor. 23 70 ANNALS OF THE MISSOURI BOTANICAL GARDEN floccose ; hyphae with or without clamp connections, 2—6 » wide, branched, sometimes anastomosing, attenuated at the septa; basidia heteromorphie, terminal or intercalary in origin; hypo- basidia conspicuous, saccate lateral (with one exception) expansions of the basidial primordia sometimes variously con- stricted, 15-20-58 x 4.5-8-16 »; epibasidia cylindrical out- growths of the primordia or hypobasidia, expanding apically to the sporogenous portion, articulate-geniculate, 3-4-septate, each segment producing a sterigma and spore, or an accessory appendage eventually tipped by a sterigma and spore, the en- tire basidium 45-140 x 4-11 »; spores hyaline, ovoid to ellipsoid, often flattened on one side, with a prominent blunt lateral apiculus, 7-19 x 4-13 »; germination by repetition. Helicogloea is a member of the Auriculariaceae, hetero- basidiomycetes which bear basidiospores in linear sequence on transversely septate basidia. The genus, by reason of a cu- rious sac-like appendage produced during basidial ontogeny, has always been noteworthy. Since the first clear morphologi- cal account (Moller, ’95) several species have been described but little detailed information concerning the developmental cycle has accumulated. Gáumann and Dodge (728) summarize the situation by saying: ‘‘A cytological study of the species is needed for the interpretation of this storage organ which is unknown in other groups of fung). "7 In 1895 Moller (1. c.) described two species collected during a stay in Brazil. Both possessed the unique saccate hypobasidia. Móller completely delineated the morphology of the genus. Further detailed studies were not made. As apparently Patouillard's material was never questioned, this work of Móller”s stood as the standard of the genus Saccoblastia, repre- sented by the species S. ovispora Moll, and S. sphaerospora Moll. The next record is that of Bresadola (703), who added a species from Poland, Helicogloea graminicola. His deserip- tion was accompanied by neither detailed studies nor illustra- tions. Likewise Bourdot and Galzin (709, ’28), who have enlarged the genus by several species and varieties, have con- tributed only taxonomie observations. 1936] BAKER—A STUDY OF HELICOGLOEA TE The first collection of the genus in the United States was made by Couch, 1920, in North Carolina. Coker (’20), puzzled by its peculiar characters, which were so suggestive of Heli- cogloea Lagerheimi yet differing in several respects, published a description of the fungus as Saccoblastia ovispora Moller var. caroliniana. In 1928, in the meantime having consulted Bresa- dola who supposedly had seen the type of the species in ques- tion, he decided to recognize it as an autonomous species, now Helicogloea caroliniana. The original description includes a careful morphological account with lucid illustrations. Linder (729) announced a new species based on material collected in Cuba, Helicogloea intermedia. This remarkable form was found to have two types of basidial development, the so-called ‘‘clavate’’ type in addition to the established saccate one. From his preparations Linder was able to augment con- siderably the cytological data. But even so they remained far from complete. The condition of his material which had been preserved in 70 per cent alcohol was hardly suitable for critical work. Helicogloea Lagerheimi Pat. was collected in Iowa during the summer of 1932 by Rogers and later reported by him (733). He made additional gatherings during the summer of 1934 and at both times killed and fixed samples of a number of collec- tions. It is material from these sources upon which this study is based. TECHNIQUE Several collections of fresh material of Helicogloea Lager- heimi were available for histological preparations. All of them were killed and fixed in Flemming’s weaker fluid. To the fixing fluid for two lots was added 1 per cent maltose. For two others dehydration by a butyl series was employed, whereas the first mentioned were run through Chamberlain’s (’32) full alcohol schedule to xylol and paraffin. Those of the maltose-xylol series proved the most valuable for critical stages. However, subsequent examination indicated that variation was due to the particular specimen as much as to fixation. Serial sections were cut at 15, 10, 8, and 5 p. Various stains were tried but [Vor. 23 72 ANNALS OF THE MISSOURI BOTANICAL GARDEN Haidenhain’s iron-alum haematoxylin with a counter stain of phloxine (a 1 per cent solution in 95 per cent alcohol) was used almost exclusively. A few slides were prepared with Flem- ming’s triple stain to serve as a check for the presence of septa at certain points. In addition whole mounts were made of herbarium specimens whenever possible. After these were revived in a moist chamber they were treated with a weak solu- tion of potassium hydroxide and stained with phloxine (1 per cent solution in water). MonPnHorLocy or H. LAGERHEIMI The indeterminate fructification of Helicogloea Lagerheimi in a living condition, or when it has been revived in a moist- chamber, appears as a thin gelatinous or mucous coat, loosely attached to its substratum of decorticated and well-decayed wood of various sorts. It ranges in thickness from 100 y to 1mm. From prepared sections it is evident that the penetra- tion within the substratum is slight. A few scattered hyphal threads are all that appear within the wood cells, and these are never deeper than a few y. Morphologically, there is no tissue differentiation into trama and hymenium, but the whole is simply a mass of hyphal and basidial elements loosely interwoven (fig. 19). This composite figure of the fructification as seen in section represents one in . which the hyphae are practically parallel, a feature character- istic of lesser developed fructifications. The development of the fruiting body ranges from a thinly spread layer to thicker ones in which the basidial elements are restricted to the upper layer. In moderately developed specimens the fruiting layer may be as much as one-third the total, or in better developed ones as little as one-sixth, as shown by the comparative scales (fig. 20). The fruiting region is the actively growing part, bearing basidia singly at various levels. The individual hyphal cells are typically binucleate. Occasionally some cells are four-nucleate, but this merely indicates a recent nuclear division not yet followed by wall formation (fig. 14a, d). Usually the two processes are concurrent. The nuclei may lie 1936] BAKER—A STUDY OF HELICOGLOEA 73 in close approximation or be widely separated within the same cell. As the cells become more mature the protoplasmic content is reduced, sometimes to only scattered remnants about the nuclei (fig. 14b, c). The mycelium is slightly irregular, char- acteristically attenuated at the septa, and its individual cells vary eonsiderably in length. Branches arise as lateral out- growths of a cell and receive daughter nuclei of the parent-cell dikaryon (fig. 15). There are no clamp connections, but the hyphae branch freely. Anastomoses also occur, although there is no evidence to support Linder's contention that such hyphal union carries a sexual stimulus to fructification. The morphological development of the basidium may be fol- lowed in the diagrammatic series, figures 1 to 13 inclusive, and seen in summation in the composite illustration, figure 19. These figures, drawn to a common scale, show the entire cycle. The first indication of basidium production is the cutting off of a terminal hyphal cell, the primordial cell, easily distinguished by its more homogeneous and denser staining protoplasm (figs. 15, 1). It contains two nuclei, descendants of the dikaryon of the subterminal cell. There may be additional basidia pro- duced by proliferation of the subterminal cell (fig. 16), just as shown by Moller in H. Lagerheimi and Coker in H. carolimana. After an increase in thickness the primordium pushes out a small lateral protuberance just above the basal septum (fig. 2). This enlarges rapidly to become the ‘‘sac,’’ or, as it will be designated henceforth, the hypobasidium, sensu Neuhoff (’24) (figs. 3, 4, 5, 6). Into this the two nuclei migrate and subse- quently fuse. Following a rather protracted resting period the fusion nucleus migrates from the hypobasidium, passing back through the primordial cell into the epibasidium, again sensu Neuhoff (fig. 7). The latter, an outgrowth of the primordium, may be apparent at an early stage. Furthermore, it may be distinctly elongated at the time the nuclei leave the hypobasidium or it may develop more slowly (figs. 7, 25). Between the primor- dium and the epibasidium there is commonly a conspicuous crimp or constriction, but never a wall (fig. 17). The epibasid- [Vor. 23 74 ANNALS OF THE MISSOURI BOTANICAL GARDEN ium is usually elongating as the nucleus moves along (fig. 8), and when it has attained its mature length the apical portion thickens and bends over (figs. 9,10). Accompanying this is a complete evacuation and withering of the hypobasidium. Later the withering includes the primordial part, and eventu- ally it extends to the lower portion of the epibasidium itself (figs. 11, 12). In mounts made with KOH these emptied por- tions may be quite swollen so that no collapse is apparent, but the regions are entirely drained of protoplasm. The nucleus now divides, the first septum coming in at once. Ina short time the two daughter nuclei divide, and the typical segmented ba- sidium is formed with four sterigmata and spores. These are so oriented commonly, through the curvature of the epibasid- ium, that the sporogenous portion lies procumbent on the sur- face of the fructification, thereby greatly favoring the chances of spore dispersal (figs. 12, 13). Bends are located below the spore-bearing portion or in it and simply adjust this region for favorable discharge (figs. 23, 63b). Excessively long thickened appendages in place of sterigmata occur frequently, but these eventually give rise to true sterigmata and spores (fig. 67b, c, d). Infrequently the basidia remain erect, in which case ap- pendages take care of the spore orientation or if the sporog- enous portion is unobstructed at the surface it need produce only sterigmata. As would be expected, there is considerable variation in the morphology of the basidium and its parts. The lateral hypo- basidium or ‘‘sac,’’ the most conspicuous feature of the fructifi- cation, is typically oblong-ovoid, but often it is constricted once or more (fig. 18a, b), a condition culminating perhaps in the extremes illustrated by f and g of the same figure. It should be noted that the general orientation of the fructification is such that the hypobasidia are only rarely pendent as extant accounts say. The fructification is resupinate, commonly effused on the under side of logs, hence the general configuration is to be con- strued inversely to the usual interpretation (fig. 19). Usually the epibasidium originates from a terminal primordial cell as a vertical outgrowth from it, but occasionally it seems to have 1936] BAKER—A STUDY OF HELICOGLOEA 75 been produced directly from the proximal end of the hypoba- sidium. The latter situation would then be equivalent to what apparently is the rule rather than the exception in H. inter- media. After examining Linder's material and his illustra- tions it is quite clear that there is a real difference in the origin of the basidial components of these two species. In H. Lager- heimi those hypobasidia which seemingly are not borne on ter- minal primordia with subsequent elongation of the epibasidia, but which appear to produce their epibasidia directly at their proximal end, are in reality borne on primordia which are short lateral branches of a proliferating complex. Hence their true origin is from a single terminal cell from which the hypobasid- ium arises in the ordinary fashion. This is clearly substanti- ated by the position of the walls (cf. figs. 21a, b; 75a, b). In H. intermedia one can easily trace the origin of the saceate type of hypobasidium from an intercalary binucleate cell which pushes out the hypobasidium, just as in H. Lagerhewma, and it in turn produees proximally the epibasidium (fig. 75b, c, d). There is, then, a fundamental difference between these two species in regard to the position and origin of the basidium and its components. In either species the type of development characteristic for one may occur as a rare exception in the other. Whether the primordium be terminal or intercalary, it is always distinguishable by its denser protoplasmic content. A study of H. caroliniana, H. Lagerheimi, and H. graminicola reveals that they are essentially identical at these stages. A third variation occurs in H. pinicola. Here the primordium is intercalary, but the epibasidium originates with a lateral out- pushing of the primordium, usually more or less apical, rather than from the proximal germination of the hypobasidium (fig. 72b, c). No ease of distal germination of the hypobasidium, such as Linder reported in his work, has been noticed in any of the material examined. Quite early the budding epibasidium is visible at the apical end of the primordium. It is seldom distinguishable before the latter attains its mature diameter or previous to the forma- tion of the hypobasidium, and it may not appear until much [Vor. 23 76 ANNALS OF THE MISSOURI BOTANICAL GARDEN later (cf. figs. 5, 22, 31b). Exceptionally, it is lateral in posi- tion. Its further development is either upright or variously bent, consequent upon its origin or the requirements for spore discharge (fig. 23). If there is a constriction between primor- dium and epibasidium it persists and is apparent in all stages, though there are numerous instances in which the epibasidium is little, if at all, distinct from the primordial portion (fig. 25b). The elongation of the epibasidium is various in relation to the accompanying nuclear behavior. It may follow one of three general patterns: I, The nucleus migrates from the hypobasid- ium as the epibasidium elongates (fig. 24) ; IT, The elongation of the epibasidium distinetly precedes the nuclear migration (fig. 25a, b) which may be delayed until a stage as late as that shown in figure 10; III, Migration precedes elongation. In such event there is little if any protoplasmic evacuation of the hypo- basidium until the epibasidium begins elongating (fig. 26a). The last scheme is infrequent except in less well developed fructifications. This results in a distinctly shorter basidium whose developmental sequence may be followed in figure 26b, c, d, and e. The other two types occur about equally, though type I probably represents the typical behavior. Type II seems more often associated with an extremely long uncon- stricted form of epibasidium and is undoubtedly a direct re- sponse to its location within the fructification. Types I and II intergrade. The type of behavior is consequent upon the pri- mordium location. In general, if the primordium is near the surface there is no need for great elongation, but if it is well within the fructification the longer form necessarily results. The crimp at the base of the epibasidium, though persistent, is never cut off by a wall. However, a second constriction is located characteristically just below the inflated sporogenous part. Such a joint has been noted by Coker in H. caroliniana; Wakefield and Pearson (’23) and Rogers (’34) show it in dif- ferent degrees for H. Lagerheimi. According to Wakefield and Pearson there is no wall at this point ; according to Rogers there is. Rogers’ figures present a condition comparable to the one Coker noticed. In H. Lagerheimi three septa are found 1936] BAKER—A STUDY OF HELICOGLOEA 77 more frequently than four in the sporogenous portion. Com- monly, protoplasmic remnants simulate a basal wall, but the actual presence of a wall is exceptional (fig. 58). Ordinarily, the second constriction lies just behind the evacuation level of the fourth segment (figs. 12, 13, ete.) ; it may be accompanied by a bend. At times the constriction is very deep, and if in addition the fourth segment has a basal septum, the condition is identical with Rogers’ illustration (fig. 27a, b). Whether the last segment does or does not have a basal wall, there is never any protoplasm left behind the segmented portion. In H. caroliniana the fourth septum is evident early in the de- velopment of the sporogenous portion (fig. 76b). Again the articulation receives varying emphasis so that one finds ex- amples ranging from a definite joint, as Coker showed, to a simple bend. CYTOLOGICAL OBSERVATIONS Each hyphal cell characteristically contains two small nuclei whose position within the cell is various. The origin of the two-nucleate condition was not determined. As the cells grow older and elongate their protoplasmic content is much reduced until it may consist only of scant bits around the nuclei. The nuclei of the hyphae, however, remain small. Each nucleus is composed of a deeply staining central nucleolus embedded in the homogeneous nucleoplasm and is surrounded by a delicate membrane which is barely differentiated from the cytoplasm in which the whole lies. The apical cells destined to become the basidial primordia are limited to the more actively growing regions near the surface of the fructification. Their contents more nearly fill the cell and stain as an even, homogeneous mass. Cohsequently these primordia and the sub-adjacent cells are the most favorable places to locate mitosis. This is typically conjugate, both nuclei being entirely synchronized in behavior, though the smallness of the cells and of the nuclei at this stage makes detailed analysis practically impossible. At times one finds the spindles in the well-known side-by-side orientation or slightly oblique (fig. 28a). Curiously enough, in by far the [Vor. 23 78 ANNALS OF THE MISSOURI BOTANICAL GARDEN majority of cases the figures are crossed at varying angles (fig. 28b, c). Future orientation of the spindles in this way is implied clearly in the early stages shown by a, b, and c of figure 29. Such behavior is probably due to the small diameter of the cell which would prohibit a side-by-side arrangement. No ad- ditional details could be discerned except that the membrane has disappeared by the time of spindle formation, and some- times the remnants of the nucleoli are visible in the nuclear cavity. Later they disappear (fig. 30). Following mitosis a wall is laid down between the two daughter dikaryons. This process may continue for several divisions but eventually the terminal cell constitutes the future primordium. The wall may not be formed immediately. In figure 31a, where it has just been laid down, the nuclei give evidence of a longer reorgani- zation. Here, too, the cell contents are quite vacuolate, the re- sult probably of rapid elongation and growth. 'The nuclei of the primordium are in a typical resting condition. When first delimited the primordium is little distinguished in size from the hyphal cells. Before the production of the hypobasidium it increases in diameter and somewhat in length, though again it is difficult to evaluate this exactly in a form so variable in size and behavior. Examples in figure 32 repre- sent an increase with reasonable surety, for they were ap- proximately adjacent in the same section. Beginning as a lateral outgrowth of the primordial cell just anterior to the basal septum, the hypobasidium rapidly as- sumes the form of an inflated sac-like appendage (figs. 31b, 33). At no time in its development is it without cytoplasm. As the hypobasidium elongates the nuclei in the primordium, which up to now have been more or less centrally located, enlarge slightly and prepare to move into the hypobasidium. "They are now somewhat ellipsoidal. Sometimes the nuclei lie in such close association that they seem to be fusing, but careful focusing reveals that their individual membranes are intact (fig. 34a, b). One nucleus of the pair may precede the other into the hypobasidium by a considerable distance (fig. 35). In any event, the movement is always tandem, with the long axes 1936] BAKER—A STUDY OF HELICOGLOEA 79 of the nuclei parallel to that of the primordium. There is no change in their shape as the nuclei pass through the narrow neck connecting hypobasidium and primordium. Following this migration the primordium is commonly more vacuolate (fig. 34b). Darker staining masses in the protoplasm are sug- gestive of a protoplasmic flow accompanying the nuclear passage (figs. 35, 37). Occasionally one nucleus of the pair may be larger, but this only signifies an earlier increase in its volume (fig. 38d, e). Once in the hypobasidium, the nuclei come together for fusion. Up to this time the hypobasidium has been growing but after karyogamy growth ceases. The two nuclei ordinarily do not penetrate below the mid-region of the hypobasidium, except in the case of the constricted hypo- basidia where they always lie in the larger portion (fig. 18a, b, f, 2). At the point of contact their membranes break down, al- lowing the components of the two nuclei to intermingle com- pletely (fig. 38c). Usually until then their structure is no more differentiated than in the resting stage, although in figure 38d one nucleus is already forming a reticulum. A little later the fusion nucleus is dumb-bell-shaped but this constric- tion is transitory (fig. 39). The fusion nucleus is bounded by © a continuous membrane; it contains two nucleoli, usually op- posite each other, and a reticulum in varying degrees of ad- vancement. The few variations of these earlier stages involve only minor deviations from the pattern. First, there is varia- bility in the position of the nuclei in the hypobasidium. Figure 38e illustrates deeper penetration than usual. Another differ- ence is that at times one nucleus is somewhat more advanced, as evidenced by its greater volume and earlier appearance of a reticulum (fig. 38e). In general these variations are rare, the migration and fusion processes being quite regular. Within the fusion nucleus, now spherical in outline, the nucleoli draw together to unite (fig. 40a, b, c). The single re- sultant nucleolus clearly has an inereased volume. By now the nucleus has very nearly attained its maximum size, with a diameter as much as five times that of the nuclei when in the primordium, almost equivalent to that of the hypobasidium. [Vor. 23 80 ANNALS OF THE MISSOURI BOTANICAL GARDEN Simultaneously the nuclear content is undergoing change. A sharp reticulum or network of threads develops throughout, highly suggestive of mitotic prophase. This increases in com- plexity until it reaches a stage illustrated in figure 41a. As the threads thicken, distinct, deeply staining beads representing aggregations of chromatic material are scattered throughout on them (fig. 41b). From this point on the nucleus is contract- ing and its entire contents become more concentrated and uni- formly deeply staining. The single nucleolus gradually shifts from an eccentric to a centric position. This contraction is preliminary to a resting stage which will be followed by nuclear migration from the hypobasidium through the primordium to the epibasidium (fig. 42a, b). This contracting or post-fusion stage is the one most commonly seen; hence it probably in- volves a considerable span of time. With its contraction the nucleus becomes more and more homogeneous until it resumes the typical resting condition. When completely contracted the nucleus has a total volume not much in excess of the unfused nuclei at the time of migration but the nucleolar material quite evidently has been doubled in volume (see figure 71 for com- parative nuclear and nucleolar ratios). The fusion nucleus ordinarily is in the center of the typical hypobasidium ; in the constricted type it lies in the largest portion. With the return to the resting condition the nucleus is ready to leave the hypobasidium. The primordial cell in the mean- time has given rise to the epibasidium, more or less greatly developed. If the migrating nucleus is at all elongate the whole body moves with its long axis parallel to that of the hypobasid- ium, later to those of the primordium and epibasidium (fig. 43a, b, c). Asinthe earlier migration, there is no attenuation or change of form when the nucleus passes through the neck, for the diameter of the migrating nucleus never exceeds that of the neck. When the nucleus leaves the hypobasidium the protoplasm also withdraws, leaving the distal portion first, until the hypobasidium is completely emptied and withers. Evacuation eventually extends to the primordium and lower 1936] BAKER—A STUDY OF HELICOGLOEA 81 portions of the epibasidium as their contents are withdrawn to augment the expanding sporogenous portion. The epibasidial development at this time follows one of the plans already outlined. Of these the first two are by far the most common. In accord with these types, I and II, the first evidence for nuclear division is the enlargement of the nucleus followed by the appearance of threads in the heretofore ho- mogeneous structure. This may begin at a level as early as that represented by figure 44b, but typically there are no signs of division until the nucleus comes to lie in the enlarged por- tion of the epibasidium, the actual sporogenous part (fig. 45a). By now all the protoplasm has withdrawn from the rest of the basidium into that region. There is seldom any further elongation or increase in diameter of the latter once division stages have been inaugurated. This first division of the sec- ondary nucleus is taken to be meiotic. Once again the nucleus increases in volume, often nearly equalling the diameter of the tube, usually much elongated, in general conformable to the space in which it lies (figs. 46, 49). An increase in volume is the first diagnostie sign of the prophase and is the more marked because of the lighter staining properties of the nucleoplasm. This is followed by the gradual appearance of threads which soon become definitely organized. The membrane is intact about the threads and nucleolus. Chromatic beads appear on the threads but they are never as pronounced as those seen in the fusion nucleus (fig. 47). The threads thicken, become denser, and frequently are localized at a pole opposite to the nucleolus, even from the beginning (fig. 48a, b). The nucleolus is usually peripheral, rarely central. Stages represented in figure 48 may be interpreted as synaptie. Following this, the chromatic material is increasingly concentrated until it is massed more or less centrally in the nuclear cavity (fig. 48c). Slightly later, individual clumps are distinguishable in the mass and probably represent pairs or groups of chromosomes. The nucleolus, as yet distinct, from this point onwards de- creases in size and finally disappears. No clear-cut evidence [Vor. 23 82 ANNALS OF THE MISSOURI BOTANICAL GARDEN was found for its direct expulsion into the cytoplasm. It usually is pushed to the edge of the nucleus quite early, and when the membrane breaks down lies free in the cavity or near by in the cytoplasm. Rarely is it distinct when the spindle is delimited ; in one exceptional case a nucleolus was still quite sharp at anaphase. Commonly, what has been interpreted as a last remnant of it is seen in the cytoplasm (figs. 50b; 51a, b). Development in accordance with type III must be considered a special case. There the prophase is distinct quite early in the development of the sporogenous portion, at times when the latter is just differentiated (fig. 45c) or even earlier. The nucleus here is less drawn out because it lies in a broader area. Aside from the earlier initiation point there is no further varia- tion in procedure. The whole is simply a shortened basidium whose developmental processes as a consequence are likewise foreshortened. A metaphase of the sort found in the higher plants has not been seen. As the chromatin mass separates into individual units, presumably chromosomes, a spindle is formed intra- nuclearly. Itis always oriented parallel to the long axis, never oblique. Because of the extremely small figure, spindle fibers are not distinguishable, but the whole region appears simply as a darker-staining area. At the poles there are minute but definite dark-staining bodies, the centrosomes—more properly centrioles, for no further differentiation can be made. "Their origin is unknown. Astral radiations do not occur. As the chromosomes pass into anaphase the nuclear membrane dis- appears, leaving the division figure in the nuclear cavity (fig. 90b, c, d, e). At anaphase the chromosomes pass to the poles in a very irregular manner, for they may be so scattered in their distri- bution as to extend from equator to poles (fig. 50). It is during this scattering, if ever, that one ean count the number of chromosomes present. The size of the figure and its chromo- somes precludes any finality of count. As nearly as can be determined there are at least five pairs of chromosomes of dif- ferent sizes. 1936] BAKER—A STUDY OF HELICOGLOEA 83 Even though the membrane has broken down, the nuclear cavity, by virtue of its lighter staining properties, remains distinct for much longer, as figure 51c, a late anaphase, illus- trates. There the spindle is stretched out and lengthened: the chromosomes are no longer discernible as individual units, for the majority are already massed together at the poles. There is no longer any trace of a nucleolus. The polar massing and the lengthening of the spindle continue until at telophase a condition represented by figure 52 is characteristic. The con- nective fibers or strands between the two polar masses are conspicuous. Similar persistent strands have been noted in the fungi by numerous authors. Tischler (’21—’22) designates them **karyodesmotische," Bensaude (718), ‘‘pont cinoplas- mique,’’ and Colley (’18), ‘‘suspension fibers." That their occurrence is far from uncommon is evident from the fre- quency of reports covering a wide distribution of forms (cf. Blackman, ’04, Colley, '18, Lindfors, ’24, and Pady, ’34, on rusts; Maire, ’08, on divers basidiomycetes; Neuhoff, '24, on the secondary nuclei of Auriculariaceae and Tremellaceae). These connections are still distinct when lighter areas around the now almost structureless chromatic masses give clear indi- cation of the reorganizing daughter nuclei (fig. 52). Gradually a delicate membrane surrounds them. Apparently the chro- matic masses prominent at telophase become the nucleoli of the daughter nuclei, a situation commented upon by Sass (’29) in his studies of Agaricaceae. The two nuclei are usually synchronized in their behavior but now and then one matures in advance of the other (fig. 56a, b,c). A wall separating the two nuclei either is laid down be- fore reorganization or much later. It is distinguishable at first only as a very delicate deposition, barely differentiated from the surrounding protoplasm; occasionally a clearer zone seems to mark its future position. Delay of the wall formation until the nuclei are undergoing the second division is the ex- ception (fig. 56b). As nearly as could be determined the wall is formed by uniform deposition. Frequently a heavier aggrega- [VoL. 23 84 ANNALS OF THE MISSOURI BOTANICAL GARDEN tion is made out mid-way on the spindle strands. This may represent the initiation of the new wall (figs. 52d, 56b). Daughter nuclei from the first division never reach a par- ticularly high degree of reorganization, for the second division follows almost immediately. That this does ensue rapidly is inferred by the fact that early-division stages of it were never seen. Spindle orientation is longitudinal, with only an occa- sional and minor deviation to the oblique. Anaphases with characteristic spindles and darker-staining bodies, the chro- mosomes, are common (fig. 55). In figure 56a the nucleolus of one dividing nucleus has been cast aside and is disinte- grating. The subsequent behavior resembles that of the first- division figures and leads to the formation of four nuclei. Moreover, the spindles are notably smaller than those of the previous division and therefore the chromosome number could not be determined with any degree of satisfaction. An exactly parallel situation has recently been recorded by Pady (735) for the division of the fusion nucleus in the promycelium of Hyalopsora. Ordinarily the two nuclei progress at equal rates, but when they are not synchronous the basal segment may have a nuclear phase as advanced as late telophase while the upper is still in anaphase, or vice versa (fig. 56a). Again a spindle may not have formed in one, although the second is already in anaphase. When all four nuclei are reorganized there results the typical segmented spore-bearing part (fig. 57). In that figure the mid-wall, the first one formed, is clearly older and better developed; the other two have just been laid down. As already noted, a fourth septum is exceptional, but if present it appears later at the base of the fourth segment. Figure 58a shows a true septum but figure 58b illustrates how protoplasmie remnants may simulate it. The four daughter nuclei often display an inequality of size, that of the basal segment being subject to the greatest varia- tion, for it is often noticeably smaller than the others. These nuclei are all in a resting condition. Though their total volume now is slightly greater than in resting nuclei elsewhere, the nucleolus is no longer enlarged. Here, if ever, two nucleoli 1936] BAKER—A STUDY OF HELICOGLOEA 85 may appear, though apparently the condition is evanescent (fig. 59). This is in accordance with the already ascertained fact that nuclei may show variability in this respect, especially after several successive divisions, such as is the case here (Sharp, ’34). That there is great variability in the nuclear and nucleolar volumes all through the cycle is evident (fig. 71). Sterigmata promptly appear as small outgrowths from each segment (fig. 60). They are not formed synchronously nor in any regular order of succession. In most instances the fourth segment is the last to produce its sterigma, but it may be the first segment in which production is longest delayed. Initia- tion commonly starts in the second unit (fig. 61). There is no eorrelation between the position of the hypobasidium and the side of the sporogenous portion which produces the spores, but the spores are borne on the convex side. Sometimes the end segment bears its sterigma apically (fig. 63a); infrequently sterigmata may be borne on opposing sides (fig. 63d). At this time the nuclei are usually to be found directly opposite the budding sterigmata, more often peripheral than central. Gradually the sterigmata lengthen, and then small vesicles, the future spores, appear at their apices (fig. 62). Sometimes the vesicle is preceded by a dark-staining cap or plug whose nature was not determined (fig. 64a, b). Colson (735) finds a similar condition during early sterigmatie pro- duction in Psalliota campestris, where a small quantity of stainable material appears in the narrowest region. She con- siders it merely a mechanical retention of stain. Such would also seem to be the case here. Only the initial spore stages are hyaline, and the expanding spore is henceforth never without a definite protoplasmie content. Protoplasmie migration from the basidium segments begins early and precedes nuclear mi- gration, which does not take place until the spore is practically mature. The latter induces striking changes in the nucleus. While the spore is forming the nucleus becomes decidedly more stainable and completely altered in shape (fig. 65a-d). Finally it becomes very dense, staining homogeneously black, and by then is quite elongate and more or less irregular in outline. It [Vor. 23 86 ANNALS OF THE MISSOURI BOTANICAL GARDEN moves into the sterigma as an undifferentiated compact mass (fig. 66a, b). Often it is long and slender during its progress into the spore (fig. 66e, d). This is not an uncommon phenom- enon in the basidiomycetes. As the nuclear mass penetrates the spore the latter may show vacuolation. By now the proto- plasm is well withdrawn from the segment, the last remnants passing in with, or just after, the nucleus. No protoplasmie lining ever remains in the cells as Buller (722) noted in the Auricularieae. Once within the spore the nucleus undergoes corresponding changes in reverse order until the true resting nuclear state is again attained (fig. 66e, f, g, h,i). The volume is slightly under that of the antecedent resting condition. In diseussing the general morphology, the formation of ex- ceedingly long thickened appendages in place of sterigmata was touched upon. At first it was thought that these might be indieative of direct germination as the nucleus early moved into them without undergoing change. But later it was found that they result directly from the position of the basidium and are merely a means of elevating the spores to the surface. Their consistent occurrence in regions of greater compactness in the fructification bear this out. Sometimes the lower seg- ments of a basidium have these appendages, whereas the upper one produces an ordinary sterigma (fig. 63a). Eventu- ally the appendages produce true sterigmata and spores, ter- minally. This procedure is well known in other heterobasidio- mycetes ; for example, in Auricularia the long appendages push through the gelatinous matrix to bring the sterigmata and spores to the surface. In the rusts normally (Gáumann and Dodge, l. c.) as well as under unfavorable conditions, the ba- sidiospores are produced only when brought to the surface. The nuclei then move into the spores, having undergone the changes described. These nuclear changes may begin early, the initiation point being correlated with the length of the appendage (fig. 67e, f). The nuclear behavior is in complete agreement with Neuhoff's criterion for a sterigma. Linder speaks of variation in sterigmatic length in H. intermedia (see his pl. 41, figs. 13-15), a situation which is probably tanta- mount to the one just described. 1936] BAKER—A STUDY OF HELICOGLOEA 87 Germination, which may take place at once, begins with the emission of a short germ tube and a sterigma upon which a secondary spore exactly like the first is borne (fig. 68b, c, d). This is the so-called ‘‘ germination by repetition,’’ long a recog- nized character of the heterobasidiomycetes (Patouillard, ’87). Not only is the morphology of the sterigma the same but the nucleus, in order to pass into the secondary spore, is altered in exactly the same fashion as characterizes the passage from basidial segment to spore (fig. 69). Nuclear division within the spore has never been seen. At times spores begin germi- nating while still attached to the sterigmata (fig. 68b). No data pertaining to the actual spore discharge were obtained. Attempts to culture H. Lagerheimi have been unsuccessful so far. Unfortunately no fresh material was available, and herbarium specimens of the 1934 collections had to be relied upon. Spores from that source sown in hanging-drop cultures of sterile distilled water gave a goodly percentage of germi- nation within from 24 to 48 hours, but transfers were never successful. From observation of spore behavior in the hang- ing-drop cultures it is readily seen that the usual mode of germination is by repetition. Möller illustrates septate spores, each segment of which is capable of producing a secondary spore. This has not been seen. Although in Patouillard’s type material a spore with two germ tubes was found, septation between was indistinct. The spores do occasionally segment in germination, but their subsequent behavior was not deter- mined. Small sub-globose conidia appear in older cultures. Their mode of origin was not seen. Frequently they are massed, just as Moller reported. Before the appearance of the germ tubes the spores are quite hyaline, in most cases with a single, rather conspicuous oil globule. As germination processes are initiated this divides into two to many with corresponding reduction in size. When the germ tube arises the contents of the spores are distinctly granular with numerous oil bodies. The tubes themselves are mostly lateral, but may be terminal, in position. At their apices a small bubble-like protuberance is the first sign of the [VoL. 23 88 ANNALS OF THE MISSOURI BOTANICAL GARDEN secondary spore. 'This remains quite hyaline until maturity when the content of the original spore is discharged into it. In some cultures within a few days delicate mycelial out- growths were evident (fig. 70). Taxonomic CONSIDERATIONS All the species of Helicogloea are entirely saprobic. They are found on well-decayed woods of various kinds, including pine, alder, birch, beech, oak, aspen, poplar, and willow, rarely on humus or soil. One species occurs on grasses from whence it takes its name, H. graminicola. Collections have been made of some species practically throughout the year; others seem to have more seasonal limitations. Any is more likely to be found in damp rather than dry weather. Geographically, the distribution is scattered. H. Lagerheimi is the most wide- spread. Besides the type locality, Ecuador, it has been re- ported from France, England, and the United States. The only other North American species are H. caroliniana, as yet found only in the type locality, North Carolina, and H. pinicola recently reported from Canada. As pointed out earlier, the chief considerations of the genus have been taxonomic. The most comprehensive treatment is that of Bourdot and Galzin. They characterize the genus as ““floccose or gelatinous-mucous, spreading ; the probasidium in the form of a sac, laterally pendent and giving rise to a ba- sidium with three transverse walls; sterigmata lateral (one terminal) ; spores hyaline, smooth." By reason of the floccose or gelatinous nature of the parent genus, these authors have erected the subgenus Saccogloea to care for the mucose-gelat- inous forms. Furthermore, they comment: ‘‘The French species of Saccoblastia do not always show the generic char- acters clearly; the contents of certain probasidia appear often to be a reserve utilized for the formation of the basidium, but it seems in many cases that it is the probasidium itself which becomes erect and is directly transformed into a basidium. At least, the section shows often all the intermediate stages be- tween the probasidium as a pendent sac and the curved or erect ERRATA Substitute the accompanying plate 11 for that published in Baker’s ‘‘A Study of the Genus Helicogloea’’ in Ann. Mo. Bot. Gard. 23: pl. 11. The following list of errata is also for Baker’s paper. p. 89: under H. gc read—Pl. 12, figs. 72 a-l, 73 a-e. p. 92: under H. read—Pl. 13, figs. 76 a-e. p. 97: first Sears utei paragraph, insert no—From the table following it is elear that there is no p.124: the explanation of idle 12 should inelude figs. 72 and 73, now on p. 126. 1936] BAKER—A STUDY OF HELICOGLOEA 89 basidium.’’ In view of such indefiniteness all obtainable species were studied. A brief description is given of each, chiefly for its morphology. Hzrrcoaroza pinicola (Bourd. € Galz.), comb. nov. Pl. 13, figs. 72a-1, 73a—. Saccoblastia pinicola Bourdot € Galzin, Soc. Mee, Fr. Bull. 25: 16. 1909. The type has not been seen, but two collections of Galzin's, and Bisby’s Canadian gathering were examined. They are all truly floccose. The scattered receptacles of Galzin’s collec- tions are Cartridge-Buff.? A few patches here and there, ge- latinous in consistency, are a deeper color—Olive-Buff to Deep Olive-Buff. The Canadian material shows no gelatinous areas. This species is clearly a member of the floccose section. Bourdot’s and Galzin’s description answers with some ex- ceptions. The hypobasidium was found to be consistently larger than the range given by these authors. They say it is 40-45 x 8-12 »; measures made here extend the range to 19-56 x 7-14 y, the majority in Galzin’s material falling around 50 y in length. The morphological plan is even less in agreement. Bourdot and Galzin refer to variation in the ‘‘basidia’’ but not to any in the hypobasidia. In the material studied the latter is extremely noticeable. Irregularities most frequently take the form of forked or branching hypobasidia (fig. 72a, i, j). Forked basidia mentioned and figured by Bourdot and Galzin have not been found, but figure 72i, for example, strongly sug- gests that these authors may have mistaken forked hypobasidia for forked basidia. The most striking feature is the mode of origin of the basidium parts. Occasionally the primordium is a terminal cell of varying length, as in H. Lagerheim, but usually it is intercalary. The hypobasidium arises in the usual way as a lateral outgrowth but at maturity presents a great variety of forms—from straight to constricted to forked (fig. 72). The size is usually great in comparison with those of other species. The epibasidium is produced as a lateral 2 Capitalized color names are from Ridgway, R., Color Standards and Color Nomenclature. 1912. [Vor. 23 90 ANNALS OF THE MISSOURI BOTANICAL GARDEN out-pushing of the primordium at or near its apex. Hence one commonly finds the basidia appearing in chains in contrast to the type of proliferation common in H. Lagerheimi (fig. 13e). At maturity the spore-bearing region of the epibasidium is three- or four-septate, and the portion posterior to it withers. Apical germination of the terminal segment is common (fig. 72e, k). Bourdot and Galzin say this species is ‘‘very close to H. graminicola, of which it is perhaps only a variety. It differs in habit and the basidia and spores are notably larger." "There seems to be little basis for the assumption of varietal status. True, both species have conspicuous clamp connections and both are supposedly floccose, but the epibasidial character and origin in the two are entirely different, nor does H. graminicola ever show the hypobasidial irregularities so marked in H. pinicola. A collection from Litschauer of this species on Alnus viridis undoubtedly belongs to the form alniviridis, although it was approved by Bourdot before he segregated the form. It shows no deviation from described characters except that there was no paucity of hypobasidia such as Bourdot reported. Morpho- logically it is identical with the species (fig. 73 a-e). The validity of forms founded on substratal attachment is question- able for pure saprobes, even more so in this case since the advent of Bisby’s specimen which was found on Populus. It, too, showed no specifice differences. RANCE: Causse Noir, April 25, 1910, Galzin, 20555 (in Herb. Univ. N. Car. ex Herb. Bourdot); Nov. 19, 1916, Galzin, 2104 (in Farlow Herb. ex Herb. Burt). AUSTRIA: Tirol, Aug. 15, 1924, Litschauer (in Herb. Univ. Iowa). CANADA: Manitoba, Victoria Beach, Aug. 23, 1935, Bisby, F5695. HerLIcocLora graminicola (Bres.), comb. nov. Pl. 13, figs. 74a-d. Saccoblastia graminicola Bresadola, Ann. Mye. 1: 112. 1903. After this is moistened it appears as a thin gelatinous coat- ing on a grass stem, in color Vinaceous-Buff. There is nothing in its structure or consistency to justify Bresadola’s 1936] BAKER—A STUDY OF HELICOGLOEA 91 adjective ““tomentose.”? From phloxine-stained mounts one can readily distinguish its salient characters. Micro-dimen- sions show close correspondence to those given by Bresadola. The primordium is terminal and it gives rise to the epibasid- ium. A conspicuously elongate connection between primordium and sporogenous portion is not found. The hypobasidia are large, coarser in appearance than those of H. Lagerheimi, more constant in size than those of H. pinicola, and are never forked. All the protoplasm withdraws from the hypobasidium but not necessarily from the primordium, which may still show a defi- nite protoplasmic content after the segmentation of the spore- bearing region (fig. 74c). Clamp connections are prominent on the hyphae, sometimes appearing on the basal septum of the primordium which just beyond on the opposite side is pro- ducing the hypobasidium (fig. 74a). Eichler (in Farlow Herb. ex Herb. Bresadola, ex Herb. Patouillard). Heticocuora intermedia (Linder), comb. nov. Pl. 13, figs. 75a-h. Saccoblastia intermedia Linder, Ann. Mo. Bot. Gard. 16: 488. 1929. A slide of the type material prepared by Linder was studied for the morphology of the species. From this it was clear that the origin of the saccate hypobasidium is intercalary, not ter- minal, or only exceptionally so. Following nuclear fusion within the hypobasidium, the latter germinates proximally to the epibasidium. After great elongation four sterigmata and spores are formed on the thickened end. Linder uses the terms ““probasidium, promycelium, and basidium”” for these parts. However, there is no reason, in spite of morphological varia- tion, why Neuhoff’s terminology is not applicable. The clavate type simply represents direct development from a terminal primordium. In this event, when a lateral sac is not produced, fusion takes place in the clavate portion itself; hence it is to be interpreted as the hypobasidium and its subsequent outgrowth as the epibasidium. Figure 75 is included merely for com- parison with the other species. [VoL. 23 92 ANNALS OF THE MISSOURI BOTANICAL GARDEN Linder speaks of the saccate hypobasidium exhibiting distal germination at times. If his figures (pl. 40, fig. 4; pl. 41, fig. 9) represent the best evidence for that phenomenon, its occur- rence is questionable. Succeeding stages would be convincing but none was found. However, the hypobasidia are frequently constricted, once or more, and it is possible that this so-called distal germination is only an extreme constriction. Distal germination was never seen in H. Lagerheimi. Hypobasidia of the form illustrated in figure 75e, H. intermedia, might be in- terpreted as unusually constricted, especially when compared with figure 18f and g of H. Lagerheimi which is so construed. Furthermore, if the orientation of the two fructifications is the same, then distal germination could only occur for those rare cases where the hypobasidia are lateral or pendent. CuBA: Soledad, Sept., 1924, Linder, TYPE (in Herb. Mo. Bot. Gard.). HxzrrcoaLoEA caroliniana (Coker), comb. nov. Pl. 13, figs. 76a-c. Saccoblastia caroliniana Coker, Jour. Elisha Mitchell Sci. Soc. 43: 233. 1928; S. ovispora Moller var. car- oliniana Coker, Jour. Elisha Mitchell Sci. Soc. 35: 121. 1920. Coker's descriptions are so adequate that little need be said here. The fructification is a particularly well-developed gelat- inous one. In its morphological behavior it resembles H. Lagerhewni very closely. A series of diagrams is included for comparison. Here, as in the other species of the genus, long appendages may precede the formation of sterigmata and spores (fig. 76d). NorTH CAROLINA: Chapel Hill, April 15, 1920, Couch, 4256 (in Herb. Univ. N. Car.) ; Chapel Hill, July 31, 1920, Couch, 4601 (in Herb. Univ. N. Car.). HzrrcoaroEA LaGERHEIMI Patouillard, Soc. Myc. Fr. Bull. 8: 121. 1892. Pls. 7-12; pl. 13, figs. 77-78; pl. 14. Saccoblastia ovispora Moller, Protobasidiomyceten, Bot. Mitt. 8: 16. 1895. Saccoblastia sebacea Bourdot € Galzin, Soc. Myc. Fr. Bull. 25: 15. 1909. 1936] BAKER—A STUDY OF HELICOGLOEA 93 Saccoblastia sebacea var. vulgaris Bourdot & Galzin, Hym. de France, p. 5. 1928. Saccoblastia sebacea var. pruinosa Bourdot E Galzin, Hym. de France, p. 5, 1928. A study of the type material resulted in the above changes in synonymy. Figures 79-81 show the development of the ba- sidium as seen in Patouillard's material, from the initiation of the primordium to the mature basidium and spores. Though micro-dimensions are consistently larger than those from the majority of collections, they do not exceed the uppermost figures obtained. There is no uncertainty about the identity of this material with Móller's or Bourdot's and Galzin’s. Therefore the earliest name takes precedence. On the basis of basidium size H. caroliniana is perhaps the nearest species. But the two are so totally different in their habit they could never be confused. In particular, the hypobasidia of the type collection are sometimes constricted and frequently are set farther forward in relation to the basal wall of the primordium (fig. 79d) than usual. No ease of intercalary origin was noted in this material. Saccoblastia sebacea B. & G., an acknowledged gelatinous form, has been separated from S. ovispora Moller largely be- cause the latter was supposedly floccose or hypochnoid. A study of all available specimens and descriptions has been made in an effort to straighten out the situation. Of Helicogloea Lagerheimi (Saccoblastia ovispora Moll.), Moller wrote: **Sie bildete einen dünnen, fast durchsichtigen lockeren weissen Ueberzug, der in ganz unregelmássiger Umgrenzung mehrere Centimeter in jeder Richtung sich ausdehnte. Bei sehr feuchtem Wetter sieht dieser Ueberzug fast schleimig aus, da das Gewirr der Fáden Wasser zwischen sich festhált, bei trocknerem Wetter dagegen bemerkt man nur einen lockeren Hyphenfilz, der bei vollstandigem "Trocknen zur Unsichtbarkeit zusammenfállt.”? (Lei, Ferry (796) the following year translated Móller's work into French, in addi- tion utilizing Móller's plates. The structure of the fructifica- tion is described as irregular patches of white loosely inter- woven hyphae. Lindau (797) introduced the adjective ‘‘wer- [Vor. 23 94 ANNALS OF THE MISSOURI BOTANICAL GARDEN gartig’’ in his characterization. As yet only Móller's species were known. Patouillard (’00) delineated the genus Sacco- blastia as floccose, soft, and said of the tribe Septobasidiés to which he assigned it: ‘‘Floconneux ou erustacé, non gélati- neux." Chronologically Bresadola’s (’03) report is next, which described H. graminicola as ‘‘tomentose.’’ Clements (’09) placed the genus (as Saccoblastia) in that group of Auriculariae having a byssoid pileus. Clements and Shear (’31) adhere to this characterization. Saccardo (’82) qualified the structure by saying, *«Est quasi Hypochnus sporis pleur- ogenis." Killerman (’28) repeated Lindau’s diagnosis for two species, having replaced H. sphaerospora with H. gramini- cola. The character of floccose as opposed to gelatinous is the distinguishing one for two sub-genera in Bourdot and Galzin. Thus there had gradually crept into the literature the con- ception of floccose or hypochnoid fructifications in Helicogloea. H. pinicola is truly floccose. The interpretation of H. Lager- heimi as floccose probably has been considerably strengthened by Bresadola's comment to Coker in reference to H. carolini- ana which was first described as a variety of H. Lagerheimi. Coker says (1. c.): ‘‘I later sent some to Bresadola who has described a sp. of Saccoblastia and seems to have seen S. ovispora. He writes as follows: *For me it is a species quite distinct from S. ovispora. The genus Saccoblastia is not ge- latinous ; your species is gelatinous like Platygloea; . . .' " An exactly parallel situation has arisen in the case of Stypella minor Móller and Tremella gangliformis Linder. As Martin (734) points out, the error is not in Móller's original description but in subsequent misplaced emphasis and conse- quent distortion. Since the type of Helicogloea Lagerheimi is mucous-gelatinous, all necessity of maintaining these species separately is removed. Unfortunately, Móller's type speci- men has not been located; Rick's material, agreeing well with Moller’s descriptions and coming from an adjacent province, gives strong likelihood of being the same fungus. The type of Bourdot's and Galzin’s species likewise has not been seen, but all the colleetions so labelled which have been at 1936] BAKER—A STUDY OF HELICOGLOEA 95 ` hand were decidedly not floccose. When any of these specimens is moistened it is mucous-gelatinous, but as it dries one can trace just such a sequence as Moller described. Throughout the genus there seems to be not only inconsist- ency of interpretation but a real variation in the nature of the fructification itself. Wakefield and Pearson (’23) note of H. Lagerheim that a series of specimens shows the ‘‘mucous- gelatinous’’ texture only in later stages. ‘‘When at its best the fungus has somewhat the consistency of Corticium confluens, and is beautifully pruinose with the projecting basidia and spores." Rick (personal correspondence, 1934) says: ‘‘The Saccoblastias are surely hypochnoid without true hymenium, but when well developed they are gelatinous, like Exidiopsis, and dry a little glancing.”? To Rick obviously hypochnoid con- notes structure rather than consistency. On that basis all the species are ““hypochnoid.”” It is significant that he con- siders the genus gelatinous, even though earlier he had been in accord with conventional treatments when in a monograph of protobasidiomycetes (733) he referred the genus to the floc- cose Auriculariales. As already remarked, H. pinicola, which is distinetly tomentose for the most part, has some scattered gelatinous patches. Móller”s statements may have left latitude for differing in- terpretations, but his figures are clear-eut and easily show the close resemblance to H. Lagerheimi. A table of measure- ments for the specimens under diseussion shows there is no reason to maintain any as distinct species because of variation in size. Only in the event that Móller's type proved to be dis- tinctly floccose would it be reasonable to retain that species. Two collections from Brazil were studied, both Rick’s. The first of these was quite inconspicuous until moistened, when it became visible as thin gelatinous film, faintly pustular, of a Buffy-Brown color; no part of it could be interpreted as floc- cose. Unfortunately, the specimen was so contaminated with hyphomycetes that mounts were far from satisfactory, the more so as it did not seem to be in a particularly advanced fruiting condition. However, all evidences indicated a struc- [Vor. 23 96 ANNALS OF THE MISSOURI BOTANICAL GARDEN tural development much like the type of H. Lagerheim. The mycelium is typical; oblong-ovoid hypobasidia—straight or constricted—are borne laterally on terminal primordia (fig. 78a). There is rarely deviation from the typical epibasidial production (fig. 78e). The second of the collections may easily be a portion of the first, to judge from appearances. Certainly when examined before moistening it is not at all floccose. If discernible at all when dry it is only as a dried mucoid layer spread thinly on the bark substratum. As it revives in a moist-chamber it quickly swells into a mucous-gelatinous layer, exactly like that of the first collection. This material offered a greater range of ba- sidia, though mature ones were scarce. Very infrequently basidia of interealary origin were found. The hypobasidia are often eonstrieted. Figure 78 a-g outlines the structural plan for both collections. Examination of the other herbarium specimens suggested emphasis of the following points. Usually the primordium of the basidium is of terminal origin, rarely it is intercalary. The hypobasidia are straight or constricted, but in a few collections irregular ones approaching the H. pinicola type occur. Micro- dimensions show a greater range than those given by Bourdot and Galzin, but they are fairly consistent for the particular collection. All the fructifications are decidedly gelatinous with a color range from Ochraceous-Buff to Fuscous, most com- monly Grayish-Olive. Bourdot and Galzin have included several varieties. Op- portunity to examine a large number of collections of the species and the varieties pruinosa and vulgaris from Bourdot’s herbarium showed that the varietal segregates are untenable on both habital and structural bases. According to Bourdot and Galzin, the variety vulgaris differs only in that it is ‘‘thinly spread in large gelatinous patches." When moistened it so appears; its color is pallid Cinnamon-Buff or Grayish-Olive. There is scarcely anything distinctive about this; certainly in structure it is identical with the parent species as shown by figure 77. 1936] BAKER—A STUDY OF HELICOGLOEA 97 The variety pruinosa is segregated by these same authors on the bases of micro-dimensions and growth form. The meas- ures of the hypobasidia in the material examined slightly exceed those cited by Bourdot and Galzin. Spore size extends from 8-11.5 x 5-6.5 p. Though the spore size of the parent species ranges from 8-18 x 4-9 y, the majority fall in a mid- dle grouping around 9-11 x 5-7 p. It would be impossible to make a separation on spore size. The miero-dimensions do not distinguish the variety from the parent species; morpho- logieally there is no difference between the two, and the color range is also closely parallel. No illustrations are included. From the table following it is clear that there is more reason to set aside separate varieties than to maintain the former species because of differences in measurements. Such a criterion is hardly reliable in a group as variable as the heterobasidiomycetes. HELICOGLOEA LAGERHEIMI PAT. Hypobasidium O Hyphae “sac?” Epibasidium Spores Type 2-6 u 18-38 x 5.5—10 469—100 x 5-7.5 u| 13-15-18 x 6.5-8 u Moller's 2-6 y 30 x Bu 100 o 13 x 7-9 u measures Rick’s colls. 2-6 y 18-33 x 6-7.5 y 65 x 4.5 u 15-18 x 6-7.5 u B.’s & G.’s 2-6 u 18-30 x 6-9 y | 45-75 x 46.9 y 8-10-15 x measures 5-6-8 pu Other specimens 2-6 u 15-39 x 5-13 4| 58-105 x 5-9 u |8-12-17 x 4-9 u var. vulgaris 2-6 u 18-28 x 5-8 y | 50-60 x 5-6 y | 9-11 x 6-7 u vir. pruinosa 2-6 u 16-32 x 5-9.5 u| 50-60 x 5-6 y |8-11.5 x 5-6.5 u Roger’s meas- 34.546 y 16-32 x 7-10 u 75-100 x 11-14.5 x 4.5-7.5 u 5-7.5 u Total range 2-6 y 15-39 x 5-13 nl 45-105 x 4-9 u| 8-18 x 4-9 u ECUADOR: Chorrera de Agoyan near Baños, Jan. 1892, Lagerheim, TYPE (in Far- low Herb. ex Herb. Patouillard). BRaziL: Rio Grande do Sul, Sao Leopoldo, 1931, Rick (in Farlow Herb.); 1931, Rick (in Herb. Mo. Bot. Gard.). UNITED STATES: Iowa, Iowa City, Oct. 5, 1932, Rogers, 242; July 8, 1934, Rogers, [Vor. 23 98 ANNALS OF THE MISSOURI BOTANICAL GARDEN $05, 306; sr 9, 1934, Rogers, 274; Missouri, Springfield, Oct. 2, 1933, Looney (all in Herb. Univ. Iow FRANCE: Allier, ee 5750 (in Farlow Herb. ex Herb, Patouillard) ; Bourdot, 5750, 5882, 39135 (in Herb. Paris ex Herb. Bourdot) ; Aveyron, Galzin, 3895, 3951, 4334, 5749, 7856, 7961, 10085, 11982, 13816, 13817, 14003, 14247, 14270, 14368, 14411, 14606, 14764, 14790, 14800, 14825, 14828, 16605, 16632, 16638, 16641, 16646, io 16704, 16938, 17810, 17454, 18828, 18899, 19067, 20134, 21132, 23717, 23718, in Her 3305, 6017, 6508, 7612, , 23288, 23289, 23240, 12151, m 12349, RE ME 23291, 23308, 23333, 23293, 12348, 12411, 23294, 23295, 23302, 23297, 23296, 23300, 23303, 23301, 23304, 23298, 23335, 14915, 15419, 15479, 15991, 18561, ..... , 24120, 24121, 24122); St. Guirol, May 6, 1910, Galzin, 5815 (in Herb. Paris, 23287 ex Herb. Bourdot); Tarn, July 23, 1909, Galzin, 4316 (in Herb. Paris, 23306 ex Herb. Bourdot); April 10, 1916, Galzin, 19561 (in Herb. Paris, 2331 ex Herb. Bourdot) ; St. Priest-en-Murat, Nov. 1914, Bourdot, 42262 (in Herb. Univ. N. Car. ex Herb. Mrd ; Orne, Oet. 1925, E. Gilbert, 1313 (in Herb. Paris, 39816 ex Herb. Bour ENGLAND: London, Feb. m Pearson (in Herb. Paris ex Herb. Bourdot). Discussion Mitotic processes of Helicogloea give no indications of ab- normalities, although a few stages encountered need some ex- position. The unusual spindle disposition in nuclear division is by no means restricted to this form. Kniep (’13) noted a similar irregularity in Hypochnus terrestris. His figures in- clude a definitely crossed type and one slightly angled, both from a germ tube; and the usual conjugate type from the mycelium. This gradation is comparable to that found in the terminal hyphal cells of Helicogloea. Tischler (l. c.) adopted Kniep’s figures; neither author offers an explanation. Neu- hoff (’24), throughout his studies on Auriculariaceae and Tremellaceae, saw mitosis only once—in Achroomyces Tiliae. Quite SE his illustration denotes that the spindles were crossed, but of this he makes no mention. However, he regards as abnormal crossed spindles in the hypobasidial division in Exidia pithya. This same phenomenon has been demonstrated frequently in the divisions of the fusion nucleus and in one instance of nuclear division in the spore (cf. Rogers, '32; Lander, 735; Colson, '35). That the clamp connection is a device to Domit the side-by-side orientation of the spindles for the conjugate division of the dikaryon and a means to bring non-sister nuclei together, is widely accepted. Since there are 1936] BAKER—A STUDY OF HELICOGLOEA 99 no clamp connections in this species, it is perhaps reasonable to assume that the crossing of the spindles is imposed by the nar- row limits of the cell diameter. Sass (’29) has commented upon the fact that apparently the chromatin masses at telophase become the nucleoli of the daughter nuclei. This is quite evidently so in the meiotic di- visions of Helicogloea where a complete sequence can be traced easily from late anaphase to the reorganizing daughter nuclei, still joined by the drawn-out spindle threads. To ex- plain this it is necessary to assume that the true nucleolus is distinct from this chromatic mass. This, then, would be in line with the second of the explanations offered by Sass for varia- tion in chromosome and reticulum behavior among the several species that he was studying. In support of this hypothesis, that the chromatin is localized or stored in the large ‘‘nucle- olus-like body’’ which therefore is to be considered a karyo- some from which the chromatin passes into the achromatic framework during pre-meiotic processes, he presents four lines of evidence. Nuclei of Helicogloea are too small to permit confirmation. Sass’s final statement, that the true nucleolus is a minute body expelled from the chromatin body just before spindle formation, offers a clue to what is evident in Helico- gloea. As figures 45, 48 (cf. earlier stages also) show, there is a decided decrease in nucleolar size during prophasie periods evident from the first, and after the spindle and chromosomes are definitely formed a minute body is still visible, often some distance removed from the spindle. Later it too disappears. Though this cannot be conclusive without tracing the origin of the reticulum and chromosomes, nevertheless the behavior is suggestive. It seems to occur commonly in many fungi; Wakayama (730) has promoted similar ideas based on observa- tions of Aspergillus. One of the most curious and striking of aberrations con- cerns the chromosomes in their passage to the poles. This consistently is irregular, so that from metaphase through anaphase they are scattered from equator to pole, sometimes more or less clumped together. Throughout the literature [Vor. 23 100 ANNALS OF THE MISSOURI BOTANICAL GARDEN this condition is commonly figured (cf. Juel, '98; Ruhland, ’01; Kniep, '13; Colley, "18: Bagchee, ’25; Wakayama, ’30, ’30a, '81, ’32) not only for basidiomycetes but for ascomycetes as well. Wakayama’s extensive works, especially on Agari- caceae, show that this phenomenon is more characteristically present than a regular separation of the chromosomes. His comment that this scattering is due to very different migration velocities for each univalent is the only stated attempt noted in elucidation of the point. He furthermore remarks that in spite of this unusual procedure it nowise follows that the nu- clear process is not typical. Even more striking than this is the behavior of the nuclei in their migration from the parent cells or accessory appendages into the spores. Their extreme chromaticity, attenuation, and loss of form mark the narrow constricted portion through which they must pass just prior to entering the spore as a true sterigma, sensu Neuhoff. Such nuclear alteration has been remarked time and time again in the basidiomycetes. Maire (702) noted that in the heterobasidiomycetes the nucleus passed into the germ tube without change but in moving into the spore through the sterigma it became constricted. It re- mained for Neuhoff, however, to crystallize the phenomenon into a definite criterion. According to his dictum, cytological in its foundation, a migrating nucleus in passing through a sterigma into a spore assumes just such an elongated form, at the same time giving a chromatin stain reaction. It is tenable then that the long inflated appendages so frequently developed in place of the shorter subulate sterigmata are accessory. This is strengthened still more by the fact that when they reach the surface or a position favorable to spore discharge true sterig- mata are formed and the nuclei go through them into the spores in the changed form. That such attenuation is not the re- sult of mechanical constriction due to the size of the sterigma is upheld by the initiation of the condition long before the nucleus reaches that portion of the sterigma (cf. Whelden, '34, on Tremella). Spore germination processes are initiated by the production from the primary spore of filaments of varying 1936] BAKER—A STUDY OF HELICOGLOEA 101 length. This in turn develops a true sterigma through which the nucleus moves into the secondary spore, having previously undergone precisely these specified changes. The sterigmata in relation to the spores by all evidence fit Buller's (’22) func- tional criteria, viz., that ‘‘the typical sterigma ...is... an organ for the violent discharge of the spores.’’ The hilum re- gion of the Helicogloea spore is well developed. Observation of the actual mechanism of discharge was impossible without fresh material. A lack of consistency in the terminology employed for the basidiomycetes and the interpretation of their diagnostic struc- ture, the basidium, has long marked the literature, the situation only recently becoming clarified and more constant as careful morphological and cytological studies have brought out true and natural relationships. Helicogloea is classified with those basidiomycetes which have transversely septate basidia as opposed to those in which the basidia are vertically divided or those which remain undivided. Historically, the basidio- mycetes have been separated into Wei subdivisions on the basis of basidium morphology, e. g., Hymé ycétes-Hétéro- basidiés and Homobasidiés (Patouillard, '87); Proto- and Autobasidiomycetes (Brefeld, '88); and Acrosporeae and Pleurosporeae (van Tieghem, 93). As Gáumann and Dodge (1. c.) point out, the first two complements carry a phylogenetic implieation and the last is neutral in its connotation. "These categories do not comprise identieal groups. Van Tieghem considered septum formation an accessory character, the fundamental one for him being the mode of spore insertion, ex- pressed by his terms Acrosporeae and Pleurosporeae. These primary divisions each included both septate and non-septate basidia, which he spoke of as Phragmo- and Holobasidia. Patouillard’s groupings have a less artificial basis. Further- more, he recognized in distinguishing between hetero- and homobasidial groups that the spores of the former usually germinate by the production of secondary spores, whereas those of the latter germinate by mycelia. Such a distinction is a more reliable criterion for primary segregation of the ba- [Vor. 23 102 ANNALS OF THE MISSOURI BOTANICAL GARDEN sidiomycetes than the character of septation. In recent years the use of the terms heterobasidiomycetes and homobasidio- mycetes has found acceptance. The heterobasidiomycetes in- clude those forms with secondary spore germination and basidia usually divided in some fashion—the notable exception being the undivided basidia found in the Dacrymycetaceae. At its earliest inception the basidium of Helicogloea, or that of almost any heterobasidiomycete, is not distinguishable from one of a homobasidiomycete at a comparable level of develop- ment. Both are usually single, terminal hyphal cells contain- ing a pair of nuclei, descendants of the dikaryon of the sub-terminal cell. From this point onwards the resemblance is diminished in a greater or lesser degree. The true homoba- sidium remains undivided and develops no accessory parts. Therefore it functions both as zeugite and gonotocont. Vari- ation in some of the heterobasidiomycetes is founded on the separation of these critical stages and consequent modification of morphological patterns. In Helicogloea the initial cell of the basidium complex has been designated as the primordium, a term not inappropriate for the same stage of the homobasidium. From here on the similarity ceases. The subsequent development in the hetero- basidiomycetes, and in particular in the genus under discussion, involves the production of distinct and often persistent morphological entities. For Helicogloea these take the form of the ‘‘sac’’ and of spore-producing out-growths. Moller called the former ‘‘tragzelle,’’ thus merely indicating its position and form with no intimation of its cytological significance. Van Tieghem’s terms—probasidium and basidium—have been more widely adopted by later workers. Their connotation does not permit the probasidium as part of the basidium proper and hence introduces a misleading idea of their true relation to the nuclear functions. For this reason they are re- jected here. Probasidium, promycelium, and basidium have also been applied to the different parts of the basidium, especially by those who were particularly impressed by the striking resemblances of the group to certain Uredinales. 1936] BAKER—A STUDY OF HELICOGLOEA 103 These terms are rejected on the same basis as the preceding ones. A cytological interpretation is credited to Wettstein (*11), who emphasized the fact that in the basidium the fusion of two nuclei takes place before the formation of the spores. As Rogers ('34) has stated, the most satisfaetory concept of the basidium—for both homo- and hetero-types—is Neuhoff's (I. c.). Itis basically cytologic, for he considers the morpho- logical development consequent upon that. According to that author, the mature basidium of the heterobasidiomycetes is composed of a hypobasidiuwm which has ‘‘germinated’’ to one or more appendages, the epibasidia. This terminology allows adequately for any morphological separation of karyogamy and meiosis. If both these critical events take place in the hypobasidium several epibasidia result, usually four, some- times eight, consequent upon the number of nuclei produced— as in the Tremellaceae and Tulasnellaceae, respectively; if meiosis is delayed until after germination there is produced, naturally, only one epibasidium. In the latter category are placed the Auriculariaceae, and, among them, Helicogloea. It is a simple matter to reconcile the situation in that genus with this conception. Neuhoff defines the hypobasidium as the ‘‘lower, commonly inflated vesicular part of the basidium””; the epibasidium as the ‘‘upper, more uniformly tubular part of the basidium.’’ The two parts are in the majority of instances in free communication at all times. If this ground-plan is ap- plied to the behavior in Helicogloea all the species resolve neatly into that scheme. That is, the ‘‘sac,’’ a lateral expan- sion of the primordium, is the place where the two nuclei come together and fuse, and thereforeitis truly hypobasidial. From the primordium is produced apically a single appendage in which the secondary nucleus undergoes meiosis, the whole eventually elongating and bearing sterigmata and spores. This quite properly is the epibasidium. The two parts are never separated but are at all times in open communieation. Even for those species in which the primordium is intercalary this holds. It is evident then that there is nothing in Neuhoff’s [Vor. 23 104 ANNALS OF THE MISSOURI BOTANICAL GARDEN eriteria to exclude their general application to Helicogloea. At first there may seem to be a morphological diserepaney which would prevent their literal adoption. Apparently only those forms with direct germination of the primordium without an accessory part—and the clavate hypobasidium of H. imter- media is the only one which answers that deseription—seem to be completely in harmony with Neuhoff's principles. How- ever, it is only necessary to include for the others the primor- dial portion as part of the hypobasidium, which therefore is to be deseribed as a hypobasidium with a lateral expansion. In the light of this interpretation the production of the epibasid- ium is seen to be from the hypobasidium, whatever the constitu- tion of the latter (cf. H. pinicola, H. intermedia, and H. Lager- heimi). There is good morphological justification for this. The pri- mordium and sac are plainly complementary parts of one structure, and the epibasidium, an outgrowth from either part, is filled with protoplasm at their expense. For reasons of con- venience the term primordium is retained to cover the primary phase of the hypobasidium; likewise, the accessory develop- ment where karyogamy occurs is referred to as the hypobasid- ium and in formation is secondary. This is the only example so far known in which there is such an obvious morphological separation of the initial portion of the hypobasidium from that in which karyogamy occurs. One minor deviation from Neu- hoff’s nuclear scheme must be noted. The secondary nuclei of those Auriculariaceae which he investigated characteristi- cally began migration into the epibasidium during the post- synaptic stage of the first meiotic division. In Helicogloea the fusion nucleus contracts to a typical resting condition and then migrates into the epibasidium where it undergoes meiosis. In other words, not only is the hypobasidium divided into two distinct units, a condition not known in other Auriculariaceae, but the nuclear cycle is interrupted by a resting stage—even though it is so short it clearly is only to allow for migration. The hypobasidium of Helicogloea has often been interpreted as a storage organ, an ecological view which Möller inaugu- 1936] BAKER—A STUDY OF HELICOGLOEA 105 rated. His statement that the contents of the hypobasidium would be withdrawn completely and migrate into the develop- ing epibasidium was based on accurate observation, it is true. The idea of a storage organ, unique for this genus, was again expressed by Gáumann (722). But obviously the time relation between fusion and migration of the secondary nucleus (it con- tracts and rests only long enough to orient itself for emigra- tion) precludes such an interpretation. One must admit, though, that by virtue of this behavior there is introduced the potentiality for a longer resting period. That the hypobasid- ium is not a storage organ, in which reserves are held in readi- ness to produce the epibasidium when favorable conditions so permit, is further substantiated by the observation that as often as not epibasidial germination and elongation occur without any visible augmentation from the hypobasidium. The complementary parts of the hypobasidium are to be con- strued, then, only in the light of the nuclear cycle. Within the genus itself there is manifested in basidial char- acters a progressive sequence from simpler to more advanced forms. Quite obviously, the least developed type would be the clavate form of Helicogloea intermedia. The basidium arises as a terminal primordium, becomes swollen to accom- modate karyogamy, and then germinates directly to the epi- basidium. Here, then, is the simplest type of hypobasidium found in the group, for it consists of only one morphological entity. It is significant that, according to Linder’s notations, this type of hypobasidium is the first produced. Since there is a saccate type of hypobasidium in the same species, the production of the clavate type takes on a vestigial aspect. Rogers (’34), going on the assumption that the ‘‘primitive auriculariaceous basidium is one with a persistent hypobasid- ium, clearly distinguishable in all stages,’’ uses as an example the saccate Helicogloea hypobasidium showing distal germina- tion. Such behavior is decidedly atypical if it occurs at all. The clavate form is much more acceptable as the primitive type of auriculariaceous basidium, for its hypobasidium is a sharply delimited, persistent organ. Second, there may be placed the [Vor. 23 106 ANNALS OF THE MISSOURI BOTANICAL GARDEN type characteristic of the majority of species, H. Lagerheimi, H. graminicola, and H. carolimiana, whose basidial primordia likewise are terminal, but which give rise to a secondary part, morphologically distinct. Advance is also indicated by the fact that the clavate type shows no adjustment for proliferation, whereas in the second type there is ample allowance through the subterminal cell development. The remaining species are all marked by this division of the hypobasidium into primary and secondary portions. H. pimicola shows an entirely new de- parture in the intercalary origin of the basidium. But never- theless the primary portion of the hypobasidium remains the source of germination. (Note that infrequently this pattern is found in the preceding species, too). In the saccate type of H. intermedia it is the secondary portion of the hypobasidium which germinates. This possibly represents the highest level, for it suggests a potentiality for a resting stage, and, through eneystment, eventual separation. In all types meiosis occurs in the epibasidium. It is evident that the hypobasidium, and in partieular the secondary portion thereof, is the critical organ of the genus. Cytologically this has been shown to be the place of karyogamy followed by contraction to the resting condition. Phylogeneti- cally it offers two lines of interpretation: either reduced or in- creasing development. Both ideas are extant in the literature as applied to relationships within the Auriculariaceae, and be- tween them and other groups. In the Auriculariaceae, Neuhoff constructs a phylogenetic scheme involving both tendencies. He considers the genus lola the crucial one, for there the basidium arises from a terminal cell and is divided into hypobasidium and epibasidium, the former remaining distinct. This gives rise to two main lines, one showing reduction in differentiation of the hypobasidium, the other increasing emphasis. The first of these is subdivided into a series in which reduction eventually leads to complete suppression, as illustrated by the genus Helicobasidium; and another in which the hypobasidium is present though it may be reduced to a vestigial condition, as read in the series 1936] BAKER—A STUDY OF HELICOGLOEA 107 Achroomyces, T jibodasia, and Auricularia. The second line emphasizes the hypobasidium through the thickening of its walls and the introduction thereby of a resting period. This plan of development is illustrated by the genera Cystobasid- ium and Septobasidium. In the latter there are especially critical characters, for the thickening of the walls in some species is so advanced that in view of that and a consequent resting period, the hypobasidium is known as a sclerobasidium. This condition then is strongly like that of the rust teliospore. Indeed, the only real basidial difference between this and such primitive rusts as Gallowaya is the catenulate nature of the teliospores of the latter. Consequently, Neuhoff sees here a direct link to the Uredinales through such a genus, for instance, as Uromyces, whose teliospores are borne singly on the hyphae. Hence the sclerobasidium (encysted hypobasidium) and the teliospore are homologous. That there is a close relationship between the Auriculari- aceae and the Uredinales is argued by more than mere analogy. Cytologically it is demonstrable that the teliospore and hypo- basidium are strictly homologous; likewise the promycelium and epibasidium; sporidia and basidiospores. Moller was the first to comment on this, but to him the hypobasidium of Helico- gloea and the teliospore were homologous because both were the assembling point of materials necessary for basidium pro- duction. Since in the former no resting period was involved there was no need for a heavy wall. This ecological interpre- tation has since been replaced by the more substantial cyto- logical one. Moreover, the ecological argument lost credibility as investigations revealed that the teliospore is not neces- sarily designed for wintering over, for there are rusts whose ` teliospores germinate at once. Again there are rusts whose teliospores are thin-walled, just as in Septobasidium of the Auriculariaceae, where the development passes from species with thin-walled, non-resting hypobasidia—hardly more than a stage in development—to thick-walled, persistent forms which rest (winter-over) before germinating. Such striking parallelisms emphasize the closeness of the auriculariaceous fungi and the rusts. [Vor. 23 108 ANNALS OF THE MISSOURI BOTANICAL GARDEN The majority of writers have adhered to de Bary's ('84) tenets that the Uredinales constitute the most primitive of the basidiomycetes. Recently Linder (729) and Pady (733) have stressed this by deriving the auriculariaceous forms from the Uredinales. Thus Linder, starting with the Uredinales would, by reduction of the hypobasidium, lead the develop- mental line through Septobasidium and Cystobasidium to Iola with secondary branches to H elicogloea and Auricularia. Sep- tobasidium holds the closest connection with rusts, since it is parasitic. It stands as a source of further development by reason of two different hypobasidial expressions: thick-walled forms which are adapted for wintering-over and thin-walled forms which have no resting period prior to germination and which really are no more than a stage in development. The latter affords a link to Zola, a parasitic form whose thin-walled hypobasidia likewise germinate at once. Helicogloea has lost both the parasitic habit and the thickened hypobasidia. Auric- ularia is even more reduced, for there the hypobasidium is rep- resented only as a stage. Specifically, by virtue of the clavate type of hypobasidium, Linder regarded Helicogloea inter- media as the form transitional between the rusts and Zola. To achieve such a step from rusts to auriculariaceous species, there must be postulated the concomitant factors—reduction of aeciospores to binucleate conidia, development of an exten- sive fruiting body, and finally a change from parasitic to saprobic habit. In opposition to such a derivative scheme it may be con- tended that it is more logical to derive aeciospores from bi- nucleate conidia. This Neuhoff has expressed when he homologized the aeciospores with diploid conidia known in the Auriculariaceae and the pycniospores with haploid conidia. Thus the variety of secondary spore forms in the rusts cannot be said to constitute an obstacle to relationships between the two groups in light of the occurrence of both haploid and diploid conidia in the Tremellales. Especially does the argu- ment of transition from parasitic to saprobic habit seem untenable. As Janchen (’23) has remarked, an attempt at 1936] BAKER—A STUDY OF HELICOGLOEA 109 phylogenetic elucidation must be not only morphologically ac- ceptable, but ecologically comprehensible as well, if it is to be satisfactory. And exactly in its ecological expression does this derivation of Auriculariaceae from Uredinales seem to fail. The high specialization attained by the rusts much more clearly marks them as an end term of a series than as a deriva- tion point. The reverse derivation—the rusts from auricu- lariaceous forms—is not inconceivable ecologically when one considers the numerous examples of parasitic species of Auriculariaceae. The foregoing discussion is intended only to summarize some of the more outstanding phylogenetic interpretations in the literature. In view of the greater defensibility of the deri- vation of the rusts from the Auriculariaceae, that plan is up- held. Helicogloea is regarded therefore as representing a simple auriculariaceous type showing tendencies which could lead to the devolpment of an independent hypobasidium and hence to a true teliospore. SuMMARY 1. Helicogloea Pat. takes precedence over Saccoblastia Moll. with corresponding transfer of recognized species. Sac- coblastia ovispora and S. sebacea are both regarded as syno- nyms of Helicogloea Lagerhevm Pat. 2. In Helicogloea Lagerheimi Pat. the basidial primordium is a two-nucleate hyphal cell, producing a lateral expansion. 3. The dikaryon migrates into the expanded portion of the hypobasidium where fusion takes place. 4. Following contraction to the resting condition the sec- ondary nucleus migrates through the primordial portion into the elongating epibasidium, which has arisen apically from the primordium. 9. The two meiotie divisions always take place in the en- larged portion of the epibasidium. The latter segments into four cells and each segment produces a sterigma and a spore or an aecessory appendage bearing a sterigma and a spore. 6. Spore germination is by repetition. [Vor. 23 110 ANNALS OF THE MISSOURI BOTANICAL GARDEN 7. Morphological development is various in relation to the nuclear cycle; it has been found to correspond to one of three patterns, designated as I, II, and III. 8. The basidium complex is composed of hypobasidium and epibasidium, sensu Neuhoff. The former comprises two mor- phologically distinct parts: the primordium, or primary por- tion, and its lateral expansion, the ‘‘sac,’’ or secondary portion. 9. Within the genus there is seen a progressive sequence from simpler to more advanced forms of basidia, the clavate hypobasidium of H. intermedia being interpreted as the most primitive. This study was done at the Henry Shaw School of Botany of Washington University. The writer takes pleasure in acknowledging the facilities thereby extended, and the assist- ance of Dr. C. W. Dodge, under whose direction the work was done. Especial acknowledgments are due Dr. Donald P. Rogers, formerly of the Department of Botany, State Univer- sity of Iowa, not only for the collections and their preparation which made possible the investigation, but for locating the type of Helicogloea Lagerheimi Pat. Thanks are also extended to those institutions which so generously lent of their herbarium materials. LITERATURE CITED Bagchee, K. (’25). Cytology of the Ascomycetes. Pustularia balarioides Ramsb. Ann. Bot. 39: 217-266. 1925. Bary, A. de. (/84). Ve ergleichende Morphologie und Biologie der Pilze, Mycetozoen, und Bacterien. Leipzig, 1884. Bensaude, M. (718). Recherches sur le cycle evolutif et la sexualité chez les Basid- iomyeétes. Thèse. Paris, 1918 Blackman, V. H. (’04). On the fertilization, alternation of ir ME and gen- eral cytology of the Uredineae. Ann. Bot. 18: 323-373. Bourdot, H. (732). a c ipie nouveaux ou peu connus. rii Mye. Fr. Bull. 48: E e ——————, EN (709). Hyménomycétes de France. I. Hétérobasidiés. Ibid. 25: ea 1909. , (728). Hyménomycétes de France. Sceaux, 1928. WEG O. (88). E aus dem Gasammtgebiete der Mykologie. VII. zig, 1888. SE J. (703). Fungi polonici. Ann. Myc. 1: 97-131. 1903. 1936] BAKER—A STUDY OF HELICOGLOEA 111 Buller, A. H. R. (’22). Researches on fungi. 2. London, 1922. hamberlain, C. J. (732). Methods in plant histology. ed. 5. Chicago, 1932. Clements, F. E. (09). The genera of fungi. Minneapolis, 1909. ——— — ————, and C. L. Shear. (731). The genera of fungi. New York, 1931. Coker, W. C. (720). Notes on the lower ro CT of North Carolina. Jour. Elisha Mitchell Sei. Soc. 35: 113-182. Colley, R. H. (18). Parasitism, Ee adam p eytology of Cronartium ribicola. Jour. Agr. Res. 15: 619—659. 1918. Colson, B. (735). The eytology of the mushroom Psalliota campestris Quel. Ann. Bot. 49: 1-18. 1935. — — — — ——, (28). Notes on Basidiomycetes. Ibid. 48: 233-248. 1928. Ferry, R. (796). Les Protobasiai omycétes du Brésil. Rev. Myc. 18: 101-113. 1896. Gáumann, E. (722). Uber die Entwicklungsgeschichte von Jola javensis Pat. Ann. Mye. 20: 272-989. 1922. , and C. W. Dodge. (728). Comparative morphology of fungi. New rk, 1928. Janchen, E. (723). Die a 3 der Uredineen und Ustilagineen im System der Pilze. Österr. Bot. Zeitse : 302-304. 1923. Juel, H. O. (798). Die E ea in den Basidien und die Phylogenie der Ba- sidiomyceten. Jahrb. f. Wiss. Bot. 32: 361—388. 1898. Killermann, E (728). Eubasidii. In Engler u. Prantl. Die Natür. Pflanzenfam. 99-283. 1928. Kniep, H. » 13). Beiträge zur Kenntnis der Hymenomyceten I & II. Zeitschr. f. Bot. 5: 593-637. 1913. Lander, C. A. (735). The development of the fruiting body of Arachnion album. Jour. Elisha Mitchell Sei. Soe. 50: 2 Lindau, G. Se Phacidiineae. In kad u. eer Die Natiir. Pflanzenfam. I. 3- rf iili, Di Fi (729). Life-history and eytology of Saccoblastia intermedia n. sp. n. Mo. Bot. Gard. 16: 94. 1929. Lindfors, T. (724). Studien über den Entwicklungsverlauf bei einigen Rostpilzen aus zytologischen und anatomischen Gesichtpunkten. Svensk. Bot. Tidskr. 18: 1-84. 1924 Maire, R. (702). Recherches cem et taxonomiques sur les Basidiomycétes. Soc. Myce. Fr. Bull. Suppl. 18: 1-210. 19 Martin, t e (734). The genus Stypella. Univ. Iowa Studies Nat. Hist. 16: 143- 149 E Ait. (795). Protobasidiomyceten. Bot. Mitt. a. d. Tropen 8: 1-179. N Ber wW. (724). erem und systematische Stellung der Auriculariaceen und Tremellaceen. . Archiv 8: 250-297. 1924. Pady, S. M. (733). Teo development in the Pucciniastreae. Can. Jour. Res. 9: 458-485. 1933 —, (?84) vo in Milesia marginalis. Bull. Torrey Bot. Club. 61: 345-354. 1934. ———— — —— , (735). A cytological study of the development and germination of the Giora of Hyalopsora aspidiotus (Pk.) Magn. Ann. Bot. 49: 71-93. 1935. [Vor. 23 112 ANNALS OF THE MISSOURI BOTANICAL GARDEN Patouillard, N. Pr Les Hyménomycétes d'Europe. Paris, 1887. t G. de Lagerheim. (7/92). Champignons de l'equateur. Soc. Myc. Fr. Bull. e 113-140. 1892. 700). Essai taxonomique sur les familles et les genres des Hymén- TM Lons-le-Saunier, 1900. Rick, J. (733). Monografia ye protobasidiomicetos Rio-Grandenses. Egatea 18: 209-215, 285-288, 343-349. 1933 Rogers, D. P. (’32). A cytological study of Tulasnella. Bot. Gaz. 94: 66-105. 1932. —————— (783). Some noteworthy fungi from Iowa. Univ. Iowa Studies Nat. Hist. 15: 9-29. 1933. —— — — ——, (784). The basidium. Ibid. 16: 160-181. 1934. Ruhland, N. (701). Zur GE der intracellularen Karyogamie bei den Basidio- myeeten. Bot. Zeit. 59: 187—206. 1901. Saecardo, P. A. (782). SEI fungorum. 14: 245. Padua, 1882. Sass, J. E. (729). The Db aie basis for homothallism and heterothallism in the Agaricaceae. Am. Jour. Bot. 16: 663-701. 1929. Sharp, L. W. (734). nodus to cytology. ed. 3. New York, 1934. Tieghem, P. van. (793). Sur la classification des M d Jour. de Bot. 7: 77-87. 1893. Tischler, G. (’21—’22). Handbuch der Pflanzenanatomie. 2: Allgemeine Pflanzen- karyologie. 1921-1922. Wakayama, K. (’30). Contributions to the cytology of fungi. I. Chromosome number in Agaricaceae. Cytologia 1: 369-388. 1930. , (730a). Ibid. II. oa E in Morchella deliciosa Fr. Ibid. 2: 27-36. 1930. —— — — ——, (31). Ibid. III. Chromosome number in Aspergillus. Ibid. 2: 291—302. 1931. , (732). Ibid. IV. Chromosome number in Autobasidiomycetes. Ibid. 3: 260-284. 1932. Wakefield, E., and A. A. Pearson. (’23). Some additional records of Surrey resupinate Hymenomycetes. Brit. Myc. Soc. Trans. 8: 216-221. 1923. Wettstein, F. (’11). Handbuch der systematischen Botanik. ed. 2. 1911. Whelden, R. M., (*34). Cytological studies in the Tremellaceae. 1. Tremella. Mycologia 26: 415-435. 1934, [Vor. 23, 1936] 114 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 7 Helicogloea Lagerheimi All figures have been drawn with the aid of an Abbé camera lucida. Magnifica- tions are given for each figure. Figs. 1-13 inclusive show the morphological development of the basidium from the primordial stage through the mature basidium with spores. The nuclear con- ditions are representative only for these particular examples. All figures x 1100. Fig. 1. Enlarged terminal hyphal cell, the basidial D M with two nuclei, Fig. 2. Produetion of the hypobasidium as a lateral expansion of the primordium. ig. Hypobasidium enlarging, the nuelei moving toward the enlarging sec- ondary portion of the hypobasidium Fig. Mature hypobasidium with two nuclei in it. Fig. > Mature hypobasidium with fusion nucleus; epibasidium arising at the apical end of the primordium. Fig. 6. ucleus migrating through the SE hypobasidium collapsing as the protoplasm withdraws; epibasidium elongating. Fig. 7. Continued elongation of the epibasidium and withdrawal of protoplasm from the hypobasidium. ig. 8. Nucleus in enlarging epibasidium; hypobasidium entirely empty. Fig. 9. Enlarging epi Fig. 10. Epibasidium bend Fig. 11. Epibasidium two- SC following the first meiotie division; proto- plasm withdrawn partly from lower portions. Fig. 12. Epibasidium three-septate, Hn es budding sterigmata; the lower portions of the epibasidium collapsing, completely dud of protoplasm Fig. 13. Mature basidium with spores in place. ig. 14. Hyphal cells. a, d, young cells in which four nuclei are present, for the wall is not yet formed; b, e, older eells with increased reduction of protoplasm. x 1865. ig. 15. Origin of a branch from a subterminal cell; the terminal cell consti- tutes a pipes primordium. x 1865. Fig. asidial proliferation. (See also fig. 18d.) x 1100. Fig. 6 er epibasidia showing a crimp between the epibasidium and Peat aes x 1100. Fig. Variations in hypobasidial form. a, b, constricted hypobasidia; e, TENS R hypobasidium; d, e, pendent hypobasidia; f, g, extreme distal Sonate ric- tion. x 1100. Käsch * ENEE ENT BAKER — HELICOGLOEA 1936 ANN. Mo. Bor. GARD., VOL. 23 E, BOSTON. EX SC KAYN [Vor. 23, 1936] 116 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 8 Helicogloea Lagerheimi ig. Composite view of the fructification on a wood substratum as seen in section. Fig. 20. Sea les to illustrate the proportion of fruiting and hyphal regions in differently developed fructifications 100. Fig. 21. a, b, basidia arising 2 lateral branches. x 1100. Fig. Sg Origin of epibasidium prior to appearance of hypobasidium. x 1865. ig. a, b, ¢, irregular, lateral production of epibasidia. Surface level of the eid marked. x 1100 Fig. 24. eI of nuelear migration from the hypobasidium. x 1100. Fig. 25. Type II of nuclear migration from the hypobasidium. a, typical form; b, unconstricted type. Fig. 26. a, Type III of nuclear migration. b, e, d, e, subsequent development. Fig. 27. a, deep constriction and septum at the base of the fourth segment of the SE b, constriction without a wall. x 1100 Fig. 28. a, conjugate mitosis, with nearly parallel orientation; b, e, crossed spindles. x 1865. ig. 29. a, b, e, mitotic division figures indicating future crossed orientation of the spindles. In c a remnant of the nucleolus is visible. 1865. Fig. 30. Mitosis. a, membranes completely broken down, nucleoli still visible; b, nuclear membranes present in part, nucleoli distinct. x 1865. ale WEI e Po PED ee o E H T S 7 ) cl SIA a=) > 19 ys BEEN Dex EA A , SE t ANN. Mo. Bor. GARD., VOL. 23, 1936 BAKER — HELICOGLOEA V E n MES S [Vor. 23, 1936] 118 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 9 Helicogloea Lagerheimi Fig. 31. a, basidial primordium with the basal wall just formed; b, older primordium with hypobasidium and epibasidium both developing. x 1865. Fig. 32. series to show the increase in size from the initiation of the primor- dium through the hypobasidium production. 1865. Fig. 33. Origin of the hypobasidium. (See also fig. 31b.) x 1865. ig. 34. a, b, nuelei in close association previous to migration into the hypo- basidium. x 1865. Fig. 35. Lag of second nucleus in passage into hypobasidium. x 1865. Fig. 36. a, b, nuclei moving into hypobasidium. In b the first nucleus is pre- maturely enlarged prior to fusion. x 1865. ig. 37. Nuclei in hypobasidium previous to fusion. Note effect of streaming protoplasm. x 1865. Fig. 38. Nuclear fusion. a, nuclei touching, nuclear membranes still intact; b, nuclei as in a, but enlarged and showing a reticulate nature; e, nuclear mem- branes broken down and the contents mingling; d, nuclei fusing, one larger than the other with slightly more pronounced reticulations; e, nuclei of different sizes corresponding to their development, both deeper in the hypobasidium than usual. ig. 39. Fused nuclei with an indentation between the two stil apparent. Within the common membrane there is an advancing net development. x 1865. ig. 40. a, b,c, stages in nucleolar fusion. x 1865. Fig. 41. a, post-fusion nucleus, with strongly EE network; b, post-fusion nucleus with ee chromatic beads scattered on the ig. 42. ucleus contracting, the network still ea b, SE in resting aem completely contracted. x 5. Fig. 43. a, b, e, examples of nuclear passage from hypobasidium; a, detail of fig. 24; e detail of fig. 25. x 1865. Fi . Variation in the initiation of prophasie activity. a, nucleus enlarged, dën X 1865. b, nucleus still in primordial portion. x 1100 with detail x 1865. Fig. 45. Prophase in sporogenous part of the epibasidium. a, nuclear membrane intact, nucleolus peripheral, network appearing opposite pole of nucleolus; b, the network appearing slightly polarized; s. faint reticulations in enlarged nucleus from epibasidium of Type III. All x VOL. 23, 1936. m CS : BAKER — HELICOGLOEA [Vor. 23, 1936] 120 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 10 Helicogloea Lagerheimi Fig. 46. Prophase of first meiotic division. The nucleus is elongate, conforming to the shape of the epibasidium. x 1100 with detail x 1865. Fig. 47. Early prophase with chromatie beads on the threads. x 1865. Fig. 48. Prophase. a, nucleolus, slightly diminished in size, lying at anterior pole and the threads grouped at the opposite pole. x 1865; b, nueleolus still prom- inent, the threads less massed and central in position. x 1100 with detail x 1865; e, later prophase, the nucleolus at the periphery, the chromatin densely massed. x 1865 Fig. 49. Early prophase from an epibasidium with a development corresponding Fig. 50. Anaphase. a, intranuclear spindle with scattered chromosomes. Rem- nants of the nucleolus are visible within the membrane and centrosomes are Weer at the poles. b, nuclear membrane present in part; nucleolus sharp; e, d, e, mem- branes broken down and spindles free in cytoplasm; the nuclei have QUIE x 186 Fig. 51. Late anaphase. a, nuclear membrane discernible in part, the nueleolus distinet in the cytoplasm; b, membrane entirely gone, chromosomes massing together as they move to the poles; e, spindle contracted, chromosomes clumped, massing at the poles. The clearer areas at the poles denote the position of the daughter nuclei. x 1865. Fig. 52. Telophase. e, chromosomes massed at the poles, the two groups still connected by distinct fibers, daughter nuclei organizing; b, daughter nuclei more prominent and the fiber connections diminishing; e, lower nucleus nearly completely organized, the upper not so advanced; fibers prominent in upper re gion where they are still attached to the nueleus; d, darker area on connecting fibers indicative of wall formation. 18 Fig. 53. Two- ARR stage. a, upper nucleus completely organized, lower nucleus still showing fiber conneetions; the wall is just eoming in; b, wall just e the nuclei farther apart than ina. x 1865 Fig. Se Two-nucleate stage. The nuclei are reorganized but the wall is not yet formed. A clearer zone shows its future position. x 1865. th spindle ig. 55. Second meiotic division. Bo es are in anaphase, the nuclear membranes have disappeared as well as the nucleoli. x 1865. . 56 ariation in the synehronization of the second meiotie division. a, . syne SEN Zerf in anaphase, nucleolus distinct; upper nucleus in telophase. x 1865; b, lower nucleus in telophase with fiber connections si visible and the wall coming basal nuclei still in telophase with connecting fibers between. x 1100 with detail x 1865. Y , di » di T 39 D 99 ] "RM E e " av ; BT CU A IAN NS ups LL h eS EE E a Ce EE, HELICOGLOEA BAKER — gt ho ^ a m wA WR. s A emet n= a cd ee [Vor. 23, 1936] 122 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 11 Helicogloea Lagerheimi Fig. 57. Four-nucleate, four-septate epibasidium, two of the walls having just been formed. x 1 Fig. 58. a, fourth: segment of epibasidium with a basal septum; b, fourth seg- ment with protoplasmic remnants simulating a septum. 1865. Fig. 59. Segment from an epibasidium with a nucleus showing two nucleoli. x 1865. Fig. 60. Budding sterigmata. The wall is not completely formed between the two end cells. x 1865. Fig. 61. Sterigma with spore developing on the second segment of the epi- basidium. x 1865. Fi . Sterigmatic formation. In segments 2 and 3 the nuclei are changing form EE to passing into the spores. 865. Fig. 63. Variations in sporogenous portion of the epibasidium. a, apical seg- ment with typical sterigma; lower segments developing longer appendages; surface level marked; b, apical segment producing long appendage; surface level marked. Note the Bend in the basal segment. c, sterigmatic initiation in the fourth segment; d, sterigmata forming on opposite sides. All x 1100 Fig. 64. a, b, developing sterigmata with dark staining tips. x 1865. Fig. 65. Nuclear behavior preparatory to passage into the spores. a, nucleu normal, spore about half developed. Note that the dark-staining tip is still ap- parent; b, nucleus becoming deeply stainable; protoplasm withdrawing into the de- ias spore; e, nucleus dark-staining, irregular, attenuate; d, nueleus a homog- eneous dark mass. x 1865 Nuclear passage ine the spore. a, nucleus in sterigma; b, nucleus of end ale in sterigma; second segment completely drained of protoplasm an nucleus reorganizing in the spore; e, d, nuclei much drawn out in passage through the sterigmata; e, nucleus mostly within the spore; f, g, nuclei not yet reorganized in the spores; h, one nucleus reorganizing in the spore, ee? de the basal seg- ment ready to migrate; i, nucleus ee in the spore. All x 1865. ig. 67. a, nucleus moving into a long appendage without change in form. x 1865; b, sporogenous portion of an epibasidium, all segments of which have produeed long appendages. x 1100 o ae aM Gt aR oe e pt XS PE A MU AS HESS T e BAKER — HELICOGLOEA [VoL. 23, 1936] 124 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 12 Helicogloea Lagerheimi Fig. 67. e, d, formation of sterigmata and spores at the ends of long ap- pendages. Nucleus unchanged at this stage. Surface level marked in d; e, f, nuclei prematurely ehanged for passage into sterigmata and spores. All x 1865 Fig. 68. a, typieal spore; b, spore germinating while still attached to the sterigma; e, d, germination by repetition. x 1865. Fig. 69. Nucleus changing for passage through germ tube and sterigma into MT spore. 1 v 70. TO EE. of spores in hanging-drop cultures. a, stages prior to ms tube produetion, 24 hours; b, produetion of germ tubes and secondary spores, 24—48 hours; e, protoplasm withdrawn into secondary spores, 72 hours. All x 734. ig. 71. ‘Nuclei showing varying sizes through the cycle and corresponding nucleolar volumes. All from the same section and x 1865. a, dikaryon from ter- minal hyphal cell; b, dikaryon prior to migration into hypobasidium; e, fusion nucleus in B DONE ium ; d, post-fusion, nucleus contracting, nucleolar RE ob- viously increased; e, nuclei from migrating stages after fusion; f, nuclei from 4-celled stage of EC prior to sterigmatic formation; g, "ebe from typical spores. nti E "TE PEERS Ls ee ES, SEAS AA ES uud A e 2 ee A a EN t , d > * O A BAKER — HELICOGLOEA [Vor. 23, 1936] 126 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 13 Fig. 72. Helicogloea pinicola, No. 20124. a, forked hypobasidium; b, budding epibasidium sub-apieal in position; e, budding epibasidium slightly sub-apical in position; d, typieal hypobasidia; e, septate epibasidium; f, spores; g, clamp con- nections on the hyphae No. 20555. h, slightly constricted hypobasidium; i, j, forked hypobasidia; k, septate p with a fourth septum present; 1, sub-apical epibasidium. All figures x 434. Fig. Ge H. pinicola f. alniviridis. a, origin of epibasidium, hypobasidium forked; b, irregular hypobasidium, typical epibasidium; c, septate epibasidium, two segments of whieh are already discharged; d, germinating spore; e, proliferation of basidia. All x 434. ig. Helicogloea graminicola. a, origin of the hypobasidium; b, mature UC TEE ES e, septate umane Note that all the protoplasm has not with- drawn from the hypobasidium. d, spor All x 434. Fig. 75. Helicogloea intermedia. a, psu hypobasidium; b, origin of the epibasidium, hypobasidium much constricted; e, basal constriction of hypobasidium ; d, typical origin of epibasidium; e, proximal production of epibasidium or basal constriction; f, g, h, stages in the development of the clavate type of hypobasidium. All x 734. Fig. 76. Helicogloea caroliniana. a, variation in the hypobasidium; b, basidium; e, epibasidium with developing sterigmata and spore; d, long PORE with sterigma and spore; e, mature spores. All x 1100 Fig. 77. Helicogloea Lagerheimi, No. 42262, marked as var. vulgaris. a, con- stricted hypobasidium, typical epibasidium of terminal origin; b, intercalary origin of the basidium; e, septate epibasidium with four septa; two segments have gid discharged their spores; d, spores. All x 1100. Fig. 78. Helicogloea Lagerheimi, wick’ material. a, typical hypobasidium and ege dium; b, e, d, variously constricted hypobasidia; e, interealary origin of ium; f, developing sporogenous Zeg of epibasidium; g, epibasidium de- Bee ala: h, spores. All x 1100. = d z x k i k e " El] e Ob H E DE > wl > xls H ü -À ix Ue El «| 9512191 Z 3| > BHE a E s DE £ ole CEE bel ei E: a = 3 5 d > ? E " i H Een e e z e et BI Aa oxo H e 2 eo” d em ^ a? * Fig. 2. Phylogenetic Pg eua of American species of Apoc num deir, des to Woodson. Broken lines represent the hypothetical of the genus, the solid. «E the supposed relationship of ecies and varieties. ccording to more recent nomenclatural corrections, ‘‘ A. androsaemifolium? SE he est xau should read A. sah Ie . gla and ‘ osaemifolium var. incanum?’ uld be Ke eege a Me n. Beie (ef. Rhodora 24: 30- 31. 1932). (VoL. 23 166 ANNALS OF THE MISSOURI BOTANICAL GARDEN be very probably crossed back to one of original parental species, at least in part, and may often resemble that species superficially. It will be remembered that the puzzling seed- lings of A. androsaemifolium came not directly from that species, but were hybrid segregates closely resembling it. In Apocynum, therefore, one of the chief effects of hybridization seems to be the enrichment of the variability of the original species taking part in the hybridization. The actual creation of a new intermediate species by that process, while certainly a possibility, does not seem to have been effected in the case of A. medium. That binomial may be a necessity for purposes of classification, but biologically it is still in a far different status from A. androsaemifolium and A. cannabinum. To reach that status it would require a period of isolation and the operation of natural selection to remove the variability and the semi-sterility which now characterize it. The behavior of A. medium No. 449 and of A. medium No. 446 is particularly interesting. The former must apparently have been a first-generation cross, for like such plants it is highly heterozygous. Its progeny (pl. 19, fig. 3) include everything from almost straight A. cannabinum to a superficially normal A.androsaemifolium. Apocynum medium No. 446, on the other hand, bred almost true (pl. 19, fig. 4). Most of its progeny were more or less like itself; in it the 4. medium type is on the way to becoming stabilized and it probably represents a hybrid of the second generation or later. Given a fair degree of isolation there seems to be no reason why such an intermedi- ate type might not in a comparatively short time reach a specific status comparable to that of A. cannabinum or of A. androsaemifolium. SUMMARY 1. Progeny tests were made of Apocynum androsaemifolium, A. cannabinum, and their putative hybrid, A. medium. 2. Apocynum androsaemifolium and A. cannabinum bred true. A. medium produced a variable set of seedlings, some of them indistinguishable from A. androsaemifolium and A. cannabinum. 1936] ANDERSON— HYBRIDIZATION IN APOCYNUM 167 3. The seedlings of A. cannabinum had uniformly high per- centages of fertile pollen and those of A. androsaemifolium were nearly as fertile. Those of A. medium were of low aver- age fertility. 4. It is concluded that A. medium is certainly a hybrid. 5. The taxonomic importance and phylogenetic consequences of interspecific hybridization in Apocynum are discussed in the light of these results. It is suggested that the chief effect of hybridization in this genus in eastern North America at the present time is to increase variability in the parental species. BIBLIOGRAPHY Anderson, Edgar (’28). The We of species in the northern blue flags, Iris versicolor L. and Iris virginic Ann. Mo. Bot. Gard. 15: 241-332. Medwedewa, G. (735). Zur Sie. der Bestäubung von Apocynum venetum L. Stout, A. B., and Clyde Chandler (733). Pollen-tube behavior in pongan. with special PINO to incompatibilities. Bull. Torrey Bot. Club 60: 397-416. Woodson, R. E., Jr. (’30). Studies in the Apocynaceae. I. A sar study of the SCC (with special reference to the genus Apocynum). Ann, Mo. Bot. Gard. 17: 1-212. [ Vor. 23, 1936] 168 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 19 Photographs of one flower from each plant for; 1, nine seedlings of A. androsaemifolium pu 503. 2, eight seedlings of A. cannabinum »" 447. 3, seven seedlings of 4. medi 4, ten seedlings of A. medium No. y In the center a portion of a centimeter scale photographed at the same magnifica- tion for comparison. ANN. Mo. Bot. GARD., Vor. 23, 1936 PLATE 19 3 ANDERSON--HYBRIDIZATION IN APOCYNUM STUDIES IN THE APOCYNACEAE. IV! THE AMERICAN GENERA OF ECHITOIDEAE ROBERT E. WOODSON, JR. Research Assistant, Missouri Botanical Garden Assistant Professor in the Henry Shaw School of Botany of Washington University X. Neosracea Britton Neobracea Britton, in Britton € Millsp. Bahama Fl. 335. 1920; Urb. Symb. Ant. 9: 240. 1924. Bracea Britton, Bull. N. Y. Bot. Gard. 3: 448. 1905, not ing. Lactescent shrubs or small trees. Stems erect to ascending, terete; branches dichotomous to alternate when adventitious. Leaves opposite, shortly petiolate, penninerved, eglandular, the petioles subtended by 2-several inconspicuous stipular ap- pendanges. Inflorescence terminal, less frequently to sub- terminal or lateral, scorpioidally corymbose to subumbellate, few- to several-flowered, inconspicuously bracteate. Calyx 5- parted, the lobes equal or subequal, cleft nearly to the recep- tacle, imbrieated, eglandular, or bearing within 5-10 alternate squamellae. Corolla infundibuliform, the tube straight, in- appendiculate within, the limb actinomorphic, 5-parted, dex- trorsely convolute. Stamens 5, the anthers connivent and ag- glutinated to the stigma, consisting of 2 parallel sporangia borne ventrally near the apex of an enlarged, narrowly sagit- tate, peltate connective; pollen granular. Carpels 2, united at the apex by a common stylar shaft surmounted by the capitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, separate, or somewhat con- crescent at the base. Follicles 2, apocarpous, terete, acuminate, dehiscing along the ventral suture, containing many dry, truncate, apically comose seeds. 1 Concluded from ANN. Mo. Bor. Garp. 22: 153-306. (187)-(340). 1935. Issued June 10, 1936. ANN. Mo. Bor. GARD., VOL. 23, 1936. — (341) (169) [Vor. 23 170 ANNALS OF THE MISSOURI BOTANICAL GARDEN Type species: Neobracea bahamensis Britton, in Britton & Millsp. Bahama Fl. 335. 1920. KEY TO THE SPECIES a. Plants glabrous throughout; calyx-lobes rather conspicuously foliaceous, ovate to ovate-oblong; plants of Cuba................... 1. N. Valenzuelana aa. Plants more or less conspicuously pubescent TOR MT ealyx-lobes scarious or only slightly foliaceous, ovate- to oblong-trigonal. b. Inflorescence corymbose, the peduncle equalling or surpassing the pedicels: plants of CUBA EE 2. N. angustifolia bb. Inflorescence umbellate or subumbellate, the peduncle markedly shorter than the pedicels. c. Squamellae evident; follicles 15-25 em. long; plants of E Bahama Islands eer foc E EE . N. bahamensis ee. Squamellae obsolete or extremely inevident; follieles de em. long; planta:0f Cubes eege ee der 4. N. Ekmanit 1. Neobracea Valenzuelana (A. Rich.) Urb. Symb. Ant. 9: 241. 1924. Echites Valenzuelana A. Rich. in Sagra, Hist. Cuba 11: 93. 1850 Rhabdadenia Wrightiana Muell.-Arg. Linnaea 30: 438. 1860. Mandevilla Wrightiana (Muell.-Arg.) Benth. & Hook. Gen. Pl. 2: 727. 1876. Angadenia Valenzuelana (A. Rich.) Miers, Apoc. So. Am. 181. 1878. Stems relatively stout, glabrous; leaves opposite, shortly petiolate, obovate to oblong-oblanceolate, apex rounded and usually minutely emarginate, base narrowly cuneate, 2.5-6.0 em. long, 0.8-2.0 em. broad, coriaceous, glabrous throughout, dark green, nitidulous above, paler, opaque beneath; petioles 0.5-0.7 cm. long; inflorescence terminal, occasionally pseudo- lateral, subumbellate, bearing 1-5 pale rose flowers; peduncle 0.2-0.8 em. long; pedicels 2.0-2.5 em. long, glabrous; bracts 0.2-0.3 em. long; calyx-lobes ovate to ovate-oblong, obtuse to broadly acute, 0.3-0.5 cm. long, rather conspicuously foli- aceous, glabrous, the squamellae in alternate groups of 3-4; corolla infundibuliform, glabrous without, the proper-tube 0.4—0.5 em. long, about 0.15 em. in diameter at the base, the (342) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 171 throat broadly conical, 0.8-0.9 em. long, about 0.6-0.65 em. in diameter at the orifice, the lobes obliquely obovate-dolabriform, obtuse, 1.4—1.5 em. long, widely spreading ; stamens inserted at the base of the corolla-throat, the anthers narrowly sagittate, 0.4—0.425 em. long, minutely tomentulose dorsally ; ovary ovoid, about 0.1 em. long, glabrous; stigma capitate, 0.1 cm. long; nectaries about half equalling the ovary, more or less conspicu- ously 3-lobed; follicles relatively slender and flexile, very obscurely articulated, 11-15 em. long, glabrous; seeds 0.7-0.8 em. long, the very pale, yellowish coma 1.1-1.3 cm. long. UBA: ORIENTE: prope Paso Estancia ad Pinales versus in collibus i May, 1860, Wright 399 (B, G, MBG, NY, US, V); swampish thieket, alt. 400—500 m., Sierra Nipe, along trail Piedra Gorda to Woodfred, Dec. 8, 1909, ess 8106 (NY); haa Baracoa, Sept. 2, 1917, Roig 109 (NY); Camp La Gloria, south of a Moa, Dec. 24-30, 1910, Shafer 8268 (NY); by water, barren savannas, Dd of Holguin, April 9, 1909, Shafer 1281 (NY); SANTA CLARA: edge o arroyo, palm barren, Santa Clara, April 8-9, 1912, Britton $ Cowell 13290 (NY); palm barren, Motembo, Aug. 10, 1918, Leon $ Roca 8237 (NY); MATANZAS: Ser: pentine hills, near Cauasi, Oct. 10, 1927, Leon 13127 (NY); bushy savanna, San Miguel de los Baños, Aug. 8, 1919, Leon $ Roca 8904 (NY); HABANA: eruptine rock soil, Madruga, March 31, 1903, Shafer 13 (NY); Baños del Boticario, not far from Campo Florido, July 18, 1912, Leon 3353 (NY); PINAR DEL RIO: San Jose de Sagua to San Marcos, on serpentine rocks, Jan. 27, 1912, Shafer 11966 (NY). The habit of this species varies from that of a shrub to that of a small tree, the height of plants being reported as from 1 to 8 meters. The corolla is white to pinkish, reddish-flushed in the throat. The calyx-lobes, although not large, are rather con- spicuously foliaceous, consisting of a definite lamina, in which the midrib is prominent. The spreading position of the calyx- lobes is characteristic. 2. Neol gustifolia Britton, Bull. Torrey Bot. Club 53: 462. 1926. Stems relatively stout, softly hirsutulose when young, even- tually becoming glabrate and inconspicuously lenticellate; leaves opposite, shortly pedicellate, narrowly oblong-elliptic, apex acute to obtuse, base rather narrowly cuneate, 5-7 cm. long, 0.7-1.3 em. broad, firmly membranaceous, the margins somewhat revolute in desiccation, either surface softly and (343) [Vor. 23 172 ANNALS OF THE MISSOURI BOTANICAL GARDEN rather densely puberulent; petioles 0.3-0.5 em. long, puberu- lent; inflorescence lateral, alternate, somewhat shorter than the subtending leaves, bearing 8-20 small, purplish (?) flowers; peduncle minutely puberulent; pedicels 0.45-0.5 em. long, mi- nutely puberulent; calyx-lobes rather narrowly trigonal, acute to acuminate, 0.2-0.25 em. long, minutely puberulent without, the solitary, alternate squamellae extremely minute; corolla infundibuliform, minutely and rather sparsely puberulent without, the proper-tube 0.3-0.35 em. long, about 0.125 em. in diameter at the base, the throat conical-campanulate, 0.35 cm. long, about 0.4-0.425 cm. in diameter at the orifice, the lobes oblong-dolabriform, obtuse, 0.7-0.725 em. long, widely spread- ing; stamens inserted at the base of the corolla-throat, the anthers rather narrowly sagittate, 0.325 cm. long, minutely tomentulose dorsally; ovary oblong-ovoid, about 0.1 em. long, glabrous; stigma subcapitate, about 0.075 em. long; nectaries compressed-ovoid, about one-third equalling the ovary; fol- licles unknown. BA: PINAR DEL RIO: rocky soil between Santa Cruz and Las Cayuelas, April 12, 1924, Roig 3227 (NY, TYPE, MBG, photograph and analytical drawings). 3. Neobracea bahamensis Britton, in Britton € Millsp. Bahama Fl. 335. 1920. Bracea bahamensis Britton, Bull. N. Y. Bot. Gard. 3: 448. 1905. Stems relatively stout, minutely puberulent when young, eventually becoming glabrate; leaves opposite, shortly peti- olate, broadly elliptic to oblong-oblanceolate, apex obtuse to rounded, base more or less narrowly cuneate, 1.5-10.0 em. long, 0.5-2.5 cm. broad, coriaceous, above yellowish-green, somewhat nitidulous, essentially glabrous, beneath paler, densely and minutely puberulent; petioles 0.2-0.6 em. long, minutely puber- ulent ; inflorescence terminal, less frequently to subterminal or lateral, umbellate, bearing 1-10 white, reddish-flushed flowers ; peduncle 0.4—1.6 cm. long, minutely puberulent ; pedicels 0.7—1.5 em. long, minutely puberulent; bracts very minute, 0.1 em. or less in length; calyx-lobes trigonal, 0.2-0.3 em. long, minutely (344) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 173 puberulent without, the squamellae obsolete or extremely in- evident; corolla infundibuliform, minutely puberulent without, the proper-tube 0.15-0.17 cm. long, about 0.1 cm. in diameter at the base, the throat rather narrowly conical, 0.5-0.8 cm. long, about 0.3-0.35 em. in diameter at the orifice, the lobes obliquely obovate, 1.5-1.8 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers narrowly sagittate, 0.35 em. long, minutely puberulent dorsally ; ovary ovoid, about 0.1 em. long, glabrous; stigma subcapitate, 0.1-0.125 cm. long; nectaries somewhat shorter than the ovary, dentiform; follicles relatively slender and flexile, obscurely and distantly monili- form, 15-25 em. long, glabrous ; seeds 0.45—0.5 em. long, the pale yellowish coma 1.4—1.5 em. long. BAHAMA ISLANDS: New Providence, June 7, 1879, Brace 493 (NY, TYPE); near Nassau, N. P., April and May, 1903, Curtiss 137 (MBG, NY); borders of salt marsh, Millers, N. P., Sept. 3, 1904, Britton $ Brace 526 (NY); border of mangrove swamp, Deep Creek, Andros, Aug. 18-Sept. 10, 1906, Brace 5177 (NY); pineland, near Fresh Creek, Andros, Jan. 28-31, 1910, Small $ Carter 8751 (NY); rocky plain, Orange Creek and vicinity, Cat Island, Febr. 27-28, 1907, Britton $ Millspaugh 5783 (NY); Landrail Point, Crooked Island, Jan. 9-23, 1906, Brace 4670 (NY) ; coastal coppice, Pinder's Point, Great Bahama, Febr. 5-13, 1905, Britton $ Millspaugh 2511 (NY); road to South Side, Long Cay, Dee. 7-17, 1905, Brace 4053 (NY); serubland, near Georgetown, Great Exuma, Febr. 22-28, 1905, Britton $ Millspaugh 2970 (NY). 4. Neobracea Ekmanii Urb. Symb. Ant. 9: 242. 1924. Stems relatively stout, essentially glabrous; leaves opposite, shortly petiolate, elliptie to elliptic-oblong, apex acute to very obsoletely acuminate, occasionally obtuse, base cuneate, 1—2 em. long, 0.4—0.8 em. broad, coriaceous, above essentially gla- brous, more or less nitidulous, beneath opaque, minutely and rather sparsely pilosulose; petioles 0.3-0.5 em. long, very mi- nutely pilosulose ; inflorescence terminal, 1-flowered ; peduncle 0.3-0.4 em. long, essentially glabrous; pedicel 0.2—0.3 cm. long, very minutely pilosulose ; calyx-lobes linear-trigonal, obtusish, about 0.25 em. long, minutely pilosulose, eglandular within; corolla, anthers, and ovary unknown; follicles terete, relatively stout, 5-7 cm. long, essentially glabrous without; seeds 0.5- 0.6 em. long, the pale tawny coma about 1.8 em. long. CUBA: ORIENTE: prope Maravi, in pinetis, Ekman 4051 (B, TYPE). (345) [Vor. 23 174 ANNALS OF THE MISSOURI BOTANICAL GARDEN I have not been able to examine the type specimen, hence the preceding description is merely an adaptation of Urban's. XI. GALACTOPHORA Woodson Galactophora Woodson, Ann. Mo. Bot. Gard. 19: 49. 1932. Lactescent, suffruticose undershrubs or suffrutescent herbs. Branches erect or ascending, opposite to occasionally alternate above. Leaves opposite, coriaceous to subcoriaceous, petiolate to subsessile, entire, penninerved, eglandular; petioles some- what girdling at the node into an obscurely appendiculate, stipular ring. Inflorescence terminal to subterminal, scor- pioidally corymbose to subumbellate, bracteate, relatively few- flowered. Calyx 5-parted, the lobes equal to subequal, cleft nearly to the receptacle, imbricated, foliaceous, bearing within several to many indefinitely distributed squamellae. Corolla infundibuliform, large and showy, the proper-tube straight, inappendiculate within, the limb aetinomorphie, 5-parted, dex- trorsely convolute. Stamens 5, the anthers connivent and ag- glutinated to the stigma, US of 2 parallel sporangia with a conspicuous protuberant base borne ventrally near the apex of an enlarged, narrowly sagittate, peltate connective; pollen granular. Carpels 2, united at the apex by a common stylar shaft surmounted by the 5-gonal, fusiform stigma; ovules many, several-seriate, borne upon an axile, binate pla- centa. Nectaries 5, coalescent, and usually somewhat adnate to the ovary. Follicles 2, apocarpous, terete, acuminate, de- hiscing along the ventral suture, containing many dry, trun- cate, apically comose seeds. Type species : Galactophora crassifolia (Muell.-Arg.) Wood- son, Ann. Mo. Bot. Gard. 19: 50. 1932. KEY TO THE SPECIES a. Inflorescence 5-10-flowered, corymbose; calyx-lobes linear-lanceolate. . EE AU ME EE el E SEN aa. DS i 1-5-flowered, subumbellate; calyx-lobes broadly ovate to te-lanceolate b. Ger and corolla puberulent, bearing irregularly interspersed, glandular aculei. (346) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 175 e. Calyx-lobes 1.5-2.0 em. long; leaves broadly oblong to oblong-lance- A A A A EE, 2. G. crassifolia ec. Calyx-lobes 0.5—0.6 em. long; leaves ovate.......... 3. G. Schomburgkiana bb. Calyx and corolla glabrous. e. Corolla-tube 3—4 em. long, the lobes about 1.5 em. long; leaves broadly ODlong- 3-5. USE Ie we ols oae euh de oce e BAT 4. G. calycina ee. Corolla-tube 5.0—5.5 em. long, the lobes about 4.5 em. long; leaves broadity-ovate42 2e rre es EE 5. G. magnifica 1. Galactophora pulchella Woodson, Ann. Mo. Bot. Gard. 19: L 1932. Stems relatively stout, softly puberulent when young, soon becoming glabrate; leaves opposite, subsessile and somewhat amplexicaul, broadly ovate, apex obtusish, base abruptly rounded and obscurely cordate, 3.5-7.0 cm. long, 2.5—4.5 cm. broad, glabrous, or either surface softly puberulent when young; petioles 0.1-0.2 em. long; inflorescence corymbose, 5-10-flowered; peduncle 3-5 em. long, softly puberulent; ped- icels 1.0-1.5 em. long, softly puberulent with small, glandular aculei interspersed rather irregularly; bracts inconspicuous; ealyx-lobes linear-lanceolate, 0.5-0.75 em. long, minutely pu- berulent with interspersed aculei; corolla infundibuliform es- sentially glabrous without, or with extremely inconspicuous indument and aculei, the proper-tube 0.5-0.6 em. long, about 0.2 cm. in diameter at the base, the throat 1.5 em. long, about 0.4 em. in diameter at the orifice, the lobes obliquely oblong- obovate, obtuse, 1.0-1.25 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers lanceolate- sagittate, 0.5-0.6 em. long, glabrous; ovary oblong-ovoid, about 0.15 em. long, glabrous; stigma about 0.15 em. long; nectaries somewhat shorter than the ovary; follieles unknown. BRAZIL: AMAZONAS [?]: Cano Pimicheiro, June, 1854, Spruce 3718 (K, TYPE, MBG, photograph and analytical drawings). 2. Galactophora crassifolia (Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 19: 50. 1932. Amblyanthera crassifolia Muell.-Arg. in Mart. Fl. Bras. 61: 143. 1860. Echites crassifolia Spruce, ex Muell.-Arg. loc. cit. 1860, nom. nud. in synon. (347) [Vor. 23 176 ANNALS OF THE MISSOURI BOTANICAL GARDEN Rhodocalyx crassifolius (Muell.-Arg.) Miers, Apoc. So. Am. 139. 1878. Mandevilla crassifolia Muell.-Arg. ex K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895, sphalm. Mandevilla crassifolia (Spruce) K. Sch. ex Mgf. in Fedde, Rep. Spec. Nov. 20: 24. 1924, sphalm. Stems relatively stout, minutely puberulent with inter- spersed, glandular aculei when young, becoming glabrate; leaves opposite, subsessile, broadly oblong to oblong-lance- olate, apex obtuse to rounded, base rather abruptly rounded and obscurely cordate, more or less amplexicaul, 5-12 em. long, 3-5 em. broad, coriaceous, glabrous; petioles 0.1-0.2 em. long, glabrous to aculeate as upon the stem; inflorescence subumbel- late, 1—5-flowered; pedicels 1.2-1.5 em. long, glabrous to mi- nutely puberulent; bracts scarious, about 0.1 em. long; calyx- lobes ovate, acute, 1.5-2.0 em. long, more or less conspicuously aeuleate; corolla infundibuliform, conspicuously aculeolate without, the proper-tube 2.5-3.0 em. long, about 0.2 em. in di- ameter at the base, the throat 3 em. long, about 2 em. in diameter at the orifice, the lobes obliquely obovate, obtuse, 2.5-3.0 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers lanceolate-sagittate, 0.6—0.7 em. long, glabrous, or infrequently sparsely barbellate at the base of the narrow auricles; ovary oblong-ovoid, about 0.3 em. long, glabrous; stigma 0.15 em. long; nectaries about half equalling the ovary; follicles relatively stout, rigid, 12-16 em. long, gla- brous ; seeds 0.3 em. long, the tawny coma about 2 em. long. VENEZUELA: AMAZONAS: Cerro Yapacana, upper Rio Orinoco, alt. 100 m., April, 1931, Holt $ Blake 716 (US). BRITISH GUIANA: exact locality lacking, Dec., 1842, Schomburgk 1551 (B). BRAZIL: AMAZONAS: prope San Carlos, ad Rio Negro, 1853-4, Spruce 3136 (B, Camb., K, V, MBG, photograph and analytical drawings) ; in eampis arenosis prope Faro, date lacking, Ducke 8434 (B); Manáos, July 31, 1900, Ule 5175 (B); MATTO Grosso: Juruena Procedencia, April, 1909, Hoehne 1804 (B) ; Cataqui-Imain, Jan., 1919, Kuhlmann 2252 (B). 3. Galactophora Schomburgkiana Woodson, Ann. Mo. Bot. Gard. 19: 50. 1932. Stems relatively stout, softly puberulent when young, soon (34 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 177 becoming glabrate; leaves opposite, subsessile, broadly ovate, apex obtusish, base abruptly rounded, obscurely cordate and more or less amplexicaul, 4—7 em. long, 2.5-4.5 em. broad, cori- aceous, softly puberulent when young, soon becoming glabrate ; petioles 0.1-0.2 cm. long; inflorescence subumbellate, 3-5- flowered; pedicels 1.0-1.5 em. long, softly puberulent with inter- spersed aculei; bracts extremely inconspicuous; calyx-lobes ovate, acutish, 0.5-0.6 em. long, minutely puberulent, evidently rarely aculeolate; corolla infundibuliform, aculeate-striate without, the proper-tube 2-3 em. long, about 0.2 cm. in diameter at the base, the throat 3.0-3.5 em. long, about 0.8 cm. in diameter at the orifice, the lobes obliquely obovate, obtuse, 1.0-1.2 cm. long; stamens inserted at the base of the corolla-throat, the anthers and gynoecium unseen; follicles relatively stout, 14 cm. long, densely aculeolate; seeds 0.6 cm. long, the tawny coma 1.5-2.0 em. long. BritisH GUIANA: exact locality and date lacking, Schomburgk s. n. (K, TYPE, MBG, photograph and analytical drawings). 4. Galactophora calycina (Hub.) Woodson, Ann. Mo. Bot. Gard. 19: 50. 1932. Dipladenia calycina Hub. ex Ducke, Archiv. Jard. Bot. Rio Janeiro 3: 247. 1922; Bol. Mus. Goeldi 7: 113. 1913, nom. nud. Plate 2. Stems relatively stout, glabrous; leaves opposite, subsessile, broadly oblong, apex obtuse or rounded, frequently somewhat retuse, base rather abruptly rounded, obscurely cordate and more or less amplexicaul, 3-6 em. long, 2-4 cm. broad, cori- aceous, glabrous; petioles about 0.1 em. long ; inflorescence sub- umbellate, 1-5-flowered; pedicels 0.5-0.7 cm. long, glabrous; bracts very inconspicuous; calyx-lobes ovate, acute, 1.5-2.0 cm. long, glabrous; corolla infundibuliform, glabrous without, the proper-tube 1.5-2.0 cm. long, about 0.2 cm. in diameter at the base, the throat 1.5-2.0 em. long, about 1 cm. in diameter at the orifice, the lobes broadly obovate, obtuse, 1.2-1.6 cm. long, (349) [Vor. 23 178 ANNALS OF THE MISSOURI BOTANICAL GARDEN widely spreading; stamens inserted at the base of the corolla- throat, the anthers narrowly sagittate, 0.6-0.7 em. long, gla- brous; ovary ovoid, about 0.4 em. long, glabrous; stigma 0.2 em. long; nectaries somewhat less than half equalling the ovary; follicles unknown. BRAZIL: PARA: Faro, inter fruticulos loeo Campos do Tigre, Dee. 31, 1919, Ducke 11393 (B, S. MBG, photograph and analytical drawings); Cachoeira, sandy ground of new clearing, 1898, Gwynne-Vaughan 33 (K, MBG, photograph and analytical drawings). 9. Galactophora magnifica Woodson, Ann. Mo. Bot. Gard. 19: 382. 1932. Stems relatively stout, glabrous; leaves opposite, sessile, broadly ovate, apex obtuse, base broadly cordate and amplexi- caul, 4.5-5.0 em. long, 3.0-3.5 em. broad, coriaceous, wholly gla- brous; inflorescence subumbellate, few- (in our specimen 3-) flowered ; pedicels 0.9-1.0 em. long, glabrous; bracts extremely minute, scarious; calyx-lobes broadly ovate-lanceolate, acumi- nate, 2.0-2.5 em. long, glabrous; corolla glabrous without, the proper-tube 2.0-2.3 cm. long, about 0.4 em. in diameter at the base, the throat broadly campanulate, 2.8-3.0 em. long, about 2.9 em. in diameter, the lobes obliquely ovate, shortly acumi- nate, 4.5 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers narrowly sagittate, 0.8 em. long, glabrous; ovary oblongoid, about 0.15 em. long, gla- brous; stigma 0.15 em. long; nectaries somewhat shorter than the ovary ; follieles unknown. BRAZIL: MATTO GROSSO: Procedencia Juruena, campo humido e pantuoso, April, 1909, Hoehne 1759 (US, TYPE, MBG, photograph and analytical drawings). XII. SarPiNcTESs Woodson Salpinctes Woodson, in Gleason, Bull. Torrey Bot. Club 58: 453. 1931. Lactescent, suffrutescent herbs or suffruticose undershrubs. Stems erect to ascending, terete. Leaves opposite to sub- verticillate, shortly petiolate to essentially sessile, penni- nerved, eglandular. Inflorescence terminal, uniflorous in our (350) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 179 specimens. Calyx 5-parted, the lobes equal to subequal, cleft nearly to the receptacle, imbricated, bearing many indefinitely distributed squamellae within. Corolla salverform, the tube straight, exappendiculate within, the limb actinomorphic, 5- parted, dextrorsely convolute. Stamens 5, the anthers con- nivent and agglutinated to the stigma, consisting of 2 parallel sporangia borne ventrally near the apex of an enlarged, nar- rowly sagittate, peltate connective; pollen granular, Carpels 2, united at the apex by an elongate, common style surmounted by the pentagonal-fusiform stigma; ovules many, several- seriate, borne upon an axile, binate placenta. Nectaries 2, separate. Follicles apocarpous, terete, acuminate, dehiscing along the ventral suture, containing many dry, truncate, apically comose seeds. Type species: Salpinctes kalmiaefolius Woodson, in Glea- son, Bull Torrey Bot. Club 58: 453. 1931. KEY TO THE SPECIES a. Leaves oblong-ovate, 0.8—1.0 em. broad, strictly opposite: stems glabrous cog mo CHO OOO Cdp a I Oo CoE E T 1. S. kalmiaefolius aa. Leaves oblong-linear, 0.2-0.4 em. broad, opposite to subverticillate; stems scabrous:bullate......22.-22 2 oo NEP slo ela ERES ems 2. S. duidae 1. Salpinctes kalmiaefolius Woodson, in Gleason, Bull. Tor- rey Bot. Club 58: 453. pl. 37. 1931. Plate 3. Stems relatively stout, 2-5 dm. tall (fide Tate), wholly gla- brous ; leaves opposite, subsessile, oblong-ovate, apex obtuse to rounded, base rounded to subtruncate, 2—4 cm. long, 0.8-1.0 em. broad, heavily coriaceous, the margin strongly revolute in desiccation, wholly glabrous, lustrous above, opaque beneath; petioles 0.1-0.15 cm. long; inflorescence terminal, bearing a solitary, showy, bright pink flower; peduncle essentially obso- lete; pedicels 0.15-0.2 cm. long, glabrous; bracts broadly triangular-ovate, about 0.1 cm. long, scarious, caducous ; calyx- lobes ovate-lanceolate, acuminate, 0.45—0.5 em. long, scarious, glabrous; corolla salverform, glabrous without, the tube 2.5-3.5 (351) [Vor. 23 180 ANNALS OF THE MISSOURI BOTANICAL GARDEN em. long, about 0.125 em. in diameter at the base, slightly dilat- ing toward the orifice, the lobes obliquely obovate, obtuse, 2.5— 3.0 em. long, widely spreading; stamens inserted somewhat below midway within the corolla-tube, the anthers narrowly sagittate, 0.45 cm. long, glabrous; ovary ovoid, about 0.125 cm. long, glabrous; stigma 0.15 cm. long; nectaries 2, somewhat shorter than the ovary; follicles unknown. BRITISH GUIANA: dryish slopes of Savanna Hills, alt. 4400 ft., Aug., 1928—April, 1929, Tate 836 (NY, TYPE, MBG, photograph and analytical E Je 2. Salpinctes (?) duidae Woodson, in Gleason, Bull. Torrey Bot. Club 58: 454. 1931. Stems relatively stout, 1.5-3.0 dm. tall, scabrous-bullate ; leaves crowded, opposite to subverticillate, essentially sessile, oblong-linear, 3-6 cm. long, 0.2-0.4 em. broad, heavily cori- aceous, the margins strongly revolute, glabrous and lustrous above, opaque and slightly scabridulous to glabrate beneath; inflorescence evidently uniflorous, but the flowers unknown at present; follicles relatively stout, essentially continuous, 10-12 em. long, glabrous; seeds unknown. BRITISH GUIANA: dry ridge tops, Savanna Hills, alt. 4400 ft., Aug., 1928—April, 1929, Tate 805 (NY, TYPE, MBG, photograph). This genus is no better understood than when first published, due to the lack of additional specimens or other data concern- ing it. Atthe time of original publication the genus Dipladenia A. DC. was considered to be distinct from Mandevilla Lindl. upon the basis of the geminate nectaries of the former. Since the discovery of manifest intergradation between these genera and their subsequent consolidation, the status of Salpinctes, also based in part upon geminate nectaries, has been somewhat uneasy. The situation of Salpinctes is not exactly parallel, however, because of its relatively few species, also char- acteristic of its immediate relatives. Hence, the probability of intergradation is somewhat less imminent than amongst the numerous and wide-ranging species of Mandevilla and Dipla- denia. Should a Galactophora be found with salverform cor- ollas, and a Salpinctes with five concrescent nectaries, however, (352) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 181 the existing criteria separating those genera would become too attenuated for practicality. XIII. PELTAsTES Woodson Peltastes Woodson, Ann. Mo. Bot. Gard. 19: 375. 1932. Lactescent, fruticose or suffruticose lianas. Branches op- posite, or the uppermost alternate. Leaves opposite, petiolate, peltate, entire, penninerved, eglandular, firmly membranaceous to subcoriaceous; petioles somewhat girdling at the node into an obscurely appendiculate, stipular ring. Inflorescence lat- eral, opposite, infrequently terminal or subterminal, aggre- gate-dichasial, bracteate, few- to several-flowered. Calyx 5- parted, the lobes subequal to more or less conspicuously dissim- ilar, cleft nearly to the receptacle, imbricated, strikingly folia- ceous, bearing within at the base many indefinitely distributed squamellae. Corolla infundibuliform, large and showy, the proper-tube straight, exappendiculate within, the limb acti- nomorphie, 5-parted, dextrorsely convolute; stamens 5, the anthers connivent and agglutinated to the stigma, consisting of 2 parallel sporangia with a conspicuous protuberant base borne ventrally near the apex of an enlarged, narrowly sagittate, peltate connective; pollen granular. Carpels 2, united at the apex by a slender stylar shaft surmounted by the capitate- fusiform stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, essentially separate. Fol- licles 2, apocarpous, terete, acuminate, dehiscing along the ventral suture, containing many dry, rostrate, apically comose seeds. Type species: Peltastes peltatus (Vell.) Woodson, Ann. Mo. Bot. Gard. 19: 376. 1932. KEY TO THE SPECIES a. Anthers glabrous. b. Calyx-lobes 1.0-1.5 em. long; corolla-lobes broadly and obliquely obovate, obtuse, 2.5-3.0 em. long; plants of southeastern Brazil..1. P. malvaeflorus bb. Calyx-lobes 1.5-2.0 em. long; corolla-lobes dolabriform, acute to acumi- -2.5 em. long; plants of Paraguay.. ..2. P. stemmadeniiflorus aa. Anthers pubescent dorsally. (353) [Vor. 23 182 ANNALS OF THE MISSOURI BOTANICAL GARDEN b. Corolla-throat broadly conieal to campanulate, at least above. e. Corolla 4.0—5.5 em. long; species of South America. d. Calyx-lobes obovate to broadly obovate-oblong; corolla proper-tube 2.2-2.5 em. long; plants of Colombia and Venezuela. .3. P. colombianus dd. Calyx-lobes narrowly oblong to oblong-elliptic; corolla ee -tube 0.9-1.0 em. long; plants of southeastern Brazil........ . P. peltatus ec. Corolla 7.8-8.3 em. long; plants of Panama and Costa Rica. P P. isthmicus bb. Corolla-throat narrowly conical to subtubular-conical; plants of Bolivia SE 6. P. giganteus 1. Peltastes malvaeflorus Woodson, Ann. Mo. Bot. Gard. 19: 376. 1932 Plate 4. Stems relatively stout, densely ferruginous-lanulose when young, eventually becoming glabrate; leaves opposite, broadly ovate, rather shortly acuminate at the apex, base broadly rounded, 10-16 em. long, 7-11 em. broad, firmly membra- naceous, above minutely and rather sparsely ferruginous- puberulent when young, eventually becoming glabrate, beneath minutely and persistently ferruginous-puberulent; petioles 2-4 cm. long, densely and persistently ferruginous-lanulose ; inflorescences lateral, opposite, 3—5-flowered, the peduncle about equalling the length of the subtending petioles, densely ferruginous-lanulose; pedicels 2.3-2.5 em. long, ferruginous- lanulose ; bracts foliaceous, obovate to obovate-lanceolate, 0.5— 2.0 em. long; calyx-lobes oblong to oblong-lanceolate, acute to obtuse, 1.0-1.5 em. long, minutely and rather inconspicuously ferruginous-puberulent without; corolla infundibuliform, gla- brous without, the proper-tube 1.0-1.2 em. long, about 0.6 cm. in diameter at the base, the throat broadly conical-campanulate, dilating almost directly above the insertion of the stamens, 2.0-2.3 em. long, 2.3-2.5 em. in diameter at the orifice, the lobes broadly and obliquely obovate, obtuse, 2.5-3.0 em. long, slightly spreading; anthers 1.1-1.3 em. long, glabrous; ovary about 0.2 em. long, essentially glabrous; nectaries fleshy, essentially equal, about equalling the ovary; stigma 0.3 em. long; follicles relatively stout, 20-25 cm. long, the tips more or less persist- ently connate, glabrous or very minutely papillate; seeds 2.3 em. long, the pale orange coma about 5 em. long. (354) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 183 BRAZIL: PARANA: Valhinos ad marginem silvae primaevae, Nov. 11, 1910, Dusén 10851 (G, MBG, TYPE, S, US); Roca Nova, ad marginem silvae, March 15, 1909, Dusén, 7884 (S); Ponta Grossa, in silvula, Jan. 15, 1909, Dusén 7552 (NY, 8); RIO GRANDE DO SUL: Silveira Martina, prope Santa Maria, in silva primaeva, March 6, 1893, Malme 690 (S, coryer, MBG, photograph); prope São Leopoldo, Febr., year lacking, Dutra 301 (S). 2. Peltastes stemmadeniiflorus Woodson, Ann. Mo. Bot. Gard. 19: 377. 1932. Stems relatively stout, densely ferruginous-lanulose when young, eventually becoming glabrate; leaves opposite, broadly ovate, apex very shortly acute to acuminate, base broadly rounded, 10-30 em. long, 7-15 em. broad, minutely and rather sparsely ferruginous-puberulent when young, eventually be- coming glabrate upon both surfaces; petioles 3-7 cm. long, minutely ferruginous-lanulose when young, eventually becom- ing glabrate ; inflorescences lateral, opposite, 8-10-flowered, the pedunele somewhat shorter than the subtending petioles, minutely ferruginous-hirtellous; pedicels 1.0-1.5 em. long, minutely ferruginous-hirtellous; bracts foliaceous, obovate- lanceolate, 0.4-1.2 em. long; calyx-lobes oblong- to elliptic- lanceolate, acute to acuminate, 1.5-2.0 cm. long, minutely ferruginous-puberulent without; corolla infundibuliform, gla- brous without, the proper-tube 1.3-1.4 cm. long, about 0.4 cm. in diameter at the base, the throat broadly conical to campanu- late, dilating almost directly above the insertion of the stamens, 3.0-3.2 em. long, 1.2-1.3 em. in diameter at the orifice, the lobes broadly dolabriform, acute to acuminate, 2.0-2.5 cm. long, slightly spreading; anthers 1.2-1.3 em. long, glabrous; ovary about 0.25 em. long, essentially glabrous; nectaries fleshy, es- sentially equal, about equalling the ovary; stigma 0.25 cm. long; follicles unknown. PARAGUAY: in altaplanitie et declivibus, Sierra de Amambay, Dec., 1907, Rojas 9838 (V, TYPE, MBG, photograph and analytical drawings); in woods, Mbuvena, Febr., 1931, Jórgensen 4711 (MBG, NY); in reg. fluminis Alto Parana, 1909-10, Fiebrig 5841 (G, US). The distinction of P. malvaeflorus and P. stemmadennflorus may well be questioned. The plants are of much the same gen- eral aspect, although it is believed that the rather slight di- (355) [Vor. 23 184 ANNALS OF THE MISSOURI BOTANICAL GARDEN mensional differences are significant. The corolla-lobes also appear to invite specific segregation. The corolla-throat of the latter species is much narrower than that of the former; and such differences, when taken into consideration with the sup- posedly distinct, if adjacent, ranges of the species have prompted their separate maintenance, at least until more ample, contradictory evidence is forthcoming. 3. Peltastes colombianus Woodson, Ann. Mo. Bot. Gard. 19: 378. 1932. Stems relatively stout, densely ferruginous-lanulose when young, eventually becoming glabrate; leaves opposite, broadly ovate, very shortly subcaudate-acuminate, base broadly rounded to subtruncate, 16-28 cm. long, 10-22 em. broad, above very minutely puberulent when young, soon becoming glabrate, beneath persistently puberulent-papillate; petioles 5-11 cm. long, minutely ferruginous-puberulent to glabrate; inflores- cences 7-15-flowered, the peduncle densely and persistently ferruginous-lanulose; pedicels 1.5-2.0 em. long, minutely fer- ruginous-lanulose; bracts foliaceous, obovate to obovate- oblong, 1.0-1.6 em. long; calyx-lobes obovate to obovate-elliptie, acute to obtuse, 2.3—3.0 em. long, obtuse to very shortly acumi- nate, very minutely and inconspicuously puberulent without; corolla infundibuliform, glabrous without, the proper-tube 2.2-2.5 em. long, about 0.4 em. in diameter at the base, the throat not directly dilated above the insertion of the stamens, differentiated into a more or less distinet upper- and lower- throat, the former 1.0—1.3 em. long, scarcely broader than the proper-tube, the latter abruptly dilated, campanulate, 0.8—1.0 em. long, about 1.2-1.5 em. in diameter at the orifice, the lobes obliquely obovate-dolabriform, obtuse, 1.2-1.4 em. long, widely spreading to somewhat reflexed; anthers 1.2-1.3 em. long, mi- nutely hirtellous dorsally; ovary about 0.2 em. long, minutely ferruginous-tomentulose; nectaries essentially equal, fleshy, about equalling the ovary; follicles unknown. COLOMBIA: MAGDALENA: forest near a stream, alt. 2000 ft., vicinity of Santa Marta, June 27, 1899, Smith 2412 (G, MBG, TYPE, NY, US). (356) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 185 VENEZUELA: CARABOBO: Guaremales, road from Puerto Cabello to San Felipe, in forest, alt. 350 m., July 2, 1920, Pittier 8920 (G, NY, US). Mr. Smith describes his plant as a vine 25 feet tall, with ““corolla green, lobes pale yellowish green.’’ 4. Peltastes peltatus (Vell.) Woodson, Ann. Mo. Bot. Gard. 19: 376. 1932. Echites peltata Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 32. 1827; A. DC. in DC. Prodr. 8: 465. 1844; Muell.-Arg. in Mart. Fl. Bras. 6*: 159. pl. 53, fig. 2. 1860. Echites plicata A. DC. loc. cit. 454. 1844. Echites macrocalyx Muell.-Arg. loc. cit. 160. 1860. Stipecoma peltata (Vell.) Miers, Apoc. So. Am. 134. 1878. Stipecoma plicata (A. DC.) Miers, loc. cit. 1878. Stipecoma pulchra Miers, loc. cit. 135. pl. 18. 1878. Stipecoma mucronata Miers, loc. cit. 1878. Stipecoma macrocalyx (Muell.-Arg.) Miers, loc. cit. 136. 1878. Stipecoma speciosa Miers, loc. cit. 1878. Stipecoma ovata Miers, loc. cit. 137. pl. 19. 1878. Stipecoma parabolica Miers, loc. cit. 1878. Peltastes macrocalyx (Muell.-Arg.) Woodson, loc. cit. 1932. Stems relatively stout, densely ferruginous-lanulose when young, becoming glabrate; leaves opposite, broadly ovate to obovate-oval, apex very shortly acuminate to essentially rounded, base broadly rounded to subtruneate, 5-30 cm. long, 2.5-18 em. broad, firmly membranaceous, above glabrous, or very indefinitely papillate when young, beneath minutely and persistently ferruginous-lanulose; petioles 1.8—10.0 cm. long, minutely and persistently ferruginous-lanulose; inflorescence 5-20 flowered, the peduncle densely ferruginous-lanulose, about equalling the subtending petioles; pedicels 1.0-1.3 cm. long, minutely ferruginous-lanulose; bracts oblong-spathulate, foliaeeous, 0.7-1.5 em. long; calyx-lobes narrowly oblong to oblong-elliptic, acute to acuminate, ka 5-1.7 cm. long, minutely and rather sparsely ferruginous-puberulent without; corolla (357) [Vor. 23 186 ANNALS OF THE MISSOURI BOTANICAL GARDEN infundibuliform, glabrous without, the proper-tube 0.9-1.0 em. long, about 0.4—0.45 cm. in diameter at the base, the throat not directly dilated above the insertion of the stamens, differenti- ated into a more or less distinct upper- and lower-throat, the former 0.6-0.7 cm. long, scarcely broader than the proper-tube, the latter abruptly dilated, broadly campanulate, 0.8-0.9 em. long, about 1.2-1.3 cm. in diameter at the orifice, the lobes ob- liquely dolabriform, shortly acuminate, 1.6-1.7 em. long, some- what spreading; anthers 1.1-1.2 em. long, minutely hirtellous dorsally; ovary about 0.2 cm. long, finely ferruginous-puber- ulent to essentially glabrate ; nectaries fleshy, essentially equal, about equalling the ovary; stigma 0.2 em. long; follicles stout, falcate, usually persistently united at the tips, 15-25 em. long, glabrous, or essentially so; seeds 2.0-2.5 em. long, the pale tawny coma 5.0—5.5 em. long. BRAZIL: MINAS GERAES: Ilheos, 1859-60, Wawra $ Maly 241 (V); Serra de Caldas, Oct. 25, 1873, Mosen 622 (C, S) ; Caldas, prope Rio de Machada, Nov., 1854 Lindberg 191 (S); Caldas, Febr. 24, 1862, Regnell III 883 (C, S, US); Lagóa Santa, date lacking Warming s.n. (C, NY); R10 DE JANEIRO: Serra d’Estrella, Nov. 15, 1874, Glaziou 7753 (US); data incomplete, Schott 5398 (V); Widgren sn. (S); SÃO PAULO: Jaragua, ‘‘Buschwald,’’ Dec. 22, 1912, Brade 5688 (S); prope Rio Grande inter Santos et urbem S. Paulo, 1902, Wacket s.n. (C) ; DATA INCOMPLETE: Mikan s.n. (V) ; Riedel s.n. (G, V) ; Glaziou 4879 (C) ; Glaziou 11188 (C). As specimens of Peltastes from southern Brazil have accu- mulated for study, the validity of P. macrocalyx has appeared more and more dubious. The pubescence of the ovary evi- dently does not distinguish it, as believed by Mueller, since the majority of specimens, doubtless of P. peltatus as shown by other characters, demonstrates to a greater or less degree much the same type of indument. Glabrate individuals are rare. Henee, it appears desirable to consolidate the two species, at least until such time when additional criteria will have been discovered. Miers’ numerous species are all certainly insig- nifieant variations of the frequent and widespread P. peltatus prompted largely by fluctuations of leaf size as well as by the relative abundance or sparsity of vegetative indument, largely reflecting the relative age of the portion of the plants selected as specimens. (358) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 187 5. Peltastes isthmicus Woodson, spec. nov. Fruticosa volubilis; ramulis teretibus crassiusculis juven- tate dense ferrugineo-lanulosis tandem glabratis; foliis op- positis longe petiolatis peltatis late ovatis apice abruptissime brevissimeque acuminatis basi late rotundatis 10-30 em. longis 8-20 cm. latis firmiter membranaceis supra juventate sparse inconspicueque puberulo-papillatis mox glabratis subtus ju- ventate dense ferrugineo-lanulosis tandem glabratis; petiolis 5-12 em. longis ut in ramulo vestitis; inflorescentiis 10—15- floris pedunculo petiolis subaequante minute denseque fer- rugineo-lanuloso; pedicellis 1.8-2.0 cm. longis ut in pedunculo vestitis; bracteis foliaceis oblongo-spathulatis 0.8-1.7 cm. longis; calycis laciniis oblongo-obovatis breviter subeaudato- acuminatis 2.8—3.0 cm. longis basi inconspicue puberulo-papil- latis; corolla infundibuliformi extus glabra tubo proprio 2.3- 2.5 em. longo basi ca. 0.5 em. diametro metiente faucibus basi prope insertionem staminum vix inflatis deinde late campanu- latis 2.0-2.3 cm. longis ostio ea. 1.9-2.0 cm. diametro metiente lobis oblique obovatis haud acuminatis 3.5 cm. longis patulis; antheris 1.2 cm. longis dorso hirtellis; ovario ca. 0.18 cm. longo glabro; nectariis carnosis ovario subaequantibus ; stig- mate 0.2 cm. longo; folliculis crassis falcatis 22-25 em. longis irregulariter ferrugineo-papillatis ; seminibus haud visis. Stems relatively stout, densely ferruginous-lanulose when young, becoming glabrate; leaves opposite, long-petiolate, pel- tate, broadly ovate, apex very abruptly and shortly acuminate, base broadly rounded, 10-30 em. long, 8-20 em. broad, firmly membranaceous, above sparsely and inconspicuously puberu- lent-papillate in youth, soon becoming glabrate, beneath densely ferruginous-lanulose when young, becoming glabrate ; petioles 5-12 cm. long, indument as upon the stem; inflores- cence 10-15-flowered, the peduncle about equalling the sub- tending petioles, minutely and densely ferruginous-lanulose ; pedicels 1.8-2.0 em. long, clothed as upon the peduncle; bracts foliaceous, oblong-spathulate, 0.8-1.7 cm. long; calyx-lobes oblong-ovate, shortly subcaudate-acuminate, 2.8-3.0 em. long, inconspicuously puberulent-papillate at the base without; (359) [Vor. 23 188 ANNALS OF THE MISSOURI BOTANICAL GARDEN corolla infundibuliform, glabrous without, the proper-tube 2.3- 2.5 em. long, about 0.5 em. in diameter at the base, the throat broadly eampanulate, 2.0-2.3 em. long, about 1.9-2.0 cm. in diameter at the orifice, the lobes obliquely obovate, not acumi- nate, about 3.5 em. long, spreading ; anthers 1.2 em. long, hirtel- lous dorsally; ovary about 0.18 em. long, glabrous; nectaries fleshy, about equalling the ovary; stigma 0.2 em. long; follicles stout, faleate, 22-25 em. long, irregularly ferruginous- papil- late; seeds not seen. PANAMA: CHIRIQUI: vicinity of San Felix, alt. 0-120 m., Dee., 1911, Pittier 5125 (US, TYPE, MBG, photograph). Costa RICA: GUANACASTE: Nicoya, 1900, Tonduz s.n. (US). The affinities of this species, notable as the sole Central American representative of a South American genus, are set forth in the key to species. The specimen collected by Tonduz upon the Nicoya Peninsula of Costa Rica is relegated to this species merely upon presumption, as it is completely sterile. 6. Peltastes giganteus Woodson, Ann. Mo. Bot. Gard. 19: 378. 1932. Stems relatively stout, densely ferruginous-lanulose when young, eventually becoming glabrate; leaves opposite, broadly ovate, apex very shortly and abruptly acuminate to obtuse, base broadly rounded, 18-35 em. long, 9-20 cm. broad, firmly membranaceous to subcoriaceous, above minutely ferruginous- lanulose when very young, soon becoming glabrate, beneath persistently ferruginous-lanulose; petioles 7-12 cm. long, min- utely ferruginous-lanulose when young, becoming glabrate ; in- florescence 4—5-flowered, the peduncle somewhat shorter than the subtending petioles, minutely and densely ferruginous- lanulose ; pedicels 1 em. long, ferruginous-lanulose ; bracts foli- aceous, opines lanceolate, shortly acuminate, 0.8-2.0 em. long; calyx-lobes Ona o obtuse, 0.8-1.5 cm. long, minutely puberulent-papillate toward the base, otherwise glabrous; corolla infundibuliform, glabrous without, the proper-tube 1.7— 1.8 em. long, about 0.35 cm. in diameter at the base, the throat narrowly conical or subtubular-conical, dilating almost di- rectly above the insertion of the stamens, 1.8-2.0 em. long, 0.8- (360) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 189 0.9 em. in diameter at the orifice, the lobes obovate-dolabriform, acute, 1.0-1.3 em. long, somewhat spreading; anthers 1.2 cm. long, lanulose dorsally; ovary about 0.25 cm. long, minutely lanulose ; nectaries only slightly fleshy, irregularly cleft, about equalling the ovary ; stigma 0.2 em. long; follicles unknown. BoLivia: data incomplete, Bang 2804 (MBG, TYPE, NY, US); Bang 2404 (C, US 'The tubular corolla-throat sets this species well apart from its congeners, as does the irregularly cleft nectary. XIV. Sripecoma Muell.-Arg. Stipecoma Muell.-Arg. in Mart. Fl. Bras. 6t: 175. 1860; Miers, Apoc. So. Am. 132. 1878, in part; K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 166. 1895. Glabrous, suffruticose lianas. Stems volubile, terete; branches usually opposite below (?), becoming alternate above. Leaves opposite, petiolate, the blade peltate, penninerved, eglandular, entire or slightly sinuous, rigidly chartaceous or subcoriaceous; petioles very obscurely girdling at the node. Inflorescence a lateral, alternate, bostrychoid raceme. Flowers pedicellate, subtended by 1-3 scarious bracts. Calyx 5-parted, the lobes equal or subequal, cleft nearly to the receptacle, sca- rious, bearing few to several internal, glandular squamellae in groups alternate with the lobes. Corolla salverform, the tube cylindrical, somewhat dilated at the insertion of the stamens, the limb actinomorphie, 5-parted, dextrorsely convolute. Sta- mens 5, epipetalous, the anthers coherent, consisting of 2 longi- tudinal, apically convergent, bilocular sporangia borne ven- trally near the apex of an enlarged, sagittate, acutely biauricu- late, dorsally pilose connective, the filament subcylindrical, pilose, the pollen granular. Carpels 2, subinferior, apocar- pous, united at the apex by an elongate, stylar shaft sur- mounted by the fusiform stigma; ovules many, borne upon an axile, binate placenta. Nectaries 5, concrescent or essentially so. Follicles 2, apocarpous, terete, continuous, dehiscing along the ventral suture, containing many dry, rostrate, apically comose seeds; embryo straight, typically dicotyledonous. (361) [Vor. 23 190 ANNALS OF THE MISSOURI BOTANICAL GARDEN Type species: Stipecoma peltigera (Stadelm.) Muell.-Arg. in Mart. Fl. Bras. 6': 176. 1860. 1. Stipecoma peltigera (Stadelm.) Muell.-Arg. in Mart. Fl. Bras. 6*: 176. pl. 53, fig. 1. 1860; Miers, Apoc. So. Am. 133. 1878; K. Sch. in Engl. € Prantl, Nat. Pflanzenfam. 4?: 166. 1895. Echites peltigera Stadelm. Flora 24': Beibl. 21. 1841; À. DC. in DC. Prodr. 8: 447. 1844. Echites tropaeolif olia A. DC. loc. cit. 1844. Glabrous, suffruticose lianas; stems terete, relatively slen- der; leaves opposite, petiolate, the blade peltate, rigidly char- taceous to subcoriaceous, ovate, apex acuminate, base rather broadly rounded, 4-6 em. long, 3.0-4.5 em. broad, the petiole 2.5-3.0 em. long; inflorescence usually somewhat surpassing the length of the subtending leaves, bearing 5-14 ““obscurely rose-colored" flowers; pedicels 0.75 em. long, somewhat ac- crescent in fruit, the subtending bracts scarious, minute ; calyx- lobes ovate, broadly acute, about 0.2 cm. long, scarious; corolla salverform, the tube cylindrical, about 1.5 em. long, somewhat dilated at the insertion of the stamens, the lobes obliquely obovate, acuminate, about equalling the length of the tube, widely spreading; stamens inserted about midway within the corolla-tube, the anthers rather narrowly sagittate, pilose dor- sally toward the tip; ovary oblongoid, glabrous, rather grad- ually produced into the style; stigma fusiform; nectaries con- crescent, variously lobed and cleft, nearly equalling the ovary; follicles as in the generic description. BRAZIL: GOYAZ: ad Serra d’Ourada, date lacking, Pohl 1592 (V, COTYPE, MBG, photograph). The description and dimensions of the flowers and reproduc- tive organs have been taken from the original account by Muel- ler, since the one specimen available for my study is sterile. Mueller cited this species from the provinces of Bahia ( Martius s.n.) and Minas Geraes (St. Hilaire s.n.), as well as listing ad- ditional collections in Goyaz by Weddell (s.n.) and Riedel (932). These specimens have not been available for study, and (362) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 191 the species, apparently rare in the field, has not been repre-. sented in recent collections available for examination. Stipecoma simulates Peltastes in its peltate leaves, but dif- fers in its salverform corolla, scarious calyx-lobes with rela- tively few, alternate squamellae, and more narrowly rostrate seeds. EXCLUDED SPECIES The following are apparently all referable to Peltastes pel- tatus (Vell.) Woodson, Ann. Mo. Bot. Gard. 19: 376. 1932 (Echites peltata Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 32. 1827) : Stipecoma macrocalyx (Muell.-Arg.) Miers, Apoc. So. Am. 136. 1878 ( Echites macrocalyx Muell.-Arg. in Mart. Fl. Bras. 61: 160. 1860). Stipecoma mucronata Miers, Apoc. So. Am. 135. 1878. Stipecoma ovata Miers, Apoc. So. Am. 137. pl. 19. 1878. Stipecoma parabolica Miers, Apoc. So. Am. 137. 1878. Stipecoma peltata (Vell.) Miers, Apoc. So. Am. 134. 1878. Stipecoma plicata (A. DC.) Miers, Apoc. So. Am. 134. 1878 (Echites plicata A. DC. in DC. Prodr. 8: 454. 1844). Stipecoma pulchra Miers, Apoc. So. Am. 135. pl. 18. 1878. Stipecoma speciosa Miers, Apoc. So. Am. 136. 1878. XV. ANGADENIA Miers, char. emend. Angadenia Miers, Apoc. So. Am. 173. 1878, in part. Lactescent, suffruticose or suffrutescent undershrubs. Stems erect, decumbent, or volubile, terete; branches alternate. Leaves opposite, petiolate to subsessile, coriaceous to sub- coriaceous, eglandular, entire, penninerved, the petioles some- what girdling at the node into a slightly dilated, minutely ap- pendiculate, stipular ring. Inflorescence lateral, infrequently terminal, alternate, scorpioid; peduncle simple or dichoto- mously compound at the base, usually more or less conspicu- ously flexuous. Flowers white to pale yellow. Calyx 5-parted, the lobes equal, imbricated, cleft nearly to the receptacle, bear- ing within 1-2 alternate squamellae. Corolla infundibuliform, (363) [Vor. 23 192 ANNALS OF THE MISSOURI BOTANICAL GARDEN the tube straight, narrowly cylindrical below, somewhat below midway staminigerous and abruptly dilated into the broad, campanulate or tubular throat, the limb actinomorphic, 5- parted, dextrorsely convolute. Stamens 5, inserted slightly below midway within the corolla-tube, wholly included; anthers connivent and agglutinated to the stigma, consisting of 2 apically convergent, basally protuberant sporangia borne ventrally near the apex of an enlarged, sagittate, acutely 2- auriculate connective; pollen granular. Carpels 2, united at the apex by an elongate, stylar shaft surmounted by the fusi- form stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, usually concrescent, occasionally more or less separate. Follicles 2, apocarpous, terete, acumi- nate, dehiscing along the ventral suture, containing many dry, narrowly rostrate, apically comose seeds. Type species: Angadenia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878. KEY TO THE SPECIES a. Corolla 3.7—4.5 em. long, the throat tubular; inflorescence usually dichoto- mons Daluw A Ee GE 1. A. Lindeniana aa. Corolla 2.5-3.4 em. long, the throat conical to eampanulate; inflorescence usually A 040 eg at ieee ty Nedstat an ee ers 2. A. Berterit 1. Angadenia Lindeniana (Muell.-Arg.) Miers, Apoc. So. Am. 180. 1878. Rhabdadenia Lindeniana Muell.-Arg. Linnaea 30: 437. 1860. Rhabdadenia Lindeniana Muell.-Arg. angustifolia Muell.-Arg. loc. cit. 438. 1860. Echites Inndeniana (Muell.-Arg.) Griseb. Cat. Pl. Cub. 173. 1866. Suffruticose or suffrutescent lianas; stems relatively slen- der, sparsely pilosulose to glabrate; leaves opposite, distinctly petiolate, broadly elliptic to oblong-lanceolate, apex acuminate, base obtuse, 2.0-6.5 em. long, 1.5-3.0 em. broad, firmly mem- branaceous, glabrous throughout, or very sparsely pilosulose when young, nitidulous above, opaque beneath; petioles 0.5-0.8 em. long, glabrous or glabrate; inflorescence lateral, or infre- (364) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 193 quently subterminal, scorpioid, usually dichotomous below, bearing 6-18 showy, cream-colored flowers; peduncle usually about twice surpassing the subtending leaves, glabrous, or rarely very minutely puberulent; pedicels 1.0-1.2 cm. long, somewhat accrescent in fruit, glabrous or essentially so; bracts scarious, minutely ovate-lanceolate; calyx-lobes ovate-lanceo- late, acuminate, 0.3-0.4 cm. long, scarious, glabrous, the squamellae 2-3; corolla infundibuliform, glabrous without, the proper-tube 0.7—1.3 em. long, about 0.2-0.25 cm. in diameter at the base, the throat tubular, 1.5-2.0 cm. long, about 0.6-0.8 cm. in diameter at the orifice, the lobes obliquely obovate, 1.2-1.5 em. long, widely spreading; anthers oblong-elliptie, narrowly sagittate, 0.45-0.5 em. long, glabrous or essentially so; ovary ovoid, about 0.15 em. long, glabrous; nectaries concrescent or essentially so, nearly twice surpassing the ovary ; stigma about 0.1 em. long ; mature follieles unknown. CUBA: MATANZAS: Arroyo Grande ad Jagüey, alt. 600 m., April, 1889, Eggers 5297 (B, US); ORIENTE: wooded hills, vicinity of Santiago, Mareh 10-25, 1912, Britton Britton $ Cowell 12848 (MBG, NY); in dense woods, exact locality lacking, July 21, 1859, Wright 1376 (S, V); DATA INCOMPLETE: Swartz s.n. (S); Linden 1699 (V). JAMAICA: upper slopes of Mt. Diablo, alt. 500-800 m., Febr. 25-28, 1920, Mazon $ Killip 394 (US). Relegation of Maxon & Killip 394 to this species is not made with great assurance, as the corolla, preserved in only one in- stance, is much narrower, with respect to both proper-tube and throat, than in the Cuban specimens. Furthermore, it must be recalled that the apocynaceous floras of Cuba and Jamaica, at least in the highlands, are ordinarily specifically distinct. 2. Angadenia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878. Echites Berterii A. DC. in DC. Prodr. 8: 447. 1844. Echites Sagraei A. DC. loc. cit. 450. 1844. Echites ferruginea A. Rich. in Sagra, Hist. Cuba 11: 92. 1850. Rhabdadenia Berteri (A. DC.) Muell.-Arg. Linnaea 30: 435. 1860. (365) [Vor. 23 194 ANNALS OF THE MISSOURI BOTANICAL GARDEN Rhabdadenia Sagraei (A. DC.) Muell.-Arg. loc. cit. 1860. Rhabdadenia cubensis Muell.-Arg. loc. cit. 1860. Echites cubensis (Muell.-Arg.) Griseb. Cat. Pl. Cub. 172. 1866. Angadema Sagraei (A. DC.) Miers, loc. cit. 181. 1878. Angadenia Havanensis Miers, loc. cit. 1878. Angadenia Cubensis (Muell.-Arg.) Miers, loc. cit. 182. 1878, where erroneously referred to Grisebach. Secondatia ferruginea (A. Rich.) Miers, loc. cit. 227. 1878. Rhabdadenia corallicola Small, Bull. N. Y. Bot. Gard. 3: 434. 1905. Suffrutescent lianas, not rarely erect or suberect in their northern range; stems sparsely pilosulose to glabrate; leaves opposite, shortly petiolate to subsessile, ovate-elliptie to ob- long-linear, apex acuminate to obtuse, not infrequently mu- cronulate, base obtuse, rarely obscurely cordate, 1—7 cm. long, 0.5-2.0 em. broad, membranaceous to subcoriaceous, above glabrous, usually somewhat nitidulous, beneath opaque, gla- brous, or minutely puberulent along the midrib; petioles 0.1-0.5 em. long, occasionally essentially suppressed; inflores- cence lateral in volubile specimens, subterminal to terminal in erect specimens, scorpioid, usually simple, infrequently dichotomous, bearing 5-30 cream-colored or yellowish flowers ; peduncle twice or more surpassing the subtending leaves, glabrous to very minutely puberulent ; pedicels 1.0-1.5 em. long, glabrous to very minutely puberulent; bracts lanceolate, 0.1 cm. long or less, scarious; calyx-lobes ovate-lanceolate, acumi- nate, 0.3-0.5 em. long, scarious, glabrous to very minutely puberulent-papillate without, the squamellae 1-2-3; corolla infundibuliform, glabrous to very indefinitely papillate with- out, the proper-tube 0.5-0.8 cm. long, about 0.2 em. in diameter at the base, the throat broadly conical to campanulate, 1.0-1.5 cm. long, about 0.5-0.9 cm. in diameter at the orifice, the lobes obliquely obovate-dolabriform to subreniform-dolabriform, 1.0-1.3 em. long, widely spreading; anthers subtrigonal, obtuse to acute, narrowly sagittate, 0.3-0.4 em. long, indefinitely pu- berulent-papillate dorsally; ovary ovoid, about 0.1 em. long, (366) 1936] WOODSON—STUDIES IN THE APOCYNACEAE, IV 195 glabrous; nectaries concrescent, about equalling the ovary, ex- tremely variable and not infrequently more or less separate; stigma about 0.1 cm. long; follicles slender, divaricate, con- tinuous, 5-10 cm. long, glabrous; seeds 0.5 cm. long, the pale yellowish coma about 2 cm. long. UNITED STATES: FLORIDA: Miami, Dade Co., June, 1877, Garber 11957 (MBG); age Miami, April 30, 1930, Duckett s.n. (MBG); in dry sandy marl, Goulds, Jan. 1930, Moldenke 399a (MBG, NY); Big Pine Key, May, 1891, Simpson 251 iB S). BAHAMA ISLANDS: near Nassau, April 25, 1903, Curtiss 178 (MBG, US); in pinetis, New Providence, Febr. 28, 1888, Eggers 4315 (US). CUBA: ISLA DE PINOS: near Nueva Gerona, March 10-April 2, 1904, Curtiss 395 BG); PINAR DEL RIO: palm barrens, west of Guane, Nov. 21-22, 1911, Shafer 10428 (MBG, US); St. Yago de Cuba [Santiago], 1844, Linden 2167 (V); moun- tains near El Guama, among grass on hillside, March 9, 1900, Palmer $ Riley 207 (US); trailing through bushes of sand swamp near seashore, near Coloma, March 18, 1900, Palmer $ Riley 337 (US); HABANA: Guanabacoa, date lacking, Poeppig 537 (V); DATA INCOMPLETE: Wright 2955 (MBG, S, V); Sagra s.n. (V). HISPANIOLA: SANTO DOMINGO: in sylvis, alt. 100 m., April 26, 1887, Eggers 1686 (B, US); La Jina, Distr. Moncion, Prov. Monte Cristy, April 19, 1933, Valeur 978 (MBG). The specimens from Santo Domingo demonstrate something of a transition to 4. Lindeniana in the shape of the corolla- throat, which is nearly subtubular. The anthers are subtri- gonal, however, as in specimens of A. Berterii as interpreted above, from Cuba, the Bahamas, and Florida. This species is bewildering in its variations, particularly with regard to pos- ture and shape of foliage. It would doubtless be indiserete to segregate upon the present state of our knowledge. The genus Angadenia as conceived by Miers is a hodge-podge of various elements, chiefly referable to Odontadenia, as the following list of excluded species testifies. After the removal of numerous species included within it by Miers but more prop- erly referable to older genera, a distinct element centering about Echites Berterw A. DC. and Rhabdadenia Lindeniana Muell.-Arg. remains as a residue of manifestly related entities which form the genus Angadenia as here emended. As out- lined by Mueller (loc. cit. 1860), these species differ from Rhabdadenia chiefly in the presence of calycine squamellae, although certain poorly defined characters of general aspect (367) [Vor. 23 196 ANNALS OF THE MISSOURI BOTANICAL GARDEN might be mentioned in support of their segregation, among the more prominent of which is the extensive, zig-zag, scorpioid inflorescence of the former. EXCLUDED SPECIES Angadenia Almadensis (Stadelm.) Miers, Apoc. So. Am. 179. 1878 (Echites almadensis Stadelm. Flora 24*: Beibl. 28. 1841) = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4^: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792, not R. & P.). Angadenia Amazonica (Stadelm.) Miers, Apoc. So. Am. 175. 1878 (Echites amazonica Stadelm. Flora 24!: Beibl. 50. 1841) = Odontadenia verrucosa (H € S.) K. Sch. ex Mgf. in Pulle, El. Surinam 4: 53. 1932 (Echites verrucosa R. & S. Syst. 4: 795. 1819). Angadema cognata (Stadelm.) Miers, Apoc. So. Am. 176. 1878 (Echites cognata Stadelm. Flora 24!: Beibl. 79. 1841) = Odontadenia cognata (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 18: 546. 1931. Angadenia coriacea (Benth.) Miers, Apoc. So. Am. 177. 1878 (Echites coriacea Benth. in Hook. Jour. Bot. 3: 249. 1841) = Odontadenia geminata (R. & S.) Muell.-Arg. in Mart. Fl. Bras. 6': 119. 1860 (Echites geminata R. € S. Syst. 4: 795. 1819). Angadenia Cururu (Mart.) Miers, Apoc. So. Am. 175. 1878 (Echites Cururu Mart. in Buchn. Rep. Pharm. 101. 1830) = Odontadenia puncticulosa (A. Rich.) Pulle, Enum. Pl. Surinam 383. 1906 (Echites puncticulosa A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792). Angadema elegans (Benth.) Miers, Apoc. So. Am. 178. 1878 (Echites elegans Benth. in Hook. Jour. Bot. 3: 249. 1841) = Odontadenia geminata (R. & S.) Muell.-Arg. in Mart. Fl. Bras. 61: 119. 1860 (Echites geminata R. & S. Syst. 4: 795. 1819). Angadenia elliptica Miers, Apoc. So. Am. 180. 1878. Based upon Gardner 2663 collected at Paranagua, State of Parana, Brazil, which has not been available for study. The unusually meager description appears to refer to Odontadenia gracilipes (368) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 197 (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 22: 294. 1935 (Echites gracilipes Stadelm. Flora 24*: Beibl. 22. 1841), which is at present known only from the State of Minas Geraes. Angadenia geminata (R. & S.) Miers, Apoc. So. Am. 178. 1878 (Echites geminata R. & S. Syst. 4: 795. 1819) = Odonta- denia geminata (R. € S.) Muell.-Arg. in Mart. Fl. Bras. 6*: 119. 1860. Angadenia grandifolia (Stadelm.) Miers, Apoc. So. Am. 175. 1878 (Echites Cururu Mart. var. P grandifolia Stadelm. Flora 241: Beibl. 79. 1841) = Odontadenia puncticulosa (A. Rich.) Pulle, Enum. Pl. Surinam, 383. 1906 (Echites puncticulosa A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792). Angadenia hypoglauca (Stadelm.) Miers, Apoc. So. Am. 173. 1878 (Echites hypoglauca Stadelm. Flora 24*: Beibl. 23. 1841) = Odontadenia hypoglauca (Stadelm.) Muell.-Arg. in Mart. Fl. Bras. 6': 118. 1860. Angadenia latifolia (Muell.-Arg.) Miers, Apoc. So. Am. 176. 1878 (Anisolobus amazonicus (Stadelm.) Muell.-Arg. f lati- folàus Muell.-Arg. in Mart. Fl. Bras. 61: 114. 1860) = Odonta- denia verrucosa (R. & S.) K. Sch. ex Mgf. in Pulle, Fl. Surinam 4: 53. 1932 (Echites verrucosa R. & S. Syst. 4: 795. 1819). Angadenia majuscula Miers, Apoc. So. Am. 174. 1878 = Odontadenia hypoglauca (Stadelm.) Muell.-Arg. in Mart. Fl. Bras. 61: 118. 1860 (Echites hypoglauca Stadelm. Flora 24!: Beibl. 23. 1841). Angadenia nitida (Vahl) Miers, Apoc. ‘So. Am. 177. 1878 (Echites nitida Vahl, Eclog. 2: 19. pl. 13. 1798) = Odontadenia nitida (Vahl) Muell.-Arg. in Mart. Fl. Bras. 6t: 118. 1860. Angadenia pandurata (A. DC.) Miers, Apoc. So. Am. 182. 1878 (Echites pandurata A. DC. in DC. Prodr. 8: 458. 1844) = Fernaldia pandurata (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 48. 1932. Angadema Poppigu ( Muell.-Arg.) Miers, Apoc. So. Am. 179. 1878 (Odontadema Poeppigii Muell.-Arg. in Mart. Fl. Bras. 6': 119. 1860) = Odontadenia geminata (R. & S.) Muell.-Arg. loc. cit. 119. 1860 (Echites geminata R. & S. Syst. 4: 795. 1819). (369) [Vor. 23 198 ANNALS OF THE MISSOURI BOTANICAL GARDEN Angadenia Prieurü (A. DC.) Miers, Apoc. So. Am. 182. 1878, err. typ. (Echites Prieurei A. DC. in DC. Prodr. 8: 458. 1844) = Mandevilla subspicata (Vahl) Mgf. Rec. Trav. Bot. Néerl. 22: 380. 1926 (Echites subspicata Vahl, Eclog. Am. 2: 18. 1798). Angadenia pruinosa Miers, Apoc. So. Am. 177. 1878. Based upon Gardner 2232, collected at Oeiras, Para, Brazil. Miers expressed the opinion (loc. cit.) that ‘‘The follicles and seeds exactly correspond with those of A. hebecarpus, showing that it belongs to this genus.” Miers evidently referred to Odonta- denia lutea (Vell.) Mgf. in Fedde, Rep. Sp. Nov. 20: 24. 1924 (Echites lutea Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 25. 1827) = Anisolobus hebecarpus Muell.-Arg. in Mart. Fl. Bras. 6!: 111. pl. 33. 1860, although he did not formally transfer that species to Angadenia, doubtless through oversight. Angadenia reticulata Miers, Apoc. So. Am. 179. 1878 = Mandevilla scabra (R. & S.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites scabra R. € S. Syst. 4: 795. 1819). Angadenia Sprucei (Muell.-Arg.) Miers, Apoc. So. Am. 176. 1878 (Anisolobus Sprucei Muell.-Arg. in Mart. Fl. Bras. 6!: 114. 1860) = Odontadenia verrucosa (R. & S.) K. Sch. ex Mgf. in Pulle, Fl. Surinam 4: 53. 1932 (Echites verrucosa R. € $. Syst. 4: 795. 1819). Angadenia sylvestris (A. DC.) Miers, Apoc. So. Am. 174. 1878 (Echites sylvestris A. DC. in DC. Prodr. 8: 464. 1844) = Odontadenia Hoffmannseggiana (Steud.) Woodson, ex Gleason & A. C. Smith, Bull. Torrey Bot. Club 60: 392. 1933 (Echites Hoff mannseggiana Steud. Nomencl. ed. 2. 1: 539. 1840). Angadema Valenzuelana (A. Rich.) Miers, Apoc. So. Am. 181. 1878 ( Echites Valenzuelana A. Rich. in Sagra, Hist. Cuba 11: 93. 1850) - Neobracea Valenzuelana ( A. Rich.) Urb. Symb. Ant. 9: 241. 1924. XVI. Urecuitres Muell.-Arg. Urechites Muell.-Arg. Bot. Zeit. 18: 22. 1860; Benth. & Hook. Gen. Pl. 2: 727. 1876; Miers, Apoc. So. Am. 124. 1878; K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895. (370) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 199 Chariomma Miers, loc. cit. 110. 1878. Lactescent fruticose or suffruticose lianas. Stems volubile, infrequently suberect, terete; branches alternate, or opposite below. Leaves opposite or rarely subverticillate, petiolate, membranaceous to subcoriaceous, eglandular, entire, penni- nerved, the petioles subtended by an inconspicuously appen- diculate, interpetiolar, stipular ring. Inflorescence lateral to subterminal or terminal, simply scorpioid, bearing few to many showy, yellowish or cream-colored flowers; peduncle straight to slightly flexuous, conspicuously bracteate. Calyx 5-parted, the lobes equal to subequal, imbricated, cleft nearly to the receptacle, bearing within paired alternate, or numerous in- definitely distributed squamellae. Corolla infundibuliform, the tube straight, about midway, or somewhat lower, stam- inigerous, and abruptly dilated into a broad, tubular throat, the limb actinomorphic, 5-parted, dextrorsely convolute. Sta- mens 5, wholly included, or the appendages somewhat exserted ; anthers connivent and agglutinated to the stigma, consisting of 2 apically convergent, basally protuberant sporangia borne ventrally near the apex of an enlarged, sagittate, acutely 2-auriculated connective bearing, except rarely, conspicuous, spirally coiled, linear, apical appendages; pollen granular. Carpels 2, united at the apex by an elongate, stylar shaft sur- mounted by the capitate-fusiform stigma; ovules many, sev- eral-seriate, borne upon an axile, binate placenta. Nectaries 5, concrescent or essentially separate. Follicles 2, apocarpous, terete, acuminate, dehiscing along the ventral suture, con- taining many dry, narrowly rostrate, apically comose seeds mingled with subtending, subscaphiform, placental chaff. Type species: Urechites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907. KEY TO THE SPECIES a. Calyx-lobes linear-lanceolate, acuminate, usually equalling or slightly sur- passing the proper-tube of the corolla; squamellae numerous, indefi- nitely distributed; nectaries essentially separate; plants of southern peninsular Florida, the Bahama Islands, the Antilles, rarely loeally in northern Atlantic coastal Central Amerien lees 1. U. lutea (371) [ Vor. 23 200 ANNALS OF THE MISSOURI BOTANICAL GARDEN aa. Calyx-lobes ovate, obtuse, much shorter than the proper-tube of the corolla ; squamellae paired, alternate with the calyx-lobes; dE concres- cent; plants of northern Central America................2. 2. U. Andrieuzii 1. Urechites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907. Vinca lutea L. Cent. IT. Pl. 12. 1756. Echites suberecta Jacq. Enum. Pl. Carib. 13. 1760. Echites domingensis Jacq. Ie. Rar. 1: 6. pl. 26. 1782. Echites heterophylla J. F. Gmel. Syst. 2: 437. 1791, not iq. Echites barbata Desv. ex Ham. Prodr. 30. 1825. Echites Catesbaei G. Don, Gen. Hist. 4: 74. 1838. Haemadictyon suberectum (Jaeq.) G. Don, loc. cit. 83. 1838. Neriandra suberecta (Jaeq.) A. DC. in DC. Prodr. 8: 422. 1844 Dipladenia flava Hook. Bot. Mag. 79: pl. 4702. 1853. Echites Peltieri Loud. Encycl. 1541. 1855, nom. nud. in synon. Echites Andrews? Chapm. Fl. So. U. S. 359. 1860. Urechites Jaegeri Muell.-Arg. Linnaea 30: 443. 1860. Urechites suberecta (Jacq.) Muell.-Arg. loc. cit. 444. 1860; Miers, Apoc. So. Am. 125. 1878. Urechites suberecta (Jacq.) Muell.-Arg. B glabrata Muell.- Arg. loc. cit. 444. 1860. Urechites suberecta (Jacq.) Muell.-Arg. y rotundifolia Muell.-Arg. loc. cit. 1860. Echites neriandra Griseb. Fl. Brit. W. I. 415. 1861. Echites jamaicensis Griseb. loc. cit. 416. 1861. Chariomma surrecta Miers, loc. cit. 111. 1878. Nerium sarmentosum P. Browne, ex Miers, loc. cit. 111. 1878, err. in synon. Chariomma Domingensis (Jacq.) Miers, loc. cit. 112. 1878. Chariomma mucronulata Miers, loc. cit. 112. 1878. Chariomma flava (Hook.) Miers, loc. cit. 113. 1878. Chariomma verticillata Miers, loc. cit. 1878. Chariomma scandens Miers, loc. cit. 114. 1878. 372 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 201 Rhabdadenia laxiflora Miers, loc. cit. 120. 1878. Rhabdadema barbata (Desv.) Miers, loc. cit. 123. 1878. Laubertia urechites Griseb. ex Miers, loc. cit. 125. 1878, err in synon. Urechites Neriandra Griseb. ex Miers, loc. cit. 126. 1878, err in synon. Laseguea Jaegeri (Muell.-Arg.) Miers, loc. cit. 254. 1878. Laseguea pubiflora Miers, loc. cit. 253. 1878. Echites obovata Sesse & Moc. Fl. Mex. 43. 1887, not Nees, fide Urb. Symb. Ant. 4: 496. 1903. Urechites Andrewsii (Chapm.) Small, Fl. Southeast. U. S. 936. 1903. Urechites dolicantha Urb. Symb. Ant. 6: 38. 1909. Urechites pinetorum Small, Addisonia 4: 21. pl. 131. 1919. Urechites lutea (L.) Britton var. angustif olia Ekm. & Hel- wig, Arkiv f. Bot. 22A"; 46. 1929. An extremely variable species. Stems volubile or occasion- ally suberect, relatively stout, glabrous to densely pilose; leaves opposite, shortly petiolate, oblong-linear to subrotund, apex shortly acuminate to obtuse or rounded, base obtuse to obscurely cordate, 3-9 em. long, 0.5-6.0 em. broad, membrana- ceous to subcoriaceous, either surface glabrous to pilose; pet- ioles 0.2-1.2 cm. long; inflorescence lateral, alternate, occasion- ally subterminal or terminal, the peduncle usually somewhat surpassing the subtending leaves, occasionally somewhat shorter, generally more or less pilose, rarely glabrate, bearing few to relatively many handsome, yellowish or cream-colored flowers; pedicels 1.0-1.5 cm. long, usually somewhat pilosulose, less frequently glabrate; bracts foliaceous, ovate to ovate- lanceolate, 0.2-0.8 em. long; calyx-lobes linear-lanceolate, acu- minate, 0.8-1.7 cm. long, usually more or less pilose or pilosu- lose, infrequently glabrate, the squamellae numerous, indefi- nitely distributed; corolla infundibuliform, glabrous to laxly pilose without, the proper-tube rather narrowly cylindrical, 0.6-1.5 em. long, about 0.2 cm. in diameter at the base, the throat broadly tubular to tubular-subconical, 1.5-3.5 em. long, about (373) [Vor. 23 202 ANNALS OF THE MISSOURI BOTANICAL GARDEN 0.8-1.2 em. in diameter at the orifice, the lobes obliquely obo- vate, 1.8-3.0 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers 0.5-0.55 cm. long, minutely puberulent dorsally, the apical appendages 1.3-1.7 em. long, usually more or less exserted; ovary ovoid, about 0.1 em. long, glabrous or essentially so; nectaries essentially separate or more or less concrescent at the base, about equal- ling the ovary ; stigma 0.125 em. long; follicles relatively stout, acuminate, continuous, 8-20 cm. long, glabrous to more or less densely pilosulose ; seeds 0.7-1.1 em. long, the pale tawny coma 1.5-3.0 em. long, the placental chaff 1.0-1.3 em. long. UNITED STATES: FLORIDA: shrub on midden, Coconut, Lee Co., Febr. 19, 1930, Moldenke 5784 (NY); in dry sandy soil along roadside, near Caxambos, Collier Co., April 17, 1930, Moldenke 1006a (NY); rieh thiekets, Upper Metaeombe Key, in eoral soil, July, year pines. Curtiss 2267 G, S) ee west 0 Coconut Grove, May 1 wa p Set 8786 (NY); in Sears between Homestead and Camp Jackson, ae , Small $ Wilson 1756 (NY); Weg Big Pine Key, Dec. 2, 1912, SECH SE (NY); ; pinelands between Miami and Kendall Station, Nov. 5, 1906, Small $ Carter 2654 (NY); hammocks, Grassy Key, Jan. 28-29, 1909, Small $ Carter 3116 (NY); Key West, Aug. > ed ae sn. (US) May B (MBG, NY, US); coppice, near Deep Creek, Long Bay Cays section, Andros, Jan. 20-22, 1910, Small 4 Carter 8605 (NY, US); serub, Mathew Town, Oct. 10, 1904, Nash $ Taylor 898 (NY); Er Cockburn Town and vicinity, Watling’s, March 12-13, 1907, Britton $ Millspaugh 6093 (NY); Spring Point, Acklin’s, Dec. 21, amos 6, 1906, Brace 4273 (NY); eoppiee, Grand ee Aug. 27-Sept. 1, 1905, h §& ee 3773 (NY, US); West End, Great Bahama, April 16-May 8, 1905, poni 3601 (NY); Crooked Is., Nov., 1881, potu s.n. (MBG); Gov- ernor’s Harbor, Nov. 15, 1890, Hitchcock s.n. (MBG) CUBA: PINAR DEL RIO: climbing over bushes of sandy swamp near the shore, near Coloma, March 18, 1900, Palmer $ Riley 364 (NY, US); oak grove north of San Diego de los Baños, June 28, 1915, Leon 5108 (NY) ; ISLA DE PINOS: swamp, climber over bushes, near Nueva Gerona, July 5, 1900, Palmer $ Riley 1010 (US); on coral soil, north of Caleta Grande, May 22, 1910, Jennings 463B (NY); HABANA: in swamp near seashore, near Playa de Marianao, June 17, 1900, Palmer $ Riley 833 (US); near Morro Castle, May 20, 1908, Leon 712 (NY); MATANZAS: Cardenas, Sept. 1, 1908, Britton $ Wilson 190 (NY); SANTA CLARA: Cienequita, May 9, 1895, Combs 13 (MBG, US); coastal thicket, Rio Arimao, March 22, 1910, Britton 4 Wilson 5809 (NY); CAMAGUEY: vicinity of La Gloria, Febr. 4, 1909, Shafer 297 (NY, US) ; Cayo Paredon Grande, Oct. 25, 1909, Shafer 2748 (NY, US) ; ORIENTE woods, Alto Cedro to Paso Estancia, April 28, 1909, Shafer 1624 (NY, US); coastal thicket, Fisherman’s Point, Guatanamo Bay, March 17-30, 1909, Britton 2110 (NY, SUB) JAMAICA: roadside, Constant Spring, Aug. 29-31, 1907, Britton 961 (NY); (374) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 203 banks, Balaclava, Cockpit Country, Sept. 13-18, 1906, Britton 423 (NY); climbing on SET banks, vicinity of Montego Bay, March 28-30, 1920, Maxon $ Killip 1440 (NY, US); rocks, Navy Island, July 14, 1897, Fredholm 3163 (US); climb- ing over shrubs, Holly Mts., Mt. Diablo, alt. 3000 ft., 3s 31, 1905, Harris 900? (NY); Lititz Savanna, July 7, 1914, Harris 11732 (MBG, NY, US); near Castle- ton, alt. 500 ft., June 26, 1915, Harris 12085 (MBG, um US). HISPANIOLA: SANTO DOMINGO: sandy roadside, Haina, April 3, 1921, Faris 119 (US); district of Moncion, Prov. Monte Cristy, Aug. 12, 1929, Valeur 175 (MBG, ; Azua, March, 1913, Rose Fitch $ Russell 3960 (US); Paradis, pr. Barahona, in Mene litoralibus, Dec., 1909, Tuerckheim 2688 (NY, US); HAITI: windswept roeky beach east of Cabaret, Jan. 14, 1929, Leonard 4 Leonard 11986 (US); dry thickets east of harbor, vicinity of La Vallee, Tortue Island, Jan. 5, 1929, Leonard d Leonard 11642 (US); Massif de la Hotte, J joue between Sources-Chaudes and Source-Cahouane, July 4, 1928, Ekman 10224 (US); Bayeux, near Port Margot, Aug. 4, 1903, Nash 135 (NY); La Brande, alt. 700 ft., Aug. 14, 1905, Nash $ Tay- lor 1683 (NY Porto Bide: Salinas de Cabo-Rojo, in sylvis litoralibus, Febr. 8, 1885, Sintenis 273C (US); in declivibus, Cabeza de San Juan, Sept. 14, 1885, Sintenis 1891 (US); climbing over bushes in thicket, sandy shore, Playa de Esperanza, Vieques Island, Febr. 7, 1914, ag 2737 (NY, US); sandy plain, Icacos Cay, Jan. 30, 1923, Brit- ton 7159 A VIR ISLANDS: ST. THOMAS: rocky point, Cowell Point, Jan. 31—Febr. 4, 1913, BE. Pew Wes 91 (NY, US); ST. CROIX: Grenard, Nov. 14, 1925, Thomp- son 1005 (NY). LEEWARD ISLANDS: TORTOLA: coastal thicket, Fish Bay to Road Town, Febr. 13- 17, 1918, Britton $ Shafer 911 (NY, US); ANEGADA: sandy plain, West End, Febr. 19-20, 1913, Britton $ Fishlock 946 (NY, US); MARTIN: March 28, 1926, Good- win $ Goodwin 4 (NY); ANGUILLA, April 6, Con Goodwin $ Goodwin 16 (NY); ST. CHRISTOPHER: gulch, Canada Estate, Sept. 8—Oet. 5, 1901, Britton $ Cowell 365 (NY). Also reported from the vicinity of Santa Marta, Colombia, by Purdie. This species is very unstable, even in e? an im- portant character as the linear appendages of the anther tips, which may occasionally be totally lacking, or extremely short. This is apparently the most frequently collected of the Amer- ican Echitoideae, and space permits only the greatly abridged representation cited above. Although segregation may appear an attractive solution of the vexatious variability, a wide selec- tion of specimens, such as that which has been available for these studies, should prove an effective deterrent. 2. Urechites Andrieuxii Muell.-Arg. Linnaea 30: 442. 1860; Miers, Apoc. So. Am. 125. Stems relatively stout, minutely hirtellous when young, (375) [Vor. 23 204 ANNALS OF THE MISSOURI BOTANICAL GARDEN eventually becoming glabrate; leaves opposite, petiolate, ob- long- to ovate- or obovate-elliptie, apex rather shortly acumi- nate, base obtuse to rounded, 5-12 em. long, 2.5-7.0 em. broad, membranaceous to subeoriaceous, glabrous above, glabrous to very finely puberulent beneath; petioles 0.8-2.0 cm. long; in- florescence lateral, alternate, the peduncle usually somewhat surpassing the subtending leaves, bearing few to numerous handsome, yellow or cream colored flowers, glabrous to very inconspicuously puberulent ; pedicels 1.5-2.2 em. long, glabrous or essentially so; bracts minute, scarious, caducous; corolla infundibuliform, SON IT the proper-tube 1.0-1.5 em. long, about 0.3 em. in diameter at the base, slightly constricted toward the insertion of the stamens, the throat broadly tubular to subtubular, 2.5-3.0 em. long, 0.5-0.8 em. in diameter at the orifice, the lobes obliquely obovate, 1.5-2.3 cm. long, spread- ing; stamens inserted at the base of the corolla-throat, the anthers 0.6-0.7 cm. long, essentially glabrous to very minutely puberulent-papillate dorsally, the apical appendages linear, 0.7—0.9 em. long ; ovary ovoid, 0.15 em. long, glabrous ; nectaries concrescent, about twice surpassing the ovary; stigma 0.2 cm. long; calyx-lobes ovate, obtuse, 0.4-0.5 cm. long, glabrous or very indefinitely papillate, the squamellae in alternate pairs; follicles relatively stout, acuminate, falcate, continuous, 15- 28 cm. long, glabrous; seeds 1.5-1.8 cm. long (including the rostrum), the pale tawny coma 2.5-3.5 em. long. MEXICO: CAMPECHE: over trees EE e river, Champoton, July 7-15, 1932, Steere ere G); Tuxpefia, Dec. 3, 1931, Lundell 842 (MBG). BRITISH HONDURAS: occupied peat Belize, June, 1933, Lundell 4056 (MBG); Jones Bank, M 1933, SE NEE MBG); Santa Rita, Oct., 1933, Gentle 800 (MBG); Northern River, Dec., 1933, Gentle 1037 (MBG); Honey Camp, Orange Walk, Nov., 1928, Lundell ps (US); small climber growing in mangrove swamp, Stann [E Dee. 8, 1929, Schipp 491 ano GUATEMALA: PETEN: La Libe rtad, May 31, 1933, Lundell 3533 eae en Polol, May 2 px pu 3445 (MBG); El Paso, common vine in lov April 24, Gen Lundell 1524 (MBG); Santa Cruz, March 27-28, 1931, Ee 12366 (MBG); ZACAPA: Gualan, alt. 620 ft., June 16, 1909, Deam 6320 (MBG, NY). The seminal rostrum of this species is much longer than that of U. lutea, and the placental chaff is less delicate, more defi- (376) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 205 nitely scaphiform, and with much less tendency to scale from the placenta upon dehiscence of the follicles. EXCLUDED SPECIES Urechites Karwinsku Muell.-Arg. Linnaea 30: 440. 1860 = Fernaldia pandurata (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 48. 1932 (Echites pandurata A. DC. in DC. Prodr. 8: 458. 1844). XVII. Ruaspapenta Muell.-Arg. Rhabdadenia Muell.-Arg. in Mart. Fl. Bras. 61: 173. 1860; Miers, Apoc. So. Am. 118. 1878; K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 170. 1895. Lactescent, suffruticose or suffrutescent lianas. Stems volubile or rarely suberect, terete; branches usually opposite below, becoming alternate above. Leaves opposite, petiolate to subsessile, the blade coriaceous to membranaceous, entire, penninerved, eglandular; petioles somewhat girdling at the node into a rather inconspicuous, minutely appendiculate, stipular ring. Inflorescence a lateral or subterminal, alter- nate, greatly reduced, dichasial cyme, frequently uniflorous. Flowers pedicellate, subtended by a solitary bract. Corolla infundibuliform, the proper-tube straight, narrowly cylindri- cal, the throat conical or tubular, the limb actinomorphic, 5- parted, dextrorsely convolute, widely spreading. Calyx 5- parted, the lobes equal or subequal, cleft nearly to the recep- tacle, subeglandular within; stamens 5, epipetalous, the anthers coherent, consisting of 2 longitudinal, apically con- vergent, bilocular sporangia borne ventrally near the apex of an enlarged, sagittate, acutely biauriculate, dorsally pilose connective, the filament subcylindrical, puberulent, the pollen granular. Carpels 2, sessile or subinferior, apocarpous, united at the apex by an elongate, stylar shaft surmounted by the fusi- form, apically pilose stigma; ovules many, anatropous, borne upon an axile, binate placenta. Nectaries 5, separate or some- what concrescent at the base. Follicles 2, apocarpous, terete, continuous, dehiscing along the ventral suture, containing (377) [Vor. 23 206 ANNALS OF THE MISSOURI BOTANICAL GARDEN many dry, subscaphiform, rostrate, apically comose seeds; embryo straight, typically dicotyledonous. Type species: Rhabdadenia Pohl Muell.-Arg. in Mart. Fl. Bras. 61: 174. 1860. KEY TO THE SPECIES a. Corolla white, the throat conical, dilating continuously toward the orifice; leaves coriaceous or firmly chartaceous; plants of the Antilles, the Bahama Islands, southern Florida, and Central America, also pe in lowlands of northern South Amerieg. eee E. biflora aa. Corolla pink to rose-colored, the throat tubular, not dilating as the orifice or only slightly so; leaves membranaceous; species of South America. b. Plants scatteringly pilose; calyx-lobes 0.2-0,4 em. long, much shorter than the proper-tube.of bereet 2. R. macrostoma bb. Plants glabrous; calyx-lobes 0.8-1.25 cm. long, about equalling the Ge of the: proper-tube of tho corolla -eaea ovre rose des ei aree ula R. Pohlü 1. Rhabdadenia biflora (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 61: 175. 1860; Miers, Apoc. So. Am. 121. 1878; K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 170. 1895. Echites biflora Jacq. Enum. Pl. Carib. 13. 1760; L. Sp. Pl. ed. 2. 307. 1762; Jaeq. Select. Stirp. Am. Hist. 1: 30; 2: pl. 21. 1763; A. DC. in DC. Prodr. 8: 450. 1844. Echites paludosa Vahl, Eclog. 2: 19. 1798; A. DC. loc. cit. 467. 1844. Exothostemon paludosum (Vahl) G. Don, Gen. Hist. 4: 83. 1838; Miers, loc. cit. 241. 1878. Echites Ehrenbergu Schlecht. Linnaea 26: 666. 1853. Echites Billbergu Beurl. Vet. Akad. Handl. Stockh. 137. 1854 (1856). Rhabdadenia be (Schlecht.) Muell.-Arg. Lin- naea 30: 454, Rhabdadenia SE (Vahl) Miers, Apoc. So. Am. 119. 1878. Rhabdadenia nervosa Miers, loc. cit. 122. 1878. Rhabdadema cordata Miers, loc. cit. 1878. Rhabdadenia macrantha Donn. Sm. Bot. Gaz. 40: 7. 1905. Glabrous, suffruticose lianas; stems terete, relatively stout; leaves opposite, petiolate, the blade coriaceous or firmly char- taceous, broadly obovate-oblong to lanceolate, apex usually (378) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 207 rather abruptly mucronulate, base obtuse, either surface yel- lowish-green, the upper somewhat nitidulous, 5-12 cm. long, 1.5-5.0 em. broad, the petiole 1-2 em. long; inflorescence lateral or rarely subterminal, cymose, the peduncle equalling or some- what surpassing the length of the subtending leaves, bearing 1-5 white flowers clustered at the apex; pedicels 1.0-1.25 em. long, somewhat accrescent in fruit, the subtending bracts scari- ous, minute; ealyx-lobes subfoliaceous, broadly ovate-oblong, mueronulate, 0.1-0.9 em. long; corolla infundibuliform, the proper-tube narrowly cylindrical, 1.5-2.0 cm. long, about 0.2 em. in diameter at the base, somewhat constricted toward the insertion of the stamens, the throat conical, 2.0-3.0 em. long, about 1.5 em. in diameter at the orifice, the lobes broadly obo- vate, 2.0-2.5 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers elliptic-oblong, densely pilose dorsally, 0.5 em. long ; ovary ovoid, 0.2 em. long, rather gradually produced into the style, glabrous; nectaries depressed-quadrate, essentially separate, about half equalling the length of the ovary; stigma 0.2 em. long; mature follicles unknown. UNITED STATES: FLORIDA: exact locality lacking, 1842-49, Rugel 114 (MBG, BM); in sandy soil at edge of bay near Brickell Hammock, Miami, Dade Co., March 4, 1930, Moldenke 725 (MBG, NY); salt marshes, Fort lege July—Aug., 1900, Hitchcock 202 (MBG); Miami, June, 1877, Garber s.n. Cusa: exact locality lacking, 1860-64, Wright 2954 (MBG); SANTA CLARA: Cien- a Rio Damuji, May 23, 1895, Combs 78 (MBG); data incomplete, Sagra n. (V). JAMAIOA: Port Antonio, Dee. 31, 1890, Hitchcock s.n. (MBG); Port Morant, Dee. 20, 1890, Hitchcock s.n. (MBG); GRAND CAYMAN: Jan., 1891, Hitchcock s.n. (MBG). HISPANIOLA: HAITI: Grande Caymite, Aug. 20, 1927, Eyerdam 321 (MBG); SANTO DOMINGO: Prov. Barahona, July, 1910, Fuertes 220 (MBG Porto Rico: prope Bayamon, in paludosis, March 22, 1885, Sintenis 935 (S) Playa de Carmelita, in fruticetis paludosis prope marem, April, 1883, Eggers 836 (C) CO: CAMPECHE: over shrubs in Pantel Aguada, E p 7-15, Tm d 1937 (MBG); YUCATAN: Port Silam, 1895, Gaum 3 (M lagoon s Las Bocas de Silam, May, 1916, PRA 23335 (MBG BRITISH HONDURAS: Stann Creek, common climber in swampy places pe. the coast, pes 15, 1929, Schipp 148 (MBG) ; mangrove swamp, Belize, March 11, 1933, Lundell 1813 (MBG); Pueblo Nuevo, New River, July 17, 1933, am 4792 (379) [Vor. 23 208 ANNALS OF THE MISSOURI BOTANICAL GARDEN PANAMA: COLON: Porto Bello, April, 1826, Billberg 254 (S); CANAL ZONE: Chagres, Jan.-March, 1850, Fendler 257 (MBG). COLOMBIA: MAGDALENA: Santa Marta, 1898-1901, Smith 1664 (MBG); BOLIVAR: region of Barranquilla, April, 1935, Elias 1291 (MBG). BRITISH GuIANA: Kabakaburi, Pomeroon Distriet, Febr. 10-15, 1923, Cruz 3274 (MBG); upper Rupununi River, near Dadanawa, May 30, 1922, Cruz 1417 (MBG). FRENCH GUIANA: Cayenne, 1859, Sagot 387 (V). DurcH GUIANA: Paramaribo, date lacking, Wullschlágel 324 (V). BRAZIL: PARA: Caripi juxta Para, Aug., 1849, Spruce s.n. (Camb., V); in maritimis inundatis ad Colares, May, 1832, Poeppig 2946 (V); MARANHÃO: on lowland, border of river, Maracassume River region, Sept. 9, 1932, Krukoff 1886 (MBG, NY); data incomplete, Gardner 6060 (Camb., V). 2. Rhabdadenia macrostoma (Benth.) Muell.-Arg. Linnaea 30: 435. 1860; Miers, Apoc. So. Am. 123. 1878; K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 170. 1895. Echites macrostoma Benth. in Hook. Jour. Bot. 3: 248. 1841; A. DC. in DC. Prodr. 8: 453. 1844. Suffruticose or suffrutescent lianas; stems terete, scatter- ingly pilose when young, eventually becoming glabrate; leaves opposite, petiolate, the blade membranaceous, obovate-oblong to broadly oblong-elliptie, apex rather abruptly mueronate, base obtuse to rounded, 3-6 cm. long, 1.5-3.5 em. broad, upper surface minutely bullate-puberulent to glabrate, lower surface sparsely pilose along the midrib and veins, the petiole 0.3-0.5 cm. long, finely pilose to glabrate; cymes lateral, the peduncle about twice surpassing the length of the subtending leaves, bearing two pink or rose-colored flowers at the apex; pedicels 0.3-0.5 em. long, somewhat accrescent in fruit, the subtend- ing braets scarious, lanceolate, minute; calyx-lobes scarious, linear-lanceolate, 0.2-0.4 em. long, minutely and sparsely pilose to glabrate; corolla infundibuliform, the proper-tube narrowly cylindrical, 0.75-1.0 em. long, about 0.1 em. in diameter at the base, the throat broadly tubular, 3.0-3.5 em. long, about 1 em. in diameter at the orifice, the lobes broadly obovate, 1.5 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers narrowly oblong-lanceolate, 0.4 em. long, pilose above; ovary oblongoid, 0.1 em. long, rather gradu- ally produced into the style, glabrous; stigma 0.2 cm. long; nectaries oblongoid, essentially separate, somewhat shorter (380) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 209 than the ovary; follicles terete, relatively slender, glabrous, 9- 12 em. long, seeds unknown. BRITISH GUIANA: data incomplete, Schomburgk 329 (Camb., V). DutcH GUIANA: locality lacking, 1846, Hostmann 494, 712 (8); Paramaribo, date lacking, Wullschlágel 326 (V). : PARA: on varzea land, near Bocca do Paru, Aug. 28, 29, 1934, Krukoff 5931 (NY); Parana do Ricardo, Aug. 28, 29, Krukoff 5915 (NY). PERU: LORETO: swampy thickets, Iquitos, alt. 100 m., Aug. 2-8, 1929, Killip $ Smith 26911 (MBG). 3. Rhabdadenia Pohlii Muell.-Arg. in Mart. Fl. Bras. 6': 174. pl. 52. 1860; Miers, Apoc. So. Am. 119. 1878: K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4*: 170. 1895. Rhabdadenia Pohlii Muell.-Arg. var. volubilis Muell.-Arg. loc. cit. 1860. Rhabdadenia Pohlii Muell.-Arg. var. suberecta Muell.-Arg. loc. cit. 175. 1860. Rhabdadenia Pohlii Muell.-Arg. var. latifolia Muell.-Arg. loc. cit. 1860. Rhabdadenia latifolia (Muell.-Arg.) Malme, Arkiv f. Bot. JIA" 1/. 192% Rhabdadenia latifolia (Muell.-Arg.) Malme var. suberecta (Muell.-Arg.) Malme, loc. cit. 18. 1927. Rhabdadenia mamorensis Rusby, Mem. N. Y. Bot. Gard. 7: 326. 1927. Glabrous, suffruticose or suffrutescent lianas, occasionally erect or suberect; stems relatively slender; leaves petiolate to subsessile, the blade membranaceous, oblong-elliptic to linear, apex rather gradually acuminate, mucronulate, base narrow- ing from about the middle, obscurely auriculate, 5-12 cm. long, 0.75-3.0 em. broad, either surface glabrous, the lower glau- cescent, the petiole 0.25-1.0 em. long, or essentially obsolete; cymes lateral to subterminal, the peduncle usually about twice surpassing the length of the subtending leaves, bearing 1-3 pink or rose-colored flowers at the apex; pedicels 0.5-1.0 cm. long, the subtending bracts filiform, minute; calyx-lobes nar- rowly oblong-lanceolate, minutely mucronulate, 0.8-1.25 em. long; corolla infundibuliform, the proper-tube narrowly cylin- drical, 0.75-1.0 em. long, about 0.1 cm. in diameter at the base, (381) [Vor. 23 210 ANNALS OF THE MISSOURI BOTANICAL GARDEN the throat rather broadly tubular, 2.5-3.5 em. long, about 1 em. in diameter at the orifice, the lobes broadly obovate, 1.5-2.0 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers narrowly lanceolate, 1.0-1.25 em. long, sparsely pilosulose at the tip; ovary ovoid, 0.1 em. long, rather gradually produced into the style, glabrous; stigma 0.4 cm. long; nectaries ovoid-oblongoid, separate, about equalling the length of the ovary; follicles unknown. OLOMBIA: BOLIVAR: river marsh, Magangue, alt. 40-45 m., Jan. 18-19, 1918, Pennell 3956 (MBG). VENEZUELA: Sacupana, lower Orinoco, April, 1896, Rusby $ Squires 20 (MBG). BRAZIL: PARANA: Piraquara, in paludosis, Jan. 7, 1909, Dusén 7781 (MBG); exact locality lacking, Nov., 1916, Brito 43 (US); RIO GRANDE DO SUL: Neu Würt- temberg, Estancia Coromel, Belisiano bei Lagão, Febr. 27, 1906, Bornmüller 693 (M); same locality, Jan. 16, 1905, Bornmüller 427 (M). PARAGUAY: swamps, exact locality lacking, Dec. 8, 1928, Jørgensen 3450 (MBG); in ni ds. Frape, Jan. 5, 1929, Jørgensen 4038 (MBG ENTINA: MISSIONES: San Ignacio, Jan., 1918, SE 3591 (MBG). This species varies greatly in its habit. Although always more or less twining in the northern portion of its range, it in- clines strongly to the erect posture of such species as Mande- villa erecta (Vell.) Woodson. Perhaps such close relatives of normally twining congeners merit varietal or formal recogni- tion, but our present evidence does not appear to warrant it. EXCLUDED OR UNCERTAIN SPECIES Rhabdadenia barbata (Desv.) Miers, Apoc. So. Am. 123. 1878 (Echites barbata Desv. ex Ham. Prodr. 30. 1825) = Urechites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907 (Vinca lutea L. Cent. IT. Pl. 12. 1756). Rhabdadenia Berterii (A. DC.) Muell.-Arg. Linnaea 30: 446. 1860 (Echites Berterii A. DC. in DC. Prodr. 8: 447. 1844) = Angadenia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878. Rhabdadenia campestris (Vell.) Miers, Apoc. So. Am. 121. 1878 (Echites campestris Vell. Fl. Flum. 113. 1830; Icon. 3: pl. 43. 1827) = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792). (382) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 211 Rhabdadenia corallicola Small, Bull. N. Y. Bot. Gard. 3: 434. 1905 = Angadenia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878 (Echites Berterii A. DC. in DC. Prodr. 8: 447. 1844). Rhabdadenia cubensis Muell.-Arg. Linnaea 30: 435. 1860 = Angadenia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878 (Echites Berterii A. DC. in DC. Prodr. 8: 447. 1844). Rhabdadema laxiflora Miers, Apoc. So. Am. 120. 1878 = Urechites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907 (Vinca lutea L. Cent. II. Pl. 12. 1756). Rhabdadenia Lindeniana Muell.-Arg. Linnaea 30: 437. 1860 = Angadenia Lindeniana (Muell.-Arg.) Miers, Apoc. So. Am. 180. 1878. Rhabdadenia madida (Vell.) Miers, Apoc. So. Am. 121. 1878 (Echites madida Vell. Fl. Flum. 112. 1830; Icon. 3: pl. 42. 1827). Probably referable to a species of Mandevilla, such as M. scabra (R. & S.) K. Sch., in which the inflorescence is fre- quently reduced to few flowers. Rhabdadenia polyneura Urb. Symb. Ant. 7: 337. 1912 = Odontadenia polyneura (Urb.) Woodson, Ann. Mo. Bot. Gard. 18: 546. 1931. . Rhabdadenia Sagraei (A. DC.) Muell.-Arg. Linnaea 30: 435. 1860 (Echites Sagraei A. DC. in DC. Prodr. 8: 450. 1844) = Angadenia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878 (Echites Berterii A. DC. in DC. Prodr. 8: 447. 1844). Rhabdadenia W rightiana Muell.-Arg. Linnaea 30: 438. 1860 = Neobracea Valenzuelana (A. Rich.) Urb. Symb. Ant. 9: 241. 1924 (Echites Valenzuelana A. Rich. in Sagra, Hist. Cuba 11: 93. 1850). XVIII. Ervrnorvus Muell.-Arg. Elytropus Muell.-Arg. Bot. Zeit. 18: 21. 1860; Benth. & Hook. Gen. Pl. 2: 728. 1876; Miers, Apoc. So. Am. 114. 1878; K. Sch. in Engl. € Prantl, Nat. Pflanzenfam. 47: 169. 1895. Lactescent, suffruticose lianas. Stems terete, volubile, said to be suberect in some instances; branches alternate above, chiefly opposite below. Leaves opposite, shortly petiolate, pen- (383) [VoL. 23 212 ANNALS OF THE MISSOURI BOTANICAL GARDEN ninerved, eglandular, the petioles subtended by 1-few incon- spieuous, dentiform, adaxial stipular appendages. Inflores- cence lateral, opposite, or infrequently alternate, determinate, 1-3-flowered, conspicuously bracteate. Calyx 5-parted, the lobes equal to sub-equal, cleft nearly to the receptacle, imbri- cated, eglandular. Corolla campanulate, the tube broad, ex- appendieulate within, the limb actinomorphic, 5-parted, dex- trorsely eonvolute. Stamens 5, the anthers connivent and ag- glutinated to the stigma, consisting of 2 parallel sporangia borne ventrally near the apex of an enlarged, narrowly sagit- tate connective; filaments separate; pollen granular. Carpels 2, united at the apex by a common stylar shaft surmounted by the capitate-fusiform stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, separate or essentially so. Follicles 2, apocarpous, terete, acuminate, de- hiscing along the ventral suture, containing many dry, trun- cate, apically comose seeds. Type species: Elytropus chilensis (A. DC.) Muell.-Arg. Lin- naea 30: 440. 1860. 1. Elytropus chilensis (A. DC.) Muell.-Arg. Linnaea 30: 440. 1860. Miers, Apoc. So. Am. 115. 1878. Echites pubescens Hook. & Arn. Bot. Beechey Voy. 34. 1841, not Willd. Echites Chilensis A. DC. in DC. Prodr. 8: 468. 1844. Echites ptarmica Poepp. Nov. Gen. 3: 69. pl. 278. 1845. Echites heterophylla Miq. Linnaea 25: 653. 1852, not Gmel. Elytropus pubescens (Hook. & Arn.) Miers, loc. cit. 114. 1878. Elytropus ptarmicus (Poepp.) Miers, loc. cit. 115. 1878. Vinca sternutatoria Poepp. ex Miers, loc. cit. 1878, nom. nud. in synon. Elytropus heterophyllus (Miq.) Miers, loc. cit. 116. 1878. Stems usually relatively stout, densely ferruginous-puberu- lent to glabrate; leaves opposite, shortly petiolate, ovate to oblong-elliptie, apex rather shortly acuminate, base obtuse to (384) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 213 rounded, 3.5-9.0 cm. long, 1.2-4.5 em. broad, subcoriaceous, above glabrous, nitidulous, beneath opaque, persistently fer- ruginous-puberulent; petioles 0.5-1.0 cm long, ferruginous- puberulent ; inflorescence lateral, opposite, or infrequently al- ternate by the abortion of an opposite bud, much surpassed by the subtending leaves, 1-3-flowered ; peduncle 0.3-0.7 cm. long, minutely puberulent to glabrate; bracts oblong-elliptic, 0.2-0.5 em. long, somewhat foliaceous; pedicels 0.1-0.2 em. long, mi- nutely puberulent; calyx-lobes oblong-elliptic, acute, 0.55—0.8 em. long, somewhat foliaceous, puberulent to pilosulose; co- rolla campanulate, glabrous without, the tube 0.5—0.9 em. long, about 0.15-0.2 cm. in diameter at the base, 0.4—0.55 em. in di- ameter at the orifice, the lobes obliquely ovate, shortly apicu- late, 0.5—1.3 em. long, spreading ; stamens inserted near the base of the corolla-tube, the anthers barely included, oblong-sagit- tate, 0.5 cm. long, tips pilose; ovary ovoid, 0.07 cm. long, glabrous; stigma 0.15 em. long ; nectaries essentially separate, truncate, about equalling the ovary; follielés relatively stout, rather obscurely articulated, 15-20 cm. long, rather densely ferruginous-hirtellous; seeds 0.8-1.1 em. long, the pale yel- lowish coma 3.0-3.5 em. long. CHILE: CHILOE: am Büschen schlingend, Piruquina, Oct. 22, 1931, Gunckel 65 (MBG); data incomplete, Nov., 1925, Joseph 3334 (US); LLANQUIHUE: Casa Panque, Dec., 1926, Shannon $ Shannon 28 (US); MALLECO: Cura Cautin, Nov., 1925, Joseph 3397 (US); La Union, Oct. 25, 1931, Behn 1179 (MBG); VALDIVIA: San Carlos, Oct. 5, 1931, Gunckel 2361 (MBG); Loncoche, Sept., 1926, Joseph 4408 (US); Panguipulli, Oct., 1923, Joseph 2397 (US); Panguipulli, alt. 200 m., Oct., 1924, Hollermayer 326 (MBG, NY, US); Ufergebüseh des Calle-Calle, Oct. 20, 1897, Buchtien s.n. (US); DATA INCOMPLETE: Gay 384 (NY). ARGENTINA: RIO NEGRO: Playa Bonita, region of Lago Nahuel Huapi, Nov. 1, 1928, Cordoni s.n. (MBG, US). The relatively small cream-colored flowers are said to have a fragrance similar to that of Jasmine. Cordoni reports the flowers as violet or reddish. EXCLUDED SPECIES Elytropus spectabilis (Stadelm.) Miers, Apoc. So. Am. 116. 1878 (Echites spectabilis Stadelm. Flora 24!: Beibl. 44. 1841) (385) [Vor. 23 214 ANNALS OF THE MISSOURI BOTANICAL GARDEN - Macropharynx spectabilis (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931. XIX. CvcraApEN1A Benth. Cycladenia Benth. Pl. Hartw. 322. 1849; Benth. & Hook. Gen. Pl. 2: 728. 1876; A. Gray, Syn. Fl. N. Am. ed. 1. 2': 83. 1878; Miers, A poc. So. Am. 263. 1878; K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 168. 1895; Jepson, Man. Fl. Pl. Calif. 769. 1925. Low, subsucculent, perennial herbs. Stems erect or more or less diffuse, terete; branches alternate or rather indefinitely clustered from the base. Leaves opposite, petiolate, the blade rather thick and subsucculent, entire, penninerved, eglandu- lar; petioles winged and concrescent at the nodes, exappen- diculate. Inflorescence rather irregularly dichasial, lateral or pseudoterminal, the pedicels subtended by solitary bracts. Flowers showy, reddish-violet or nearly cream-colored. Calyx 5-parted, the lobes subequal, imbricated, cleft nearly to the receptacle, eglandular. Corolla infundibuliform, the tube straight, greatly inflated at the insertion of the stamens into a conical throat, the limb 5-parted, actinomorphic, dextrorsely convolute. Stamens 5, inserted at the base of the corolla- throat, wholly included; anthers connivent and agglutinated to the stigma, consisting of 2 parallel, basally protuberant spo- rangia borne ventrally near the apex of an enlarged, sagittate, narrowly 2-auriculate connective; pollen granular; filament short, ligular, minutely pilose. Carpels 2, united at the apex by an elongate, stylar shaft surmounted by the subcapitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectary annular, completely concrescent. Follicles apocarpous or occasionally united at the tips, terete, relatively stout, falcate, dehiscing along the ventral suture, con- taining many dry, compressed, apically comose seeds; embryo straight, the cotyledons ovate, concave. Type species: Cycladenia humilis Benth. Pl. Hartw. 322. 1849. (386) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 219 1. Cycladenia humilis Benth. Pl. Hartw. 322. 1849; A. Gray, Syn. Fl. N. Am. ed. 1. 2!: 83. 1878; K. Sch. in Engl. € Prantl, Nat. Pflanzenfam. 4?: 168. 1895; Jepson, Man. Fl. Pl. Calif. 769. 1925. Low, subsucculent, perennial herbs from a stout tap-root ; stems 1-2 dm. tall; leaves opposite, petiolate, the blade rather thick, ovate to suborbicular, apex obtuse to rounded, base ob- tuse to obscurely cordate, rather gradually produced into the petiole, 3-7 cm. long, 2-6 cm. broad, greatly reduced below, the petiole 0.5-3.0 em. long; cymes few- to several-flowered, about equalling or somewhat surpassing the subtending leaves; pedicels 0.75-1.25 cm. long, the subtending bracts narrowly lanceolate, 0.2-0.5 cm. long; calyx-lobes ovate-lanceolate to linear, acuminate, 0.5-0.75 cm. long; corolla showy, reddish- violet to cream-colored, the proper-tube 0.25-0.5 em. long, about 0.2 cm. in diameter at the base, the throat conical, 0.5—0.75 em. long, about 0.4—0.6 cm. in diameter at the orifice, the lobes obovate-oblong, 0.5-0.75 em. long, spreading; anthers 0.35 cm. long, minutely papillate dorsally ; ovary ovoid, rather abruptly produced into the style, about 0.15 cm. long; nectary annular, about half as high as the ovary; stigma subcapitate, 0.1 cm. long; follicles relatively stout, essentially continuous, falcate or slightly divaricate, 3.5-7.0 cm. long, about 0.3-0.5 cm. in diameter ; seeds 0.75 cm. long, the pale tawny coma 1.5-2.0 em. long. Var. typica. Plants glabrous and glaucous throughout. UNITED STATES: CALIFORNIA: edge of lavas, Medicine Peak, Siskiyou Co., alt. 6000 ft., Aug. 5, 1893, Baker s.n. (UC); Sisson, Siskiyou Co., lower Canadian zone, July 16, 1902, Setchell $ Dobie s.n. (UC); Snow Mt., Lake Co., alt. 7800 ft., Aug., 1894, Purpus 851 (UC); Cobb Mt., Lake Co., July, 1893, Leithold s.n. (UC, LS); high slopes, Greenville and Lassen Lake, Plumas Co., July, 1872, Lemmon 119 (UC); Lassen’s Peak, Lassen Co., Aug., 1896, Austin s.n. (MBG) ; Sierra Co., 1875, Lemmon 187 (MBG) ; peaks of Plumas Co., June, year lacking, Lemmon s.n. (MBG, UC); Greenville, Plumas Co., June 4, 1920, Clemens s.n. (CA); Mt. Lassen, Lassen Co., July 10, 1923, Bassett s.n. (CA) ; Lassen Voleanie Park, Lassen Co., June, 1927, Sutliffe s.n. (CA); southeast side of Snow Mt. above Bonnie View, Lake Co., June 7, 1919, Heller 13229 (CA, LS, MBG); Cedar Glen, Sierra Co., May 25, 1920, V. Jones s.n. (CA); Prospect Peak, 1929, Kramer s.n. (CA); Cobb Mt., near top, (387) [Vor. 23 216 ANNALS OF THE MISSOURI BOTANICAL GARDEN alt. 4000 ft., May 31, 1927, Baker 2180A (LS); Lassen's Peak, Lassen Co., alt. 6000 ft., July 8, 1897, M. E. Jones s.n. (LS, PC); mesa near Black Butte, Siskiyou Co., Aug. 9, 1899, Dudley s.n. (LS); Silver Lake, Lassen Co., July 30, 1894, Baker s.n. (LS, UC); top of ridge west of Camp, Sta. Lucia Mts., Aug. 11, 1903, Dudley s.n. (LS); Indian Valley, June 2, 1873, Lemmon s.n. (LS); Mt. Balley, Shasta Co., Sept. 26-28, 1862, pei 1448 (UC); Dutch Hill, Butte Co., 1879, Austin 370 (UC); Doe Mill, Butte Co., May, 1898, Austin $ Bruce 2422 (PC); data incom- plete, Hartweg s.n. (Cane Var. venusta (Eastw.) Woodson, in Munz, Man. So. Calif. Bot. 379. Cycladenia venusta Eastw. Bull. Torrey Bot. Club 29: 77. 1902. Calyx and eorolla-tube conspieuously pilose; in all other essential characters similar to ae typical variety. UNITED STATES: CALIFORNIA: Summit, San Antonio Mt., Los Angeles Co., June, 1899, Hall s.n. (LS); same locality, E p m., June, 1897, Hasse s.n. (LS) ; Mt. San Antonio, San Bernardino Co., July, 1901, ys 1920 (LS, PC); south slope of Baldy, San EE Mts., dry SE under pines, alt. 7500 ft., July 4, 1917, John- ston 1428 (LS ); Cucamonga Peak, San Antonio M sunny exposure, well-paeked Eu s. soil, alt. 8200 ft., July 31, 1917, Johnston 1550 (LS, UC, PC); Devil's Baekbone, San Antonio Mts., dry rocky ground, alt. 9000 ft., July 4, 1917, Johnston 1431 (LS, UC, PC); dry soil on Mt. San Antonio, alt. 8000 ft., July 16, 18 rnardino July 8, 1926, M. E. Jones s.n. (LS); Santa Lucia Mts., date lacking, Abbott s.n. (CA); Santa Lucia Peak, Oct. 3, 1921, Clemens s.n. (CA); Santa Lucia Mts., June 9, 1893, Eastwood s.n. (UC); southern slope of Mt. San Antonio, San Gabriel Mts., Los Angeles Co., on open talus slopes, alt. 10,000 ft., July E 1930, Goodman d Hitchcock 1720 (MBG) ; gravelly ridges near summit, San Antonio Mt., alt. 9500 ft., June, 1899, Hall 1231 (UC) ; San Antonio Mt., alt. 8000 ft., Aug. 20, 1905, Wilder 591 (UC); ridge east of Ontario Peak, San Gabriel Mts., occasional on dry ridge, alt. 8300 ft., July 18, 1922, Munz 6094 (PC); UTAH: San Rafael Swell, May 19, 1914, M. E. Jones s.n. (MBG, LS Var. tomentosa A. Gray, Syn. Fl. N. Am. ed. 2. 21: 400. 1886. Cycladenia tomentosa A. Gray, in Bot. Calif. Geol. Surv. 1: 474. 1876 Cycladenia humilis Benth. var. tomentosa Jepson, Man. Fl. Pl. Calif. 769. 1925, in err. Plants densely tomentose throughout; in all other essential characters similar to the typical variety with which it is oc- casionally found. UNITED STATES: CALIFORNIA: Cedar Glen, Sierra Co., May 25, 1920, V. Jones s.n. (CA); Mt. Dyer, 1883, Austin s.n. (CA); Mt. Shasta, June, 1887, Brandegee s.n. (388) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 217 (MBG); Black Cone, near Shasta, date lacking, Lemmon s.n. (MBG, G, TYPE, UC, LS); Dutch Hill, Butte Co., 1879, Austin 376 (UC); between Sisson’s and Edge- wood, Siskiyou Co., July, 1887, Brandegee s.n. (UC); trail between Indian Valley and Mountain Meadows, July, 1872, Lemmon s.n. (UC); mesa, near Black Butte, Siskiyou Co., Aug. 9, 1899, Dudley s.n. (LS). Although its relatively isolated geographical distribution might appear to argue its specificity, C. venusta Eastw. has been reduced to varietal rank since it has no distinctive char- acteristic except its relatively restricted indument. XX. Ecnrtres P. Br. Echites P. Br. Hist. Jam. 2: 182. 1756; Jacq. Enum. PI. Carib. 13. 1760, in part; Select. Stirp. Am. Hist. 1: 30. 1763; L. Sp. Pl. ed. 2. 307. 1762; Gen. Pl. ed. 6. 117. 1764; A. DC. in DC. Prodr. 8: 446. 1844; Miers, Apoc. So. Am. 191. 1878; K. Sch. in Engl. € Prantl, Nat. Pflanzenfam. 4*: 165. 1895, all in part, as to E. umbellata Jacq. Lactescent, suffruticose or suffrutescent lianas. Stems vol- ubile, terete; branches usually opposite below, becoming alternate above. Leaves opposite, petiolate, entire, penni- nerved, eglandular; petioles somewhat girdling at the node into an inconspicuous, minutely appendiculate, stipular ring. Inflorescence lateral, or rarely subterminal or terminal, alter- nate, amore or less modified dichasium, bracteate, bearing soli- tary to numerous flowers. Calyx 5-parted, the lobes equal or subequal, cleft nearly to the receptacle, bearing within at the base a solitary, frequently deeply dissected, opposite squa- mella. Corolla salverform, the tube straight, exappendiculate, not annulate, the limb actinomorphic, 5-parted, dextrorsely eonvolute. Stamens 5, the anthers wholly included, connivent and agglutinated to the stigma, consisting of 2 parallel spor- angia with a conspicuous, protuberant base borne ventrally near the apex of an enlarged, sagittate, narrowly 2-auriculate connective; pollen granular. Carpels 2, apocarpous, united at the apex by an elongate, stylar shaft surmounted by the fusi- form-subeapitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, separate or more (389) [Vor. 23 218 ANNALS OF THE MISSOURI BOTANICAL GARDEN or less concrescent. Follicles 2, apocarpous, terete, continuous or only slightly articulated, dehiscing along the ventral suture, containing many dry, rostrate, apically comose seeds. Type species: Echites umbellata Jacq. Enum. Pl. Carib. 13. 176 12 KEY TO THE SUBGENERA AND SECTIONS A. Corolla 1.25-2.5 em. long, the lobes narrowly oblong- to elliptic-lanceolate, reflexed at anthesis; inflorescence relatively lax and many-flower AA. Corolla 5-8 em. long, the lobes obliquely obovate, spreading at anthesis; inflorescence relatively compact, few- to several-flowered EE Subgen. II. EvEcHITES B. Corolla-tube not spirally contorted; species of Central America........ BREE ee Cee ery a eer ee Sect. 1. YUCATANENSES BB. Corolla-tube spirally RAE plants of southern peninsular Florida, the Bahama Islands, the Antilles, and locally of the peninsula of Yueatan and Atlantie eoastal Colombia............ Sect. 2. UMBELLATAE Subgen. I. Pseupecurres Woodson, subgen. nov. Corolla 1.25-2.5 cm. long, the lobes narrowly oblong- to elliptic-lanceolate, long-acuminate, reflexed at anthesis; inflo- rescence relatively lax and several- to many-flowered. Spp. 1-2. KEY TO THE SPECIES a. Corolla 1.25-1.8 em. long, the lobes somewhat shorter than the tube; ovary and noctaries glabrous. viv exse pe sensed: 1. E. tuxtlensis aa. Corolla 2.0-2.5 em. long, the lobes about twice surpassing the tube; ovary puberulent-papillate, nectaries minutely pilosulose-barbate....2. E. turbinata 1. Echites tuxtlensis Standl. Contr. U. S. Nat. Herb. 23: 1164, 1924. Stems relatively slender, glabrous; leaves opposite, shortly petiolate, narrowly lanceolate- to obovate-elliptic, apex acumi- nate, base cuneate, 6-10 cm. long, 1.5-3.5 em. broad, firmly membranaceous to subcoriaceous, glabrous; petioles 0.3-0.8 em. long; inflorescence lateral, alternate, a lax, dichotomous, bostrychoid dichasium bearing several to many small, yel- lowish-green flowers, much surpassing the subtending leaves, wholly glabrous; pedicels 0.7-1.0 em. long; bracts minutely trigonal, about 0.1 cm. long, or somewhat less; calyx-lobes ovate-trigonal, acute to acuminate, 0.1—0.15 em. long, glabrous, the squamellae deltoid, minutely erose; corolla salverform, (390) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 219 glabrous without, the tube 0.8-1.0 em. long, about 0.1 cm. in diameter at the base, slightly dilating somewhat above the base at the insertion of the stamens, thence slightly constricting toward the orifice, the lobes oblong- to elliptic-lanceolate, acu- minate, somewhat shorter than the tube, more or less reflexed at anthesis; stamens inserted near the base of the corolla-tube, the anthers narrowly elliptic-sagittate, 0.55 em. long, glabrous; ovary oblong-ovoid, about 0.1 cm. long, glabrous; stigma 0.1 em. long; nectaries separate, about half equalling the ovary; follicles known. MEXICO: CHIAPAS: near Tuxtla, alt. 2400-2800 ft., Sept. 1, 1895, Nelson 3080 (US, TYPE, MBG, photograph and analytical drawin BRITISH HONDURAS: high ridge, Corozal District, 1981-88, Gentle 439 (MBG) ; Maskall, Northern River, Dee., 1933, Gentle 1022 (MBG). Costa Rica: Heredia, in Monte Barba, date lacking, Ørsted 15544 (C). 2. Echites turbinata Woodson, Ann. Mo. Bot. Gard. 21: 615. 1934 Stems relatively stout, glabrous; leaves opposite, petiolate, elliptic, apex acuminate, base broadly obtuse, 7.5—12.0 em. long, 3.5-7.0 em. broad, firmly membranaceous, evidently somewhat subsucculent, glabrous; petioles 0.8-1.3 em. long; inflorescence lateral, alternate, rather irregularly dichasial, about equalling the subtending leaves, bearing several rather small, yellowish- green (?) flowers, glabrous throughout; pedicels 0.8-1.0 em. long; bracts ovate-lanceolate, 0.13-0.3 em. long; calyx-lobes ovate, acuminate, 0.3—0.32 cm. long, glabrous, the squamellae deltoid, minutely denticulate; corolla salverform, minutely puberulent-papillate without, the tube 0.7—0.8 cm. long, about 0.15 em. in diameter at the base, slightly constricting toward the orifice, the lobes narrowly elliptic-lanceolate, acuminate, 1.4—1.6 em. long, somewhat turbinate; stamens inserted about midway within the corolla-tube, the anthers rather narrowly trigonal-sagittate, 0.45 em. long, glabrous; ovary ovoid, very minutely puberulent-papillate, about 0.15 cm. long; stigma 0.15 em. long; nectaries separate, somewhat shorter than the ovary, minutely pilosulose-barbate at the tips; follicles unknown. (391) [ Vor. 23 220 ANNALS OF THE MISSOURI BOTANICAL GARDEN CosTA Rica: forets du Rancho Flores, Febr. 22, 1890, Tonduz 2147 (B, TYPE, MBG, photograph and analytical drawings). A drawing of the inflorescence of this species, with remarks on its strueture, will be found in Ann. Mo. Bot. Gard. 22: 14. pl. 2, fig. 2. 1935. Subgen. II. EvEcurres Woodson, subgen. nov. Corolla 5-8 em. long, the lobes obliquely obovate, obtuse to shortly acuminate, spreading at anthesis; inflorescence rela- tively compact, few- to several-flowered. Sects. 1-2. Sect. 1. YUCATANENSES Woodson. Corolla relatively delicate in texture, the buds attenuate, the tube straight, not spirally contorted. Species of Central America. Spp. 3-5. KEY TO THE SPECIES a. Inflorescence almost perfectly dichasial to simply helicoid, not subumbel- AO ADO Dis De AOI D EC 3. E. turrigera aa. Inflorescence subumbellate. b. Leaves coriaceous or subcoriaceous, more or less pandurate, conspicuously nitidulous above with verrueose venation.............. 4. E. yucatanensis bb. Leaves rather delicately membranaceous, evidently never pandurate opaque above, the venation not verrueose............ ues. ó. E. elegantula 3 .Echites turrigera Woodson, Ann. Mo. Bot. Gard. 19: 381. 1932. Echites cincinnalis Woodson, loc. cit. 21: 616. 1934. Stems relatively slender, rather inconspicuously pilosulose when young, eventually becoming glabrate; leaves opposite, petiolate, broadly oblong- to ovate- or obovate-elliptie, apex rather shortly acuminate, base obtuse to rounded, rarely ob- scurely cordate, 5-15 em. long, 3-10 em. broad, membranaceous, wholly glabrous, somewhat nitidulous and the venation some- what verrucose above, rather pale yellowish-green; petioles 0.8-4.0 em. long, glabrous; inflorescence lateral to subterminal, alternate, almost perfectly dichasial to simply helicoid, some- what surpassing the subtending leaves, bearing 6-20 rather showy, white or cream-colored flowers; peduncle minutely pilosulose-scabridulous to glabrate; pedicels 1.0-1.2 cm. long, minutely pilosulose-scabridulous to nearly glabrate; bracts oblong-lanceolate, acuminate, somewhat foliaceous, 0.1—0.3 em. (392) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 221 long; calyx-lobes lanceolate, acuminate, 0.3—0.7 cm. long, rather sparsely pilosulose, the squamellae erose to lacerate; corolla salverform, glabrous without, the tube 2.8-3.8 em. long, about 0.15-0.2 cm. in diameter at the base, rather conspicuously and abruptly dilated somewhat below midway at the insertion of the stamens, thence rather gradually constricting toward the orifice, the lobes obliquely obovate, shortly acuminate, 1.5-2.0 em. long, spreading; anthers rather narrowly oblong-sagittate, 0.75-0.8 em. long, glabrous; ovary oblongoid, about 0.15-0.2 em. long, glabrous; stigma 0.15 em. long; nectaries separate or rather indefinitely conerescent, somewhat less than half equal- ling the ovary ; immature follicles rather sparsely pilosulose. GUATEMALA: ZACAPA: Gualan, alt. 620 ft., June 20, 1909, Deam 6376 (MBG, TYPE, NY); JUTIAPA: Laguna de Guija, alt. 1200 m., April, 1894, Heyde $ Luz 6345 (NY). NICARAGUA: GRANADA: environs de Granada, alt. 40 m., autumn, 1869, Lévy 1074 (C, MBG, photograph and analytieal drawings); low hills near Granada, edge of thicket, July 2, 1923, Maxon Harvey $ Valentine 7614 (US). When E. cincinnalis was originally described, only two of the specimens cited above were known to me, Lévy 1074, the type of E. cincinnalis, and Deam 6376, the type of E. turrigera. The former species was erected upon the basis of a simple, cin- cinnal inflorescence, that of E. turrigera being an almost per- fect, compound dichasium. Geographical distribution was also a fancied support for the distinction of the species. How- ever, with the examination of Heyde € Lux 6345 and Maxon Harvey € Valentine 7614, it appears probable that the inflo- rescence of the collective species is much more variable than is usual, and it has been decided to consolidate the two former species, at least until additional evidence warranting their seg- regation is forthcoming. d Echites yucatanensis Millsp. ex Standl. Field Mus. Publ. Bot. 8: 35. 1930. Stems relatively stout, glabrous; leaves opposite, ovate to oblong, frequently irregularly pandurate, apex acuminate, base obtuse to rounded, 7-12 cm. long, 2.5-7.0 em. broad, coriaceous to subcoriaceous, dark green and nitidulous above, the venation (393) [Vor. 23 222 ANNALS OF THE MISSOURI BOTANICAL GARDEN conspicuously verrucose, opaque beneath, glabrous through- out; petioles 1-2 em. long; inflorescence lateral, alternate, sub- umbellate, bearing 3-9 greenish-yellow (?) flowers; peduncle glabrous, usually somewhat shorter than the subtending leaves; pedicels 1.0-1.3 em. long, glabrous; bracts ovate- lanceolate, minute, scarious ; calyx-lobes ovate-lanceolate, acu- minate, 0.18-0.2 em. long, glabrous, the squamellae deltoid, minutely erose; corolla salverform, glabrous without, the tube 4.0-4.5 em. long, about 0.2 cm. in diameter at the base, somewhat below midway abruptly dilated at the insertion of the stamens, thence gradually constricting toward the orifice, the lobes obliquely obovate, shortly acuminate, 2.5-3.0 cm. long, spread- ing; anthers lanceolate-sagittate, 1.0-1.2 em. long, glabrous; ovary oblongoid, about 0.15 em. long, glabrous; stigma 0.2 em. long; nectaries more or less concrescent at the base, less than half equalling the ovary ; follicles relatively slender, acuminate, continuous, rigidly divaricate, 16-25 em. long, glabrous; seeds 1.0-1.2 em. long, the tawny coma 1.5-2.0 em. long. MEXICO: CAMPECHE: small vine, Tuxpeña, Febr. 18, 1932, Lundell 1350 (MBG, NY, US); YUCATAN: Chichankanab, date lacking, Gawmer 1979 (C, MBG, S, 1s0- TYPES); in clearing, Chichen-Itza, June 22, 1932, Steere 1471 (MBG); Xkantunil, date lacking, Gaumer 817 (S) ; Izamal, date laeking, Gaumer 817 (C). 9. Echites elegantula Woodson, Am. Jour. Bot. 22: 686. 1935. Stems relatively slender, glabrous; leaves opposite, shortly petiolate, ovate-elliptie, apex acutely subcaudate-acuminate, base obtuse and somewhat decurrent, 5-8 em. long, 2.0-3.5 em. broad, rather delicately membranaceous, opaque, glabrous, the veins not verrucose above; petioles 0.5—1.0 em. long ; inflores- cence lateral, alternate, subumbellate, bearing 4—8 showy, greenish cream-colored flowers; peduncle about half equalling the subtending leaves, essentially glabrous; pedicels 1.5-2.0 cm. long, glabrous ; bracts linear, 0.2-0.3 em. long; ealyx-lobes lanceolate, acuminate, 0.3-0.5 em. long, indistinctly papillate without, the squamellae subquadrate, essentially entire; co- rolla salverform, very minutely papillate without, the tube 5.0-5.5 cm. long, about 0.15 cm. in diameter at the base, some- (394) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV Soo what below midway abruptly dilated at the insertion of the stamens, thence gradually constricting toward the orifice, the lobes obliquely obovate, obtuse, 3.0-3.5 em. long, spreading; anthers elliptic-lanceolate, obtusely auriculate, 0.9 em. long, glabrous; ovary oblongoid, about 0.3 cm. long, glabrous ; stigma 0.3 cm. long; nectaries concrescent, somewhat less than half equalling the ovary; follicles unknown. MEXICO: YUCATAN: over tree in forest, Chichen-Itza, June 23, 1932, Steere 1477 (MBG, TYPE). The obtusely auricled anthers are an exception to the general rule of this genus. Sect. 2. UMBELLATAE Woodson. Corolla relatively fleshy in texture, the buds obtuse, the tube spirally contorted. Plants of southern peninsular Florida, the Bahama Islands, the Antilles, and locally of the peninsula of Yucatan and Atlantic coastal Colombia. Sp. 6. 6. Echites umbellata Jacq. Stems relatively stout, glabrous; leaves opposite, shortly petiolate, narrowly oblong-elliptic to suborbicular, apex acumi- nate to rounded and somewhat retuse, base obtuse to rounded, occasionally somewhat cordate, firmly membranaceous or somewhat subsucculent, glabrous; inflorescence lateral to sub- terminal, alternate, irregularly dichasial to subumbellate, glabrous throughout; peduncle somewhat shorter than the sub- tending leaves, or virtually lacking, bearing relatively few greenish-yellow flowers; pedicels 1.0-1.5 em. long, glabrous, greatly accrescent in fruit; bracts ovate to ovate-lanceolate, 0.15-0.3 em. long, only slightly foliaceous; calyx-lobes ovate- to narrowly oblong-trigonal, acute to acuminate, 0.15-0.5 em. long, glabrous, scarious or only slightly foliaceous, the squa- mellae very deeply lacerate ; corolla salverform, glabrous with- out, 2.0—5.5 em. long, about 0.15—0.2 cm. in diameter at the base, somewhat dilated slightly below midway at the insertion of the stamens, thence gradually constricting toward the orifice, spirally contorted above the insertion of the stamens, the lobes (395) [Vor. 23 224 ANNALS OF THE MISSOURI BOTANICAL GARDEN obliquely obovate, obtuse to very inconspicuously apiculate, 1.1-3.0 em. long, spreading; stamens inserted slightly below midway within the corolla-tube, the anthers narrowly lance- olate-sagittate, 0.9-1.2 cm. long, glabrous; ovary oblongoid, about 0.15 cm. long, glabrous; stigma 0.2 em. long; nectaries essentially separate, about equalling to about half equalling the ovary; follicles relatively stout, continuous, acuminate, rigidly divaricate, 9-26 em. long, glabrous; seeds 0.5-0.85 cm. long, the tawny coma 1.5—5.0 em. long. Var. typica. Echites umbellata Jacq. Enum. Pl. Carib. 13. 1760; Select. Stirp. Am. Hist. 1: 30; 2: pl. 22. 1763; A. DC. in DC. Prodr. 8: 447. 1844; Miers, Apoc. So. Am. 193. 1878; K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 165. 1895. Tabernaemontana Echites L. Syst. Pl. ed. 10. 945. 1759. Echites litorea HBK. Nov. Gen. 3: 212. 1819; A. DC. loc. cit. 448. 1844; Miers, loc. cit. 199. 1878. Echites umbellata Jacq. var. longiflora Griseb. Cat. Pl. Cub. 172. 1866. Echites ovata P. Br. ex Miers, loc. cit. 192. 1878, in err. Echites obliqua Miers, loc. cit. 193. 1878. Echites longiflora (Griseb.) Miers, loc. cit. 194. 1878. Echites Echites (L.) Britton, in Small, Fl. Miami, 147. 1913. Leaves narrowly oblong-elliptic to suborbicular, 4-12 cm. long, 2.0-7.6 cm. broad; petioles 0.3-1.5 em. long; inflorescence 2-7-flowered, the peduncle manifest, somewhat shorter than the subtending leaves. UNITED STATES: FLORIDA: sandy field at Cape Sable, July, year lacking, Curtiss 2266 (MBG, NY, US); climbing over bushes, Old Rhodes Key, July 2, 1895, Curtiss 5448 (MBG, NY, US); Miami, May 7, 1904, Tracy 9185 (MBG, NY); pinelands between Miami and Kendall Station, Nov. 5, 1906, Small $ Carter 2657 (NY); Key West, 1874, Palmer 438 (MBG); on dunes, uncommon here, Palm Beach, June 7, 1896, Webber 434 (MBG); pinelands, Big Pine Key, Febr. 27, 1911, Small Carter d Small 3545 (NY); pinelands between Peter’s Prairie and Homestead, Nov. 10, 1906, Small § Carter 2655 (NY); pinelands about Cox Hammock, Dade Co., June (396) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 225 24, 1915, Small Mosier $ Small 6582 (NY); Snapper Creek, south of Coconut no bn 27-31, 1901, Small $ Nash s.n. (NY). A ISLANDS: The Bight and vicinity, Cat, March 1-6, 1907, Britton d gees 5887 (NY); Governor's Harbor, Nov. 14, 1890, Hitchcock 16 (MB Abraham Bay and vicinity, Mariguana, Dec. 6-8, 1907, Wilson 7458 (NY); Eleu- thera Bluff, Eleuthera, Jan. 10, games mg irchild s.n. (US); Water Key, Salt Key Bank, May 22 3 Men on 8148 (MBG, NY); southeast end, Watling's, Nov. 27- 28, 1907, Wilson 7 3 (NY); Grand er Aug. E 1, 1905, Nash $ Taylor 3890 (NY); ane Green Turtle Cay, Abaco, Dec. 4, 1904, Brace 1492 (NY); west E Great Bahama, April 16-May 8, 1905, Brace ns (NY); vicinity of Blue Hills, w Providence, May 28-29, 1909, Wilson Mrd (NY); Mangrove Cay, Andros Se ug. 18-Sept. 10, 1906, Brace 4897 (MBG); South Caicos, Dec. 14-16, 1907, Wilson 7595 (MBG, NY); Gold Rock, Acklin’s, Dec. 21, 1905-Jan. 6, 1906, Brace 4399 (NY, We Eer PINAR DEL : palm-barrens west of Guane, Nov. 21-22, 1911, Shafer 0 (MBG, NY, $5; limestone hills, vieinity of Sumiduro, July 28—31, 1912, miht 13419 (NY, US); ISLA DE PINOS: over bushes on river bank, near Nueva Gerona, July 8, 1900, Palmer $ Riley 1041 (US); HABANA: near Morro Castle, May 20, 1908, Leon 713 (NY); climbing through bushes in swamp near the sea, Playa de Marianao, June 17, 1900, Palmer $ Riley 842 (US); MATANZAS: coral rock soil east of Matanzas, March 13, wë Britton Britton $ Shafer 176 (NY); Playa, Aug. 28, Se Britton $ Wilson 59 (NY); SANTA CLARA: Cienequita, May 9, 1895, Combs 17 (MBG); dry field, is of Saneti Spiritus, Febr. 15-24, 1912, Shafer 12167 A ); CAMAGUEY: Cayo Paloma, Oct. 12, 1909, Shafer 2590 (MBG, NY, US); vicinity of La Gloria, Febr. 4, 1909, Shafer 302 (NY, US); ORIENTE: Cave Hill, limestone SCH south of Ho SR April 11, 1909, Shafer 1236 (NY, US); Punta Maisi, Dec. 14, 1910, Shafer 7930 (NY). JAMAICA: fee over shrubs, Constant Spring to Bardowie, Febr. 8, 1915, Harris 12109 (MBG, NY, US); in sand, coastal region east of Montego Bay, sea- level, March 28, 1920, Maxon $ Killip 1615 (US); Port Royal, Dec. 18, 1890, Hitch- cock s.n. (MBG); Bog Walk, Dec. 17, 1890, Hitchcock s.n. (MBG). HISPANIOLA: SANTO DOMINGO: Paradis, prope Barahona, in fruticetis litoralibus, Dec., 1909, Tuerckheim 2689 (NY, US); southeast of town, San Pedro de Macoris, March 31, 1913, Rose Fitch $ Russell 4292 (NY, US); semi-arid pine region, Mon- pande ei Monte Cristy, alt. 375 m., June 5, 1930, Valeur 444 (MBG, US) ; Lajana, Sam ula, alt. about 100 m., Apr. 30-May 2, 1922, Abbott 2275 (US); HAITI: s om on shrubs, bluff east of Bord de Mer, vicinity of Jean Rabel, Jan. 27—Febr. 9, 1929, Leonard $ Leonard 12883 (NY, US); downs of eliff-bordered shore west of Cabaret, Baie des Moustiques, Jan. 13, 1929, Leonard $ Leonard 11923 (US) ; thickets on coral rocks east of La Tete Linne, vicinity of Basse Tierre, Tortue Island, March 24, 1929, Leonard $ Leonard 14070 (US); a common vine, Miragoane and vicinity, April 12, 1927, Eyerdam 431 (US); hillside, Bayeux, near Port Mar- got, Aug. 4, 1903, Nash 138 (NY, US). MEXICO: YUCATAN: Isla Mujeres, March pti m Goldman 646 (US); in elearing, Uxmal, July ES 21, SE Steere 1987 (M. BriTIsH HONDURAS: coastal region, Honey Aai cm 1929, Lundell 515 (MBG, NY, US). COLOMBIA: San Andres Island, June, 1929, Toro 28 (NY). (397) [Vor. 23 226 ANNALS OF THE MISSOURI BOTANICAL GARDEN Var. crassipes (A. Rich.) Gomez, Anal. Soc. Espan. Hist. Nat. 23: 274. 1894. Echites crassipes A. Rich. in Sagra, Hist. Cuba 11: 91. 1850. Rhodocalyx crassipes (A. Rich.) Miers, Apoc. So. Am. 140. 1878. Leaves narrowly oblong-elliptic to nearly linear, 1-6 cm. long, 0.2-1.5 em. broad; petioles 0.1—0.3 em. long; peduncle ob- solete or scarcely manifest, the solitary pedicels apparently sessile. CUBA: HABANA: Cuabal de Jesus Maria, Minas, June 24, 1915, Leon 5211 (NY); MATANZAS: serpentine hills near Cauasi, Oct. 10, 1927, Leon 13138 (NY); SANTA CLARA: on bushes, Sagua, Sept. 4, 1903, Britton $ Wilson 327 (NY); CAMAGUEY: Santayana, in palm barrens in serpentine, Oct. 4, 1922, Ekman 1533 (S) ; ORIENTE: Holguin, Cerro de Fraile, in fruticetis, Sept. 25, 1916, Ekman 7549 (S); barren savannas, southeast of Holguin, rocky places, Nov. 26-29, 1909, Shafer 2951 (NY). These plants appear to be no more than microphyllous, de- pauperate individuals of var. typica, and it is doubtful whether they should even be recognized as a variety. The reduction of the primary peduncle is perhaps their greatest distinction. EXCLUDED OR UNCERTAIN SPECIES Echites acuminata R. & P. Fl. Peruv. 2: 19. pl. 134. 1799 = Mesechites acuminata (R. & P.) Muell.-Arg. Linnaea 30: 446. 1860. Echites acutiloba A. DC. in DC. Prodr. 8: 451. 1844 = Mande- villa acutiloba (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 54. 1932. Echites adglutinata Jacq. Select. Stirp. Am. Hist. 1: 915 2: pl. 23. 1763, err. typ. = Prestonia agglutinata (J acq.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931 (Echites agglutinata Jacq. Enum. Pl. Carib. 13. 1760). Echites agglutinata Jacq. Enum. Pl. Carib. 13. 1760 = Prestonia agglutinata (Jacq.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931. Echites albiflora Miers, Apoc. So. Am. 204. 1878 = Tabernae- montana sp. The present taxonomic confusion of the genus (398) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 227 Tabernaemontana precludes a precise relegation of this synonym. Echites alexicaca Mart. ex Stadelm. Flora 241: Beibl. 68. 1841 - Mandevilla illustris (Vell.) Woodson var. glabra (Muell.- Arg.) Woodson, Ann. Mo. Bot. Gard. 20: 729. 1933 (Dipla- denia illustris (Vell.) Muell.-Arg. P glabra Muell.-Arg. in Mart. Fl. Bras. 61: 125. pl. 38. 1860). Echites almadensis Stadelm. Flora 24!: Beibl. 28. 1841 = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792, not R. & P.). Echites altescandens H. Winkl. in Fedde, Rep. Spec. Nov. 7: 243. 1909 = Mandevilla antennacea (A. DC.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 47: 171. 1895 (Echites antennacea A. DC. in DC. Prodr. 8: 456. 1844). Echites amazonica Stadelm. Flora 24': Beibl. 50. 1841 = Odontadenia verrucosa (R. € S.) K. Sch. ex Mef. in Pulle, Fl. Surinam 4: 53. 1932 (Echites verrucosa R. & S. Syst. 4: 795. 1819). Echites andina (Muell.-Arg.) Miers, Apoc. So. Am. 204. 1878 (Amblyanthera andina Muell.-Arg. Linnaea 30: 425. 1860) - Mandevilla riparia (HBK.) Woodson, Ann. Mo. Bot. Gard. 19: 58. 1932 (Echites riparia HBK. Nov. Gen. 3: 214. 1819). Echites Andrewsu Chapm. Fl. So. U. S. 359. 1860 = Urechites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907 (Vinca lutea L. Cent. IT. Pl. 12. 1756). Echites Andrieuxit (Muell.-Arg.) Miers, Apoc. So. Am. 206. 1878 (Amblyanthera Andrieuau Muell.-Arg. Linnaea 30: 422. 1860) = Mandevilla Andrieuxii (Muell.-Arg.) Hemsl. Biol. Centr.-Am. Bot. 2: 316. 1882. Echites angustifolia Benth. in Hook. Jour. Bot. 3: 247. 1841, not Poir. - Mandevilla Benthamii (A. DC.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. A7: 171. 1895 (Echites Benthamu A. DC. in DC. Prodr. 8: 467. 1844). Echites angustifolia Poir. Encycl. Suppl. 2: 537. 1812 = Mesechites angustifolia (Poir.) Miers, Apoc. So. Am. 230. 1878. (399) [Vor. 23 228 ANNALS OF THE MISSOURI BOTANICAL GARDEN Echites annularis L. f. Suppl. 166. 1781 = Prestonia annu- laris (L. f.) G. Don, Gen. Hist. 4: 84. 1838. Echites antennacea A. DC. in DC. Prodr. 8: 456. 1844 = Mandevilla antennacea ( A. DC.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895. Echites ( Amblyanthera ?) apocynifolia A. Gray, Proc. Am. Acad. 22: 435. 1887 = Mandevilla apocynifolia (A. Gray) Woodson, Ann. Mo. Bot. Gard. 19: 65. 1932. Echites arborea Vell. Fl. Flum. 114. 1830; Icon. 3: pl. 47. 1827 = Malouetia arborea (Vell.) Miers, Apoc. So. Am. 89. 1878. Echites aspera Mart. € Gal. Bull. Acad. Roy. Brux. 11!: 359. 1844. This may possibly refer to Mandevilla subsagittata (R. & P.) Woodson. The type specimen, Galeotti 1587, was un- available during these studies although special search was made for it in the principle collections of Europe and America. Echites asperuginis Sw. Prodr. 52. 1788 = Anechites lap- pulacea (Lam.) Miers, Apoc. So. Am. 237. 1878 (Echites lap- pulacea Lam. Encycl. 2: 341. 1786). Echites assimilis K. Sch. in Engl. Bot. Jahrb. 25: 724. 1898 - Mandevilla riparia (HBK.) Woodson, Ann. Mo. Bot. Gard. 19: 58. 1932 (Echites riparia HBK. Nov. Gen. 3: 214. 1819). Echites atropurpurea Lindl. in Paxt. Mag. Bot. 9: 199. 1842 - Mandevilla atroviolacea (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 20: 724. 1933 (Echites atroviolacea Stadelm. Flora 24!: Beibl. 75. 1841). Echites atroviolacea Stadelm. Flora 24!: Beibl. 75. 1841 = Mandevilla atroviolacea (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 20: 724. 1933. Echites augusta Vell. Fl. Flum. 114. 1830; Icon. 3: pl. 48. 1827 = Macrosiphonia longiflora (Desf.) Muell.-Arg. in Mart. Fl. Bras. Gr: 140. pl. 43. 1860 (Echites longiflora Desf. Mem. Mus. Paris 5: 177. pl. 20. 1819). Echites auriculata Pohl ex Stadelm. Flora 24': Beibl. 25. 1841 - Mandevilla hirsuta ( A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792). Echites axillaris Besse € Moc. Fl. Mex. 45. 1887. Descrip- tion impossible to interpret. (400) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 229 Echites Bang Rusby, Bull. N. Y. Bot. Gard. 4: 409. 1907 = Prestonia acutifolia (Benth.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 188. 1895 (Haemadictyon acutifolium Benth. ex Muell.-Arg. in Mart. Fl. Bras. 6': 167. 1860). Echites barbata Desv. in Ham. Prodr. 30. 1788 = Urechites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907 (Vinca lutea L. Cent. II. Pl. 12. 1756). Echites barbata Sesse & Moc. Fl. Mex. 45. 1887, non Desv. Deseription impossible to interpret. Echites Benthami A. DC. in DC. Prodr. 8: 467. 1844 = Mandevilla Benthami (A. DC.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895. Echites Berteru A. DC. in DC. Prodr. 8: 447. 1844 = Anga- denia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878. Echites bicolor Miq. Stirp. Surinam. Select. 154. 1851 = Mandevilla scabra (R. & S.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites scabra R. € S. Syst. 4: 195. 1819). Echites bicorniculata Rusby, Descr. So. Am. Pl. 86. 1920 = Mesechites bicorniculata (Rusby) Woodson, Ann. Mo. Bot. Gard. 19: 387. 1932. Echites bicornis Spruce, ex Muell.-Arg. in Mart. Fl. Bras. 6!: 114. 1860, nom. nud. in synon. - Odontadenia verrucosa (R. & S.) K. Sch. ex Met in Pulle, Fl. Surinam 4: 53. 1932 (Echites verrucosa R. & S. Syst. 4: 795. 1819). Echites biflora Jacq. Enum. Pl. Carib. 13. 1760 - Rhab- dadenia biflora (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 61: 175. 1860. Echites bignoniaeflora Schlecht. Linnaea 26: 372. 1853 = Stemmadenia Galeottiana (A. Rich.) Miers, Apoc. So. Am. 76. 1878 (Odontostigma Galeottianum A. Rich. in Sagra, Hist. Cuba 11: 87. pl. 56. 1850). Echites Billbergu Beurl. Vet. Akad. Handl. Stockh. 137. 1854 (1856) = Rhabdadenia biflora (Jacq.) Muell.-Arg. in Mart. Fl. Bras. Di: 175. 1860 (Echites biflora Jacq. Enum. Pl. Carib. 13. 1760). Echites Blanchet A. DC. in DC. Prodr. 8: 448. 1844 = (401) [Vor. 23 230 ANNALS OF THE MISSOURI BOTANICAL GARDEN Prestonia coalita (Vell.) Woodson, Ann. Mo. Bot. Gard. 18: 992. 1931 (Echites coalita Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 40. 1827). Echites bogotensis HBK. Nov. Gen. 3: 215. pl. 243. 1819 = Mandevilla bogotensis (HBK.) Woodson, Ann. Mo. Bot. Gard. 19: 73. 1932. Echites Boliviana Britton, in Rusby, Mem. Torrey Bot. Club 4: 219. 1895 = Mandevilla antennacea (A. DC.) K. Sch. in Engl. € Prantl, Nat. Pflanzenfam. A3: 171. 1895 (Echites an- tennacea A. DC. in DC. Prodr. 8: 456. 1844). Echites brachyloba (Muell.-Arg.) Miers, Apoc. So. Am. 203. 1878 = Mandevilla brachyloba (Muell.-Arg.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Amblyanthera brachyloba Muell.-Arg. Linnaea 30: 423. 1860). Echites Brachysiphon Torr. Bot. Mex. Bound. Surv. 158. 1859 = Macrosiphonia Brachysiphon (Torr.) A. Gray, Syn. Fl. N. Am. 2*: 83. 1878. Echites brachystachya Benth. in Hook. Jour. Bot. 3: 248. 1841 = Mandevilla scabra (R. € S.) K. Sch. in Engl. € Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites scabra R. & S. Syst. 4: 795. 1819). Echites bracteata HBK. Nov. Gen. 3: 217. 1819 = Mande- villa bracteata (HBK.) O. Ktze. Rev. Gen. 2: 414. 1891. Echites bracteata Vell. Fl. Flum. 112. 1830; Icon. 3: pl. 41. 1827, not HBK. or Mart. = Forsteronia Velloziana (A. DC.) Woodson, Ann. Mo. Bot. Gard. 21: 622. 1934 (Echites Vellozi- ana A. DC. in DC. Prodr. 8: 474. 1844). Echites bracteosa Rusby, Mem. N. Y. Bot. Gard. 7: 325. 1927 = Mandevilla bracteosa (Rusby) Woodson, Ann. Mo. Bot. Gard. 20: 742. 1933. Echites breviflora Urb. Symb. Ant. 5: 464. 1908 = Mesechites angustifolia (Poir.) Miers, Apoc. So. Am. 230. 1878 (Echites angustifolia Poir. Eneyel. Suppl. 2: 537. 1812). Echites brevipes Benth. Pl. Hartw. 216. 1845 = Mesechites citrifolia (HBK.) Woodson, Ann. Mo. Bot. Gard. 19: 387. 1932 (Echites citrifolia HBK. Nov. Gen. 3: 216. 1819). Echites Brownet (A. DC.) Muell.-Arg. Linnaea 30: 446. (402) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 231 1860 (Echites torosa Jacq. B Brownei A. DC. in DC. Prodr. 8: 449. 1844) = Mandevilla torosa (Jacq.) Woodson, Ann. Mo. Bot. Gard. 19: 64. 1932 (Echites torosa Jacq. Enum. Pl. Carib. 13. 1760). Echites calycosa A. Rich. in Sagra, Hist. Cuba 11: 94. 1850 = Asketanthera calycosa (A. Rich.) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932. Echites campanulata Sesse E Moc. Fl. Mex. 44. 1887. De- seription impossible to interpret. Echites campestris Vell. Fl. Flum. 113. 1830; Icon. 3: pl. 43. 1827 = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4^: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792). Echites canescens Willd. ex R. & S. Syst. 4: 795. 1819 = Mandevilla scabra (R. & S.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites scabra R. & S. Syst. 4: 795. 1819). Echites Catesbaei G. Don, Gen. Hist. 4: 74. 1838 = Urechites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316..1907 (Vinca lutea L. Cent. II. Pl. 12. 1756). Echites Chilensis A. DC. in DC. Prodr. 8: 468. 1844 = Elytropus chilensis (A. DC.) Muell.-Arg. Linnaea 30: 140. 1860. Echites chlorantha Schlecht. Linnaea 26: 663. 1853 = Mes- echites trifida (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 6*: 151. 1860 (Echites trifida Jacq. Enum. Pl. Carib. 13. 1760). Echites Christophoriana Ham. Prodr. 31. 1825. This may refer to Mesechites repens (Jacq.) Miers, as indicated by the deseription of small oval leaves, dichotomous lateral inflores- cence, and small yellow flowers. Echites ciliata Stadelm. Flora 24!: Beibl. 32. 1841 = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzemfam. 4?: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792). Echites cinerea A. Rich. in Sagra, Hist. Cuba 11: 93. 1850 = Haplophyton cinereum (A. Rich.) Woodson, comb. nov. (H. cimicidum A. DC.). An unnumbered collection by Sagra bear- (403) [Vor. 23 232 ANNALS OF THE MISSOURI BOTANICAL GARDEN ing the notation ''Echites cinerea’’ is apparently the type specimen of this species. It is at present incorporated in the Natural History Museum at Vienna with the types of many others of Richard’s Cuban species; and although the identity of the plant (represented in duplicate) is scarcely open to ques- tion, one can hardly avoid doubting the place of collection, as the genus Haplophyton is apparently limited in distribution to the semi-arid portions of northern Mexico and the south- western United States. Echites circinalis Sw. Prodr. 52. 1788 = Prestonia aggluti- nata (Jacq.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931 (Echites agglutinata Jacq. Enum. Pl. Carib. 13. 1760). Echites citrifolia HBK. Nov. Gen. 3: 216. 1819 = Mesechites citrifolia (HBK.) Woodson, Ann. Mo. Bot. Gard. 20: 637. 1933. Echites citrina A. DC. in DC. Prodr. 8: 474. 1844 = Mars- denia sp. (perhaps closely related to M. fusca Wright). Echites coalita Vell. Fl. Flum. 112. 1830; Icon. 3: pl. 40. 1827 = Prestonia coalita (Vell.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931. Echites Cobanensis Donn. Sm. Bot. Gaz. 40: 6. 1905 = Mande- villa tubiflora (Mart. & Gal.) Woodson, Ann. Mo. Bot. Gard. 19: 52. 1932 (Echites tubiflora Mart. € Gal. Bull. Acad. Roy. Brux. 11:: 358. 1844. Echites coccinea Hook. & Arn. in Hook. Jour. Bot. 1: 286. 1834 = Mandevilla coccinea (Hook. & Arn.) Woodson, Ann. Mo. Bot. Gard. 20: 734. 1933. Echites cognata Stadelm. Flora 24': Beibl. 79. 1841 = Odontadenia cognata (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 18: 546. 1931. Echites comosa O. Ktze. Rev. Gen. 2: 414. 1891 = Mande- villa villosa (Miers) Woodson, Ann. Mo. Bot. Gard. 19: 70. 1932 (Laseguea villosa Miers, Apoc. So. Am. 250. 1878). Echites concolor Ham. Prodr. 31. 1825. This may refer to Mesechites angustifolia (Poir.) Miers, Apoc. So. Am. 230. 1878 (Echites angustifolia Poir. Encycl. Suppl. 2: 537. 1812). Echites congesta HBK. Nov. Gen. 3: 214. 1819 - Mande- villa congesta (HBK.) Woodson, Ann. Mo. Bot. Gard. 20: 675. 1933 (404) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 233 Echites convolvulacea A. DC. in DC. Prodr. 8: 451. 1844 = Mandevilla convolvulacea (A. DC.) Hemsl. Biol. Centr.-Am. Bot. 2: 316. 1882. Echites cordata A. DC. in DC. Prodr. 8: 451. 1844. This is probably referable to Mandevilla convolvulacea (A. DC.) Hemsl. Biol. Centr.-Am. Bot. 2: 316. 1882 (Echites convol- vulacea A. DC. loc. cit. 1844). Echites coriacea Benth. in Hook. Jour. Bot. 3: 249. 1841 = Odontadenia geminata (R. & S.) Muell.-Arg. in Mart. Fl. Bras. 6': 119. 1860 ( Echites geminata R. € S. Syst. 4: 796. 1819). Echites corymbosa Jacq. Enum. Pl. Carib. 13. 1760 = For- steronia corymbosa (Jacq.) G. F. W. Meyer, Prim. Fl. Esseq. 134. 1818. Echites Coulteri S. Wats. Droe, Am. Acad. 18: 113. 1882-83 = Mandevilla Karwinskii (Muell.-Arg.) Hemsl. Biol. Centr.- Am. Bot. 2: 316. 1882 (Amblyanthera Karwinskii Muell.-Arg. Linnaea 30: 426. 1860). Echites crassinoda Gardn. ex Hook. Lond. Jour. Bot. 1: 544. 1842 = Mandevilla crassinoda (Gardn.) Woodson, Ann. Mo. Bot. Gard. 20: 703. 1933. Echites crassipes A. Rich. in Sagra, Hist. Cuba 11: 91. 1850 = Echites umbellata Jacq. var. crassipes (A. Rich.) Gomez, Anal. Soc. Espan. Hist. Nat. 23: 274. 1894. Echites cubensis (Muell.-Arg.) Griseb. Cat. Pl. Cub. 172. 1866 (Rhabdadenia cubensis Muell.-Arg. Linnaea 30: 435. 1860) = Angadenia Berterii (A.DC.) Miers, Apoc. So. Am. 180. 1878 (Echites Berteru A. DC. in DC. Prodr. 8: 447. 1844). Echites Cururu Mart. in Buchn. Rep. Pharm. 101. 1830 = Odontadenia puncticulosa (A. Rich.) Pulle, Enum. Pl. Surinam 383. 1906 (Echites puncticulosa A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792). Echites cuspidifera Blake, Contr. Gray Herb. 52: 79. 1917 = Mandevilla subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. & P. Fl. Peruv. 2: 19. 1799). Echites Cuyabensis A. DC. in DC. Prodr. 8: 462. 1844 = Mandevilla scabra (R. & S.) K. Sch. in Engl. & Prantl, Nat. (405) [Vor. 23 234 ANNALS OF THE MISSOURI BOTANICAL GARDEN Pflanzenfam. 47: 171. 1895 (Echites scabra R. € S. Syst. 4: 195. 1819). Echites cyamphylla Rusby, Bull. N. Y. Bot. Gard. 4: 409. 1907 = Prestonia cyaniphylla (Rusby) Woodson, Ann. Mo. Bot. Gard. 23: 284. 1936. Echites densevenulosa Stadelm. Flora 24': Beibl. 47. 1841 = Odontadenia lutea (Vell.) Mgf. in Fedde, Rep. Sp. Nov. 20: 24. 1924 (Echites lutea Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 25.1821). Echites densiflora Pohl, ex Stadelm. Flora 24!: Beibl. 56. 1841 - Mandevilla pycnantha (Steud.) Woodson, Ann. Mo. Bot. Gard. 19: 60. 1932 (Echites pycnantha Steud. Nomencl. ed. 2. 1: 540. 1840). Echites denticulata Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 30. 1827 = Prestonia denticulata (Vell.) Woodson, Ann. Mo. Bot. Gard. 23: 328. 1936. Echites dichotoma HBK. Nov. Gen. 3: 217. 1819. This ap- parently refers to Mesechites trifida (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 61: 151. 1860 (Echites trifida Jacq. Enum. PI. Carib. 13. 1760). Echites didyma Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 27. 1827 = Prestonia didyma (Vell.) Woodson, Ann. Mo. Bot. Gard. 23: 308. 1936. Echites difformis Walt. Fl. Carol. 98. 1788 = Trachelo- spermum difforme (Walt.) A. Gray, Syn. Fl. N. Am. 2!: 85. 1878. Echites disadena Miq. Stirp. Surinam. Select. 156. 1851 = Mesechites trifida (Jacq.) Muell.-Arg. in Mart. Fl. Bras. Gi: 151. 1860 (Echites trifida Jacq. Enum. Pl. Carib. 13. 1760). Echites dolichopetala Urb. Symb. Ant. 7: 335. 1912 = Asket- anthera dolichopetala (Urb.) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932. Echites domingensis Jacq. Icon. Pl. Rar. 1: 6. pl. 53. 1782 = Urechites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907 (Vinca lutea L. Cent. II. Pl. 12. 1756). Echites dubia Vell. Fl. Flum. 114. 1830; Icon. 3: pl. 50. 1827 = Cissampelos ovalifolia DC. Syst. 1: 537. 1818. (406) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 200 Echites Dusemü Malme, Arkiv f. Bot. 22A?: 9. 1928 = Prestonia Dusenii (Malme) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931. Echites Eggersu Mgf. Notizblatt 9: 78. 1924 = Laubertia Boissierii A. DC. in DC. Prodr.8: 487. 1844. Echites Ehrenbergu Schlecht. Linnaea 26: 666. 1853 = Rhabdadenia biflora (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 6': 175. 1860 ( Echites biflora Jacq. Enum. Pl. Carib. 13. 1760). Echites elegans Benth. in Hook. Jour. Bot. 3: 249. 1841 = Odontadenia geminata (R. & S.) Muell.-Arg. in Mart. Fl. Bras. 61: 119. 1860 (Echites geminata R. & S. Syst. 4: 795. 1819). Echites emarginata Vell. Fl. Flum. 113. 1830; Icon. 3: pl. 46. 1827 = Mandevilla erecta (Vell. Woodson, Ann. Mo. Bot. Gard. 19: 62. 1932 (Echites erecta Vell. loc. cit. 1830; loc. cit. pl. 45. 1827). Echites erecta Vell. Fl. Flum. 113. 1830; Icon. 3: pl. 45. 1827 = Mandevilla erecta (Vell.) Woodson, Ann. Mo. Bot. Gard. 19: 62. 1932. Echites erecta A. DC. in DC. Prodr. 8: 469. 1844, non Vell. = Rhodocalyx rotundifolius Muell.-Arg. in Mart. Fl. Bras. 6': 173. pl. 51. 1860. Echites exilicaulis Sesse € Moc. Fl. Mex. 45. 1887. Deserip- tion impossible to interpret. Echites ferruginea A. Rich. in Sagra, Hist. Cuba 11: 92. 1850 = Angadenia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878 ( Echites Berteru A. DC. in DC. Prodr. 8: 447. 1844). Echites floribunda Sw. Prodr. 52. 1788 = Forsteronia flori- bunda (Sw.) G. F. W. Meyer, Prim. Fl. Esseq. 135. 1818. Echites fluminensis A. DC. in DC. Prodr. 8: 452. 1844 = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792, non R. & P). Echites fragrans Stadelm. Flora 24': Beibl. 71. 1841 = Mandevilla fragrans (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 20: 713. 1933. Echites Franciscea A. DC. in DC. Prodr. 8: 452. 1844 = Temnadenia violacea (Vell.) Miers, Apoc. So. Am. 208. 1878 (407) [Vor. 23 236 ANNALS OF THE MISSOURI BOTANICAL GARDEN (Echites violacea Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 31. 1827). Echites Fraseri Willd. ex R. & S. Syst. 4: 796. 1819. The identity of this species cannot be ascertained from the ambigu- ous reference. Echites funiformis Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 29. 1827 = Mandevilla funiformis (Vell. K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895. Echites galegiformis Rudolph, ex Ledeb. Pl. S. Dom. 6. 1805. Reference too vague for identification. Echites geminata R. & S. Syst. 4: 795. 1819 = Odontadenia geminata (R. & S.) Muell.-Arg. in Mart. Fl. Bras. 6*: 119. 1860. Echites glandulosa Poir. Encycl. Suppl. 2: 537. 1812. Ref- erence too vague for identification. Echites glandulosa R. & P. Fl. Peruv. 2: 19. pl. 135. 1799 = Mandevilla glandulosa (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 66. 1932. Echites glaucescens Mart. € Gal. Bull. Acad. Roy. Brux. LU: 358. 1844 - Mandevilla oaxacana (A. DC.) Hemsl. Biol. Centr.- Am. Bot. 2: 316. 1882 (Echites Oaxacana A. DC. in DC. Prodr. 8: 451. 1844). Echites glomerata Poir. Encycl. Suppl. 2: 536. 1812. Pos- sibly asclepiadaceous. Echites gracilipes Stadelm. Flora 241: Beibl. 22. 1841 = Odontadenia gracilipes (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 22: 294. 1935. Echites gracilis HBK. Nov. Gen. 3: 219. 1819 = Mandevilla subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. & P. Fl. Peruv. 2: 19. 1799). Echites grandiflora Desf. ex Hook. Jour. Bot. 1: 286. 1834 = Macrosiphonia longiflora (Desf.) Muell.-Arg. in Mart. Fl. Bras. 67: 140. 1860 (Echites longiflora Desf. Mem. Mus. Paris 5: 275. pl. 20. 1819). Echites grandiflora G. F. W. Meyer, Prim. Fl. Esseq. 131. 1818 = Odontadenia Hoffmannseggiana (Steud.) Woodson, in Gleason & A. C. Smith, Bull. Torrey Bot. Club 60: 392. 1933 (Echites Hoffmannseggiana Steud. Nomencl. ed. 2. 1: 539. 1840). (408) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV Zot Echites Guarantica St. Hil. Mem. Mus. Paris 12: 324. 1825 = Macrosiphonia longiflora (Desf.) Muell.-Arg. in Mart. Fl. Bras. 6': 140. 1860 (Echites longiflora Desf. Mem. Mus. Paris 5: 275. 1819). Echites Guianensis A. DC. in DC. Prodr. 8: 458. 1844 = Mandevilla subspicata (Vahl) Mgf. Rec. Trav. Bot. Néerl. 22: 380. 1926 (Echites subspicata Vahl, Eclog. Am. 2: 18. 1798). Echites heterophylla J. F. Gmel. Syst. 2: 437. 1791 = Ur- echites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907 (Vinca lutea L. Cent. II. Pl. 12. 1756). Echites heterophylla Miq. Linnaea 25: 653. 1852, non Gmel. = Elytropus chilensis (A. DC.) Muell.-Arg. Linnaea 30: 440. 1860 (Echites Chilensis A. DC. in DC. Prodr. 8: 468. 1844). Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792 = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895. Echites hirsuta R. € P. Fl. Peruv. 2: 19. pl. 136. 1799, non A. Rich. - Mandevilla Pavonii (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 73. 1932 (Echites Pavonii A. DC. in DC. Prodr. 8: 463. 1844). Echites hirtella HBK. Nov. Gen. 3: 213. 1819 - Mandevilla subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. & P. Fl. Peruv. 2: 19. 1799). Echites hirtiflora A. DC. in DC. Prodr. 8: 456. 1844 - Mande- villa subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. & P. Fl. Peruv. 2: 19. 1799). Echites hispida Willd. ex R. & S. Syst. 4: 795. 1819 - Mande- villa hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792). Echites Hoff mannseggiana Steud. Nomencl. ed. 2. 1: 539. 1840 = Odontadenia Hoffmannseggiana (Steud.) Woodson, in Gleason & A. C. Smith, Bull. Torrey Bot. Club 60: 392. 1933. Echites holosericea Sesse E Moc. Fl. Mex. 45. 1887. De- scription impossible to interpret. Echites Hookeri A. DC. in DC. Prodr. 8: 476. 1844. Pos- sibly a species of Mandevilla. Type specimen (Tweedie s.n. in hb. Kew.) fragmentary. (409) [Vor. 23 238 ANNALS OF THE MISSOURI BOTANICAL GARDEN Echites Hulkiana Pulle, Rec. Trav. Bot. Néerl. 9: 160. 1912 = Prestonia acutifolia (Benth.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 188. 1895 (Haemadictyon acutifolium Benth. ex Muell.-Arg. in Mart. Fl. Bras. 61: 167. 1860). Echites hypoglauca Stadelm. Flora 24!: Beibl. 23. 1841 = Odontadenia hypoglauca (Stadelm.) Muell.-Arg. in Mart. Fl. Bras. Di: 118. 1860. Echites hypoleuca Benth. Pl. Hartw. 23. 1839 = Macro- siphonia hypoleuca (Benth.) Muell.-Arg. Linnaea 30: 452. 1860. Echites illustris Vell. Fl. Flum. 114. 1830; Icon. 3: pl. 49. 1827 = Mandevilla illustris (Vell.) Woodson, Ann. Mo. Bot. Gard. 20: 727. 1933. Echites insignis Spreng. Syst. 1: 632. 1825. Possibly refers to Odontadenia Hoff mannseggiana (Steud.) Woodson. Echites istmica Vell. Fl. Flum. 112. 1830; Icon. 3: pl. 39. 1827 = Condylocarpon sp. Echites jamaicensis Griseb. Fl. Brit. W. I. 416. 1861 = Urechites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907 (Vinca lutea L. Cent. II. Pl. 12. 1756). Echites japurensis Stadelm. Flora 24': Beibl. 19. 1841. = Mesechites trifida (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 6!: 151. 1860 ( Echites trifida Jacq. Enum. Pl. Carib. 13. 1760). Echites jasminiflora Mart. € Gal. Bull. Acad. Roy. Brux. 11!: 357. 1844 = Mandevilla subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. € P. Fl. Peruv.2: 19. 1799). Echites javitensis HBK. Nov. Gen. 3: 220. 1819 = Mande- villa javitensis (HBK.) K. Sch. in Engl. & Prantl, Nat. Pflan- zenfam. 4?: 171. 1895. Echites Karwinskii (Muell.-Arg.) Miers, Apoc. So. Am. 206. 1878 ( Amblyanthera Karwinskii Muell.-Arg. Linnaea 30: 426. 1860) = Mandevilla Karwinskii (Muell.-Arg.) Hemsl. Biol. Centr.-Am. Bot. 2: 316. 1882. Echites lanata Mart. & Gal. Bull. Acad. Roy. Brux 11!: 359. 1844. Possibly refers to Mandevilla subsagittata (R. & P.) Woodson. The type specimen evidently has been lost. (410) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 239 Echites lanuginosa Mart. € Gal. Bull. Acad. Roy. Brux. 111: 357. 1844 = Macrosiphonia lanuginosa (Mart. & Gal.) Hemsl. Biol. Centr.-Am. Bot. 2: 316. 1882. Echites lanuginosa Sesse & Moc. Fl. Mex. 44. 1887, non Mart. & Gal. Impossible to interpret. Echites lappulacea Lam. Encycl. 2: 341. 1786 = Anechites lappulacea (Lam.) Miers, Apoc. So. Am. 237. 1878. Echites lasiocarpa A. DC. in DC. Prodr. 8: 463. 1844 = Mandevilla lasiocarpa (A. DC.) Malme, Bihang till K. Sv. Vet. Akad. Handl Afd. III. 241°: 25. 1899. Echites lateriflora Sesse & Moc. La Naturaleza II. 1: Suppl. 28. 1888. Impossible to interpret. Echites Laurentiae-disca Rusby, Descr. So. Am. Pl. 85. 1920 = Prestonia acutifolia (Benth.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 188. 1895 (Haemadictyon acutifolium Benth. ex Muell.-Arg. in Mart. Fl. Bras. 6': 167. 1860). Echites laxa R. & P. Fl. Peruv. 2: 19. pl. 134b. 1799 = Mande- villalaxa (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 68. 1932. Echites leptoloba Stadelm. Flora 24!': Beibl. 15. 1841 = Prestonia agglutinata (Jacq.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931 (Echites agglutinata Jacq. Enum. Pl. Carib. 13. 1760). Echites leptophylla A. DC. in DC. Prodr. 8: 455. 1844 = Mandevilla leptophylla ( A. DC.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895. Echites Lindeniana (Muell.-Arg.) Griseb. Cat. Pl. Cub. 173. 1866 (Rhabdadenia Lindeniana Muell.-Arg. Linnaea 30: 437. 1860) = Angadenia Lindeniana (Muell.-Arg.) Miers, Apoc. So. Am. 180. 1878. Echites linearifolia Ham. Prodr. 31. 1825 = Mesechites an- gustifolia (Poir.)Miers, Apoc. So. Am. 230. 1878 (Echites angustifolia Poir. Encycl. Suppl. 2: 537. 1812). Echites linearifolia Stadelm. Flora 24': Beibl. 18. 1841, non Ham. = Mandevilla leptophylla (A. DC.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites leptophylla A. DC. in DC. Prodr. 8: 455. 1844). (411) [Vor. 23 240 ANNALS OF THE MISSOURI BOTANICAL GARDEN Echites linearis Vell. Fl. Flum. 111. 1830; Icon. 3: pl. 36. 1827. Possibly referable to Forsteronia. Echites longiflora Desf. Mem. Mus. Paris 5: 275. 1819 = Macrosiphonia longiflora (Desf.) Muell.-Arg. in Mart. Fl. Bras. 6': 140. 1860. Echites longiflora Ekm. € Helwig, Arkiv f. Bot. 22A": 45. 1929, non Desf. = Asketanthera Ekmaniana Woodson, Ann. Mo. Bot. Gard. 23: 267. 1936. Echites longifolia Besse E Moc. Fl. Mex. 45. 1887. May pos- sibly refer to E. tuxtlensis Standl. Echites lucida R. & S. Syst. 4: 795. 1819 = Odontadenia nitida (Vahl) Muell.-Arg. in Mart. Fl. Bras. 6': 118. 1860 (Echites nitida Vahl, Eclog. 2: 19. pl. 13. 1798). Echites lutea Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 25. 1827 = Odontadenia lutea (Vell.) Mgf. in Fedde, Rep. Sp. Nov. 20: 24. 1924. Echites macrantha R. & S. Syst. 4: 795. 1819, non Spreng. = Odontadenia Hoffmannseggiana (Steud.) Woodson, in Gleason & A. C. Smith, Bull. Torrey Bot. Club 60: 392. 1933 (Echites Hoff mannseggiana Steud. Nomencl. ed. 2. 1: 539. 1840). Echites macrocalyx Muell.-Arg. in Mart. Fl. Bras. 6': 160. 1860 = Peltastes peltatus (Vell.) Woodson, Ann. Mo. Bot. Gard. 19: 376. 1932 (Echites peltata Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 32. 1827). Echites ? macrocarpa A. Rich. in Sagra, Hist. Cuba 11: 94. 1850, non Wall. = Catalpa macrocarpa (A. Rich.) Ekman, in Urb. Symb. Ant. 9: 254. 1924. Echites macrophylla HBK. Nov. Gen. 3: 218. 1819, non Roxb. = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792. Echites macrophylla A. Zahlbr. Ann. K. K. Naturh. Hofmus. Wien 7: 5. 1892, nec HBK., nec Roxb. = Mandevilla subpanicu- lata Woodson, Ann. Mo. Bot. Gard. 19: 71. 1932. Echites Macrosiphon Torr. Bot. Mex. Bound. Surv. 158. pl. 43. 1859 = Macrosiphonia Macrosiphon (Torr.) A. A. Heller, Muhlenbergia 1: 2. 1900. (412) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 241 Echites macrostoma Benth. in Hook. Jour. Bot. 3: 248. 1841 = Rhabdadenia macrostoma (Benth.) Muell.-Arg. Linnaea 30: 435. 1860. Echites maculata (Descourt.) A. DC. in DC. Prodr. 8: 474. 1844 (Apocynum maculatum Descourt. Fl. Med. Antill. 3: 176. pl. 190. 1827). Probably referable to the asclepiadaceous genus Marsdenia. Echites madida Vell. Fl. Flum. 112. 1830; Icon. 3: pl. 42. 1827. Possibly referable to Mandevilla. Echites Mansoana A. DC. in DC. Prodr. 8: 448. 1844 = Mesechites Mansoana (A. DC.) Woodson, Ann. Mo. Bot. Gard. 20: 636. 1933. Echites mapirensis H. Winkl. in Fedde, Rep. Sp. Nov. 7: 113. 1909. Possibly referable to Mesechites acuminata (R. & P.) Muell.-Arg. Linnaea 30: 446. 1860. The type specimen (Buch- tien 1954) has not been available for study. Echites Maranhamensis G. Don, Gen. Hist. 4: 74. 1838 = Mandevilla scabra (R. & S.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4%: 171. 1895 (Echites scabra R. € S. Syst. 4: 795. 1819). Echites Martiana Stadelm. Flora 24!: Beibl. 31. 1841 = Mandevilla Martiana (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 20: 702. 1933. Echites Martii Muell.-Arg. in Mart. Fl. Bras. 6': 155. 1860 — Prestonia coalita (Vell.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931 (Echites coalita Vell. Fl. Flum. 112. 1830; Icon. 3: pl. 40. 1827). Echites Maximilianea Stadelm. Flora 24*: Beibl. 43. 1841 = Temnadenia violacea (Vell.) Miers, Apoc. So. Am. 208. 1878 (Echites violacea Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 31. 1827). Echites Meg’agros Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 33. 1827 = Prestonia Meg'agros (Vell. Woodson, Ann. Mo. Bot. Gard. 23: 329. 1936. Echites membranacea A. DC. in DC. Prodr. 8: 457. 1844 = Mandevilla subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. & P. Fl. Peruv. 2: 19. 1799). (413) [Vor. 23 242 ANNALS OF THE MISSOURI BOTANICAL GARDEN Echites mexicana (Muell.-Arg.) Miers, Apoc. So. Am. 205. 1878 (Amblyanthera mexicana Muell.-Arg. Linnaea 30: 424. 1860) = Mandevilla mexicana (Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 19: 65. 1932. Echites microcalyx A. DC. in DC. Prodr. 8: 456. 1844 = Mandevilla subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. & P. Fl. Peruv. 2: 19. 1799). Echites microphylla Stadelm. Flora 24!: Beibl. 35. 1841 = Mandevilla funiformis (Vell.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites funiformis Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 29. 1827). Echites minima Britton € Wilson, Mem. Torrey Bot. Club 16: 94. 1920 = Mesechites minima (Britton & Wilson) Wood- son, Ann. Mo. Bot. Gard. 19: 386. 1932. Echites mollissima HBK. Nov. Gen. 3: 218. 1819 = Mande- villa mollissima (HBK.) K. Sch. in Engl. & Prantl, Nat. Pflan- zenfam. 4?: 171. 1895. Echites montana HBK. Nov. Gen. 3: 213. 1819 = Mandevilla montana (HBK.) Mgf. Notizblatt 9: 82. 1924. Echites mucronata R. & S. Syst. 4: 796. 1819 = Mandevilla subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. & P. Fl. Peruv. 2: 19. 1799). Echites ? muricata A. DC. in DC. Prodr. 8: 474. 1844. Based upon Descourt. Fl. Med. Antill. 3: 171. pl. 189. 1827. Evidently asclepiadaceous. Echites myrtifolia R. & S. Syst. 4: 795. 1819, non Poir. = Mesechites rosea (A. DC.) Miers, Apoc. So. Am. 232. 1878 (Echites rosea A. DC. in DC. Prodr. 8: 450. 1844). Echites neriandra Griseb. Fl. Brit. W. I. 415. 1861 = Ur- echites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907 (Vinca lutea L. Cent. IT. Pl. 12. 1756). Echites nitida Vahl, Eclog. 2: 19. pl. 13. 1798 = Odontadenia nitida (Vahl) Muell.-Arg. in Mart. Fl. Bras. 6*: 118. 1860. Echites nodosa R. & S. Syst. 4: 796. 1819. Based upon a sterile specimen which is impossible to determine accurately. (414) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 243 Echites nutans Anders. Trans. Soc. Arts London 25: 203. 1807 = Prestonia quinquangularis (Jacq.) Spreng. Syst. 1: 637. 1825 (Echites quinquangularis Jacq. Enum. Pl. Carib. 13. 1760). Echites Oaxacana A. DC. in DC. Prodr. 8: 451. 1844 = Mandevilla oaxacana (A. DC.) Hemsl. Biol. Centr.-Am. Bot. 2: 316. 1882. Echites oblongifolia Ham. Prodr. 30. 1825. Perhaps refer- able to Mesechites. Echites obovata Nees, ex Steud. Nomencl. ed. 2. 1: 540. 1840. Based upon E. variegata Schrad. Goett. Gel. Anz. 1: 707. 1821, a nomen subnudum incapable of interpretation. Echites obovata Sesse E Moe. Fl. Mex. 43. 1887. May refer to Urechites lutea (L.) Britton. Echites obtusifolia Sesse & Moc. Fl. Mex. 45. 1887. Impos- sible to interpret. Echites odorifera Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 28. 1827. The plate cited has somewhat the aspect of a Temna- denia, but is not capable of identification. Echites ornata Hoehne, Comm. Linh. Telegr. Estrat. Matto Grosso, Annexo 5, Bot. 6: 82. pls. 120; 131, fig. 1. 1915 = Tem- nadenia ornata (Hoehne) Woodson, Ann. Mo. Bot. Gard. 19: 983. 1932. Echites ? ovalifolia Hook. & Arn. in Hook. Jour. Bot. 1: 286. 1834. Type specimen (Tweedie s.n. in Hb. Kew.) fragmentary, possibly asclepiadaceous. Echites ovalifolia Poir. Encycl. Suppl. 2: 535. 1812. Con- sidered by Miers ( Apoc. So. Am. 248. 1878) to be a Forsteronia. If so, it possibly represents F. spicata (Jacq.) G. F. W. Meyer, of which, however, we have no authenticated specimens from Hispaniola. Echites ovalis Mgf. Notizblatt 9: 79. 1924 = Allomarkgrafia ovalis (Mgf.) Woodson, Ann. Mo. Bot. Gard. 19: 45. 1932. Echites pallida Miers, Apoc. So. Am. 195. 1878 = Mesechites trifida (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 6': 151. 1860 ( Echites trifida Jacq. Enum. Pl. Carib. 13. 1760). (415) [Vor. 23 244 ANNALS OF THE MISSOURI BOTANICAL GARDEN Echites paludosa Vahl, Eclog. 2: 19. 1798 = Rhabdadenia biflora (Jacq.) Muell.-Arg. in Mart. Fl. Bras. Gi: 175. 1860 (Echites biflora Jacq. Enum. Pl. Carib. 13. 1760). Echites pandurata A. DC. in DC. Prodr. 8: 458. 1844 = Fernaldia pandurata (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 48. 1932. Echites paniculata Poir. Encycl. Suppl. 2: 536. 1811. Diag- nosis impossible to interpret. Echites parviflora Sesse & Moc. Fl. Mex. 44. 1887. Impos- sible to interpret. Echites pastorum Mart. ex Stadelm. Flora 24!: Beibl. 52. 1841 = Mandevilla tenuifolia (Mikan) Woodson, Ann. Mo. Bot. Gard. 20: 679. 1933 (Echites tenuifolia Mikan, Fl. & Faun. Bras. fasc. 3. 1820). Echites Pavonu A. DC. in DC. Prodr. 8: 463. 1844 = Mande- villa Pavonii (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 73. 1932. Echites peduncularis Stadelm. Flora 24': Beibl. 54. 1841 = Mandevilla tenuifolia (Mikan) Woodson, Ann. Mo. Bot. Gard. 20: 679. 1933 (Echites tenuifolia Mikan, Fl. & Faun. Bras. fasc. 3. 1820). Echites peltata Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 32. 1827 = Peltastes peltatus (Vell.) Woodson, Ann. Mo. Bot. Gard. 19: 376. 1932. Echites peltigera Stadelm. Flora 24': Beibl. 21. 1841 = Stipecoma peltigera (Stadelm.) Muell.-Arg. in Mart. Fl. Bras. 6': 176. 1860. Echites petraea St. Hil. Mem. Mus. Paris 12: 322. 1825 = Macrosiphonia petraea (St. Hil.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 168. 1895. Echites Picardae Urb. Symb. Ant. 5: 466. 1908 — Asket- anthera Picardae (Urb.) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932. Echites pilosa Vell. Fl. Flum. 112. 1830; Icon. 3: pl. 38. 1897 = Forsteronia pilosa (Vell.) Muell.-Arg. in Mart. Fl. Bras. 6': 99. 1860. Echites pinguifolia Standl. Field Mus. Publ. Bot. 8: 35. 1930 (416) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 245 = Fernaldia pandurata (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 48. 1932. Echites pinifolia St. Hil. Mem. Mus. Paris 12: 325. 1825 = Macrosiphonia petraea (St. Hil.) K. Sch. var. minor (Hook.) Woodson, Ann. Mo. Bot. Gard. 23: 376. 1936 (Echites grandi- flora Desf. var. minor Hook. Jour. Bot. 1: 286. 1834). Echites plicata A. DC. in DC. Prodr. 8: 454. 1844 = Peltastes peltatus (Vell.) Woodson, Ann. Mo. Bot. Gard. 19: 376. 1932 (Echites peltata Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 32. 1827). Echites Pohliana Stadelm. Flora 24': Beibl. 73. 1841 = Mandevilla velutina (Mart. Woodson var. angustifolia (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 23: 376. 1936 (Echites Pohliana Stadelm. var. a angustifolia Stadelm. loc. cit. 1841). Echites portobellensis Beurl. Vet. Akad. Handl. Stockh. 137. . 1854 (1856) = Prestonia portobellensis (Beurl.) Woodson, Ann. Mo. Bot. Gard. 18: 553. 1931. Echites Prieurei A. DC. in DC Prodr. 8: 458. 1844 - Mande- villa subspicata (Vahl) Mgf. Rec. Trav. Bot. Néerl. 22: 380. 1926 (Echites subspicata Vahl, Eclog. 2: 18. 1798). Echites psidufolia Mart. ex Stadelm. Flora 24!: Beibl. 46. 1841. Possibly refers to Odontadenia. The type specimen has apparently been lost. Echites ptarmica Poepp. Gen. 3: 69. pl. 278. 1845 = Ely- tropus chilensis (A. DC.) Muell.-Arg. Linnaea 30: 440. 1860 (Echites Chilensis A. DC. in DC. Prodr. 8: 468. 1844). Echites puberula Michx. Fl. Bor. Am. 1: 120. 1803 - Trache- lospermum difforme (Walt.) A. Gray, Syn. Fl. N. Am. 2: 85. 1878 (Echites difformis Walt. Fl. Carol. 98. 1788). Echites pubescens Hook. & Arn. Bot. Beechey Voy. 34. 1830, non R. & S. - Elytropus chilensis (A. DC.) Muell.-Arg. Linnaea 30: 440. 1860 (Echites Chilensis A. DC. in DC. Prodr. 8: 468. 1844). Echites pubescens R. € S. Syst. 4: 796. 1819 = Mandevilla congesta (HBK.) Woodson, Ann. Mo. Bot. Gard. 20: 675. 1933 (Echtes congesta HBK. Nov. Gen. 3: 214. 1819). (417) [Vor. 23 246 ANNALS OF THE MISSOURI BOTANICAL GARDEN Echites pubiflora G. Don, Gen. Hist. 4: 73. 1838 = Mande- villa scabra (R. & S.) K. Sch. in Engl. &. Prantl, Nat. Pflan- zenfam. 4?: 171. 1895 (Echites scabra R. & S. Syst. 4: 795. 1819). Echites pulchella Gardn. ex Hook. Icon. P1. 5: pl. 470. 1842 = Mandevilla spigeliaeflora (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 20: 736. 1933 (Echites (?)spigeliaeflora Stadelm. Flora 24': Beibl. 58. 1841). Echites puncticulosa A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792 = Odontadenia puncticulosa (A. Rich.) Pulle, Enum. Pl. Surinam, 383. 1906. Echites pycnantha Steud. ex A. DC. in DC. Prodr. 8: 469. 1844 - Mandevilla pycnantha (Steud.) Woodson, Ann. Mo. Bot. Gard. 19: 60. 1932. Echites quinquangularis Jacq. Enum. Pl. Carib. 13. 1760 = Prestonia quinquangularis (Jacq.) Spreng. Syst. 1: 637. 1825. Echites repens Jacq. Enum. Pl. Carib. 13. 1760 = Mesechites repens (Jacq.) Miers, Apoc. So. Am. 229. 1878. Echites revoluta A. DC. in DC. Prodr. 8: 457. 1844 = Pres- tonia agglutinata (Jacq.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931 (Echites agglutinata Jacq. Enum. Pl. Carib. 13. 1760). Echites revoluta Sesse & Moe. Fl. Mex. 44. 1887. Impossible to interpret. Echites Richardi R. & S. Syst. 4: 391. 1819 = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 42: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. .1792). Echites (?) Riedelii (Muell.-Arg.) Malme, Bull. Herb. Boiss. II. 4: 196. 1904 (Haemadictyon Riedelii Muell.-Arg. in Mart. Fl. Bras. Di: 170. 1860) = Prestonia Riedelii (Muell.-Arg.) Mgf. in Fedde, Rep. Spec. Nov. 20: 26. 1924. Echites rigida Rusby, Mem. N. Y. Bot. Gard. 7: 325. 1927 = Mesechites trifida (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 6': 151. 1860 (Echites trifida Jacq. Enum. Pl. Carib. 13. 1760). Echites riparia HBK. Nov. Gen. 3: 214. 1819 = Mandevilla riparia (HBK.) Woodson, Ann. Mo. Bot. Gard. 19: 58. 1932. (418) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 247 Echites Rosa-campestris Endl. in Harting, Parad. Vindob. 1: pl. 51. 1844-47 = Mandevilla illustris (Vell.) Woodson, Ann. Mo. Bot. Gard. 20: 727. 1933 (Echites illustris Vell. Fl. Flum. 114. 1830; Icon. 3: pl. 49. 1827). Echites Rosana Donn. Sm. Bot. Gaz. 40: 6. 1905 - Mande- villa Rosana (Donn. Sm.) Woodson, Ann. Mo. Bot. Gard. 20: 652. 1932. Echites rosea A. DC. in DC. Prodr. 8: 450. 1844 = Mesechites rosea (A. DC.) Miers, Apoc. So. Am. 232. 1878. Echites rubricaulis Poir. Encycl. Suppl. 2: 535. 1812. Ap- parently refers to Mesechites trifida (Jacq.) Muell.-Arg. or a related species. Echites Rugeliana Urb. Symb. Ant. 5: 465. 1908 — Asket- anthera calycosa (A. Rich.) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932 (Echites calycosa A. Rich. in Sagra, Hist. Cuba 11: 94. 1850). Echites rugellosa A. Rich. Actes Soc. Nat. Hist. Paris 1: 107. 1792. Incapable of identification. Echites rugosa Benth. in Hook. Jour. Bot. 3: 248. 1841 = Mandevilla rugosa (Benth.) Woodson, Ann. Mo. Bot. Gard. 19: 384. 1932. Echites sagittata Poir. Encycl. Suppl. 2: 537. 1812. Inca- pable of identifieation. Perhaps referable to Mandevilla sub- sagittata (R. & P.) Woodson. Echites Sagraei A. DC. in DC. Prodr. 8: 450. 1844 = Anga- denia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878 (Echites Berterii A. DC. loc. cit. 447. 1844). Echites salicifolia Raf. New Fl. N. Am. 4: 59. 1856, non Willd. = Trachelospermum difforme (Walt.) A. Gray, Syn. Fl. N. Am. 2: 85. 1878 (Echites difformis Walt. Fl. Carol. 98. 1788). Echites salicifolia Willd. ex R. & S. Syst. 4: 796. 1819 = Allemanda cathartica L. Mant. 214. 1771. Echites sancta Stadelm. Flora 24!: Beibl. 59. 1841 = Mande- villa sancta (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 20: 726. 1933. Echites Sanctae-Crucis S. Moore, Trans. Linn. Soc. Bot. II. (419) [VoL. 23 248 ANNALS OF THE MISSOURI BOTANICAL GARDEN 4: 396. 1895 = Mesechites Sanctae-Crucis (S. Moore) Wood- son, Ann. Mo. Bot. Gard. 19: 387. 1932. Echites Sanctae-Martae Rusby, Deser. So. Am. Pl. 85. 1920 = Laubertia Sanctae-Martae (Rusby) Woodson, Ann. Mo. Bot. Gard. 18: 555. 1931. Echites sanguinolenta Tussac, Fl. Ant. 95. pl. 11. 1808 = Prestonia agglutinata (Jacq.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931 (Echites agglutinata Jacq. Enum. Pl. Carib. 13. 1760). Echites scabra R. & S. Syst. 4: 795. 1819 = Mandevilla scabra (R. & S.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895. Echites secunda Sesse € Moc. Fl. Mex. 44. 1887. May refer to Mandevilla subsagittata (R. & P.) Woodson. Echites secundiflora A. DC. in DC. Prodr. 8: 457. 1844 = Mandevilla subsagittata (R. & P.) Woodson (Echites sub- sagittata R. & P. Fl. Peruv. 2: 19. 1799). Echites semidigyna Berg. Verh. Zeeusch. Gen. Wetens. 3: 583. 1773. Probably a species of Tabernaemontana. Echites sessilis Vell. Fl. Flum. 111. 1830; Icon. 3: pl. 35. 1827. Incapable of identification. Supposed by Mueller to be a Malouetia; a Thyrsanthus (Forsteronia) according to Miers. Echites Smithii Greenm. Proc. Am. Acad. 40: 29. 1904 = Mandevilla mexicana (Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 19: 65. 1932 (Amblyanthera mexicana Muell.-Arg. Lin- naea 30: 424. 1860). Echites speciosa HBK. Nov. Gen. 3: 219. 1819 = Mandevilla subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. & P. Fl. Peruv. 2: 19. 1799). Echites spectabilis Stadelm. Flora 24': Beibl. 44. 1841 = Macropharynx spectabilis (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931. Echites spicata Jacq. Enum. Pl. Carib. 13. 1760 = Forster- onia spicata (Jacq.) G. F. W. Meyer, Prim. Fl. Esseq. 135. 1818. Echites (?) spigeliaeflora Stadelm. Flora 24!: Beibl. 58. 1841 = Mandevilla spigeliaeflora (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 20: 736. 1933. (420) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 249 Echites splendens Hook. f. Bot. Mag. n.s. 16: pl. 3976. 1842 = Mandevilla splendens (Hook. f.) Woodson, Ann. Mo. Bot. Gard. 20: 707. 1933. Echites Stadelmeyeri Mart. ex Stadelm. Flora 24*: Beibl. 29. 1841 - Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 47: 171. 1895. Echites stellaris Lindl. Bot. Reg. 20: pl. 1664. 1834 = Temnadenia stellaris (Lindl.) Miers, Apoc. So. Am. 210. 1878. Echites stellulifera Lem. Jard. Fleur. 1: pl. 67. 1851. Ap- parently referable to T'emnadenia stellaris (Lindl.) Miers. Echites suaveolens (Lindl.) A. DC. in DC. Prodr. 8: 452. 1844 (Mandevilla suaveolens Lindl. Bot. Reg. n.s. 3: pl. 7. 1840 - Mandevilla laxa (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 68. 1932 (Echites laxa R. & P. Fl. Peruv. 2: 19. pl. 134b. 1799). Echites suaveolens Mart. € Gal. Bull. Acad. Roy. Brux. LU: 356. 1844, non A. DC. = Macrosiphonia hypoleuca (Benth.) Muell.-Arg. Linnaea 30: 452. 1860 (Echites hypoleuca Benth. Pl. Hartw. 23. 1839). Echites subcarnosa Benth. in Hook. Jour. Bot. 3: 247. 1841 = Mandevilla subcarnosa (Benth.) Woodson, in Gleason, Bull. Torrey Bot. Club 58: 455. 1931. Echites subcordata Sesse & Moc. Fl. Mex. 44. 1887. Impos- sible to interpret. Echites suberecta Jacq. Enum. Pl. Carib. 13. 1760 = Ur- echites lutea (L.) Britton, Bull. N. Y. Bot. Gard. 5: 316. 1907 (Vinca lutea L. Cent. II. Pl. 12. 1756). Echites suberosa Vell. Fl. Flum. 111. 1830; Icon. 3: pl. 34. 1827 = Prestonia denticulata (Vell. Woodson, Ann. Mo. Bot. Gard. 23: 328. 1936 (Echites denticulata Vell. loc. cit. 110. 1830; Icon. 3: pl. 30. 1827). Echites subsagittata R. & P. Fl. Peruv. 2: 19. 1799 = Mande- villa subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932. Echites subsessilis A. DC. in DC. Prodr. 8: 451. 1844 = Mandevilla subsessilis (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 59. 1932. (421) [Vor. 23 250 ANNALS OF THE MISSOURI BOTANICAL GARDEN Echites subspicata Vahl, Eclog. 2: 18. 1798 = Mandevilla subspicata (Vahl) Mgf. Rec. Trav. Bot. Néerl. 22: 380. 1926. Echites sulphurea Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 26. 1827. Possibly referable to Prestonia coalita (Vell.) Woodson. Echites surinamensis Miq. Stirp. Surinam. Select. 155. 1850 = Mesechites trifida (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 61: 151. 1860 (Echites trifida Jacq. Enum. Pl. Carib. 13. 1760). Echites sylvestris A. DC. in DC. Prodr. 8: 464. 1844 = Odontadenia Hoffmannseggiana (Steud.) Woodson, in Gleason & A. C. Smith, Bull. Torrey Bot. Club 60: 392. 1933 (Echites Hoff mannseggiana Steud. Nomencl. ed. 2. 1: 539. 1840). Echites symphitocarpa G. F. W. Meyer, Prim. Fl. Esseq. 132. 1818 -Mandevilla symphitocarpa (G. F. W. Meyer) Woodson, Ann. Mo. Bot. Gard. 19: 70. 1932. Echites syphilitica L.f. Suppl. 167. 1781. Incapable of identification. Echites tenuicaulis Stadelm. Flora 24': Beibl. 40. 1841 = Mandevilla scabra (R. & S.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites scabra R. € S. Syst. 4: 195. 1819). Echites tenuifolia Mikan, Fl. € Faun. Bras. fase. 3. 1820 = Mandevilla tenuifolia (Mikan) Woodson, Ann. Mo. Bot. Gard. 20: 679. 1933. Echites thyrsoidea Vell. Fl. Flum. 111. 1830; Icon. 3: pl. 37. 1827 = Forsteronia thyrsoidea (Vell.) Muell.-Arg. in Mart. Fl. Bras. 6t: 105. 1860. Echites tomentosa Raf. Fl. Ludovic. 46. 1819, non Vahl. Perhaps referable to Trachelospermum difforme (Walt.) A. Gray. Echites tomentosa Vahl, Symb. Bot. 3: 44. 1794 = Mande- villa hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflan- zenfam. 4?: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792). Echites torosa Jacq. Enum. Pl. Carib. 13. 1760 = Mandevilla torosa (Jacq.) Woodson, Ann. Mo. Bot. Gard. 19: 64. 1932. Echites torulosa L. Sp. Pl. ed. 2. 307. 1762 = Mandevilla torosa (Jacq.) Woodson, Ann. Mo. Bot. Gard. 19: 64. 1932 (Echites torosa Jacq. Enum. Pl. Carib. 13. 1760). (422) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 201 Echites trifida Jacq. Enum. Pl. Carib. 13. 1760 = Mesechites trifida (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 61: 151. 1860. Echites tropaeolifolia A. DC. in DC. Prodr. 8: 447. 1844 = Stipecoma peltigera (Stadelm.) Muell.-Arg. in Mart. Fl. Bras. 6': 176. 1860 (Echites peltigera Stadelm. Flora 24*: Beibl. 21. 1841). Echites tubiflora Mart. E Gal. Bull. Acad. Roy. Brux. 111: 398. 1844 - Mandevilla tubiflora (Mart. & Gal.) Woodson, Ann. Mo. Bot. Gard. 19: 52. 1932. Echites tubulosa Benth. in Hook. Jour. Bot. 3: 249. 1841 = Mesechites trifida (Jacq.) Muell.-Arg. in Mart. Fl. Bras. 6!: 151. 1860 ( Echites trifida Jacq. Enum. Pl. Carib. 13. 1760). Echites Tweediana Hieron. Bol. Acad. Nac. Cordova 4: 370. 1881. Probably referable to Mandevilla erecta (Vell.) Wood- son. Echites umbellata Sesse & Moc. Fl. Mex. 43. 1887, non Jacq. Apparently refers to a species of Thenardia. Echites undulata Besse E Moc. Fl. Mex. 44. 1887. Impos- sible to interpret. Echites uniflora Besse E Moc. La Naturaleza II. 1: Suppl. 28. 1888. Impossible to interpret. Echites Valenzuelana A. Rich. in Sagra, Hist. Cuba 11: 93. 1850 = Neobracea Valenzuelana (A. Rich.) Urb. Symb. Ant. 9: 241. 1924. Echites varia Stadelm. Flora 241: Beibl. 17. 1841 = Temna- denia stellaris (Lindl.) Miers, Apoc. So. Am. 210. 1878 ( Echites stellaris Lindl. Bot. Reg. 20: pl. 1664. 1835). Echites variegata Schrad. Goett. Gel. Anz. 1: 707. 1821. Perhaps referable to Prestonia agglutinata (Jacq.) Woodson. Echites Vauthieri A. DC. in DC. Prodr. 8: 457. 1844 = Pres- tonia coalita (Vell.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931 (Echites coalita Vell. Fl. Flum. 112. 1830; Icon. 3: pl. 40. 1827). Echites Velame St. Hil. Bull. Soc. Phil. 77. 1824 = Macro- siphonia Velame (St. Hil.) Muell.-Arg. in Mart. Fl. Bras. 6!: 138. 1860. Echites Velloziana A. DC. in DC. Prodr. 8: 474. 1844 = (423) [Vor. 23 252 ANNALS OF THE MISSOURI BOTANICAL GARDEN Forsteronia Velloziana (A. DC.) Woodson, Ann. Mo. Bot. Gard. 21: 622. 1934. Echites velutina Mart. ex Stadelm. Flora 24': Beibl. 72. 1841 = Mandevilla velutina (Mart.) Woodson, Ann. Mo. Bot. Gard. 20731. 1933; Echites venenosa Stadelm. Flora 24': Beibl. 66. 1841 = Mandevilla illustris (Vell.) Woodson, Ann. Mo. Bot. Gard. 20: 727. 1933 (Echites illustris Vell. Fl. Flum. 114. 1830; Icon. 3: pl. 49. 1827). Echites Veraguasensis Seem. Bot. Voy. Herald, 168. 1854 = Mandevilla veraguasensis (Seem.) Hemsl. Biol. Centr.-Am. Bot. 2: 317. 1882 (where misspelled veraguensis). Echites verrucosa R. € S. Syst. 4: 795. 1819 = Odontadenia verrucosa (R. & S.) K. Sch. ex Mgf. in Pulle, Fl. Surinam 4: 53. 1932. Echites versicolor Mart. ex Stadelm. Flora 24!: Beibl. 38. 1841 - Mandevilla scabra (R. & S.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 47: 171. 1895 (Echites scabra R. € S. Syst. 4: 195. 1819). Echtes verticillata Besse € Moc. Fl. Mex. 43. 1887. The authors propose two species under the same specific adjective: one reported from Mexico and one from Porto Rico. Both are impossible to interpret. Echites violacea Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 31. 1827 = Temnadenia violacea (Vell.) Miers, Apoc. So. Am. 208. 1878. Echites virescens St. Hil. Bull. Soc. Phil. 77. 1824 = Macro- siphonia virescens (St. Hil.) Muell.-Arg. in Mart. Fl. Bras. Bi: 139. 1860. Echtes (?) xanthostoma Stadelm. Flora 24': Beibl. 55. 1841 = Mandevilla coccinea (Hook. & Arn.) Woodson, Ann. Mo. Bot. Gard. 20: 734. 1933 (Echites coccinea Hook. & Arn. in Hook. Jour. Bot. 1: 286. 1834). Echites Zuccariniana Stadelm. Flora 24': Beibl. 76. 1841 = Odontadenia lutea (Vell.) Mgf. in Fedde, Rep. Sp. Nov. 20: 24. 1924 (Echites lutea Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 25. 1827). (424) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 253 XXI. Temnapenta Miers, char. emend. Temnadenia Miers, Apoc. So. Am. 207. 1878, in part. Lactescent, fruticose lianas. Stems volubile, rarely suberect, terete; branches alternate. Leaves opposite, petiolate to sub- sessile, entire, penninerved, eglandular, the petioles somewhat girdling at the nodes into a slightly dilated, minutely appendic- ulate, stipular ring. Inflorescence lateral, infrequently sub- terminal, alternate, scorpioid; peduncle di- or trichotomously compound, infrequently very obscurely so, bearing several to numerous relatively showy, rose or purplish, rarely greenish flowers. Calyx 5-parted, the lobes equal to subequal, imbri- cated, cleft nearly to the receptacle, bearing within solitary, opposite, more or less erose or lacerate squamellae. Corolla salverform to infundibuliform, the tube not appendiculate nor annulate within, the limb aetinomorphie, 5-parted, dextrorsely convolute. Stamens 5, included, the anthers connivent and agglutinated to the stigma, consisting of 2 parallel, basally protuberant sporangia borne ventrally near the apex of an enlarged, narrowly sagittate, peltate connective; pollen gran- ular. Carpels 2, united at the apex by a common stylar shaft surmounted by the fusiform-subeapitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, separate or somewhat concrescent at the base. Follicles 2, apocarpous, terete, dehiscing along the ventral suture, contain- ing many dry, rostrate, apically comose seeds. Type species: T'emnadenia violacea (Vell.) Miers, Apoc. So. Am. 208. 1878 KEY TO THE SPECIES a. Corolla salverform, 2.5—4.0 em. long; inflorescence dichotomous, usually re- atedly so, rarely trichotomous. b. Corolla bright yellow, minutely and densely ferruginous- ipa wena without; plants of Colombia..........oooooooonooomono.. . T. stenantha bb. Corolla cream suffused with maroon or rose in the throat, ids with- o species of southern Brazil. e. Plants puberulent or hispidulous to gäre leaves ovate- be in- orescence relatively ee dichot EE stellaris cc. Plants glabrous; leaves oblong-lanceolate ; diuo e and irregularly di- or trichotomons, relativo o 3. T. ornata aa. Corolla infundibuliform, 5—6 em. long, crimson-purple to rose; m obscurely EK E to essentially TOMO re e's s . T. violacea (425) [Vor. 23 254 ANNALS OF THE MISSOURI BOTANICAL GARDEN 1. Temnadenia stenantha Woodson, Ann. Mo. Bot. Gard. 21: 613. 1934. Stems relatively stout, minutely ferruginous-tomentulose when young, eventually becoming glabrate; leaves opposite, petiolate, oblong-elliptic, apex shortly acuminate, base broadly obtuse, 11-13 em. long, 4.5-5.0 em. broad, firmly membrana- ceous to subcoriaceous, glabrous above, beneath minutely and inconspicuously puberulent toward the base and otherwise glabrous; petioles 1.7-1.9 em. long, minutely ferruginous- puberulent; inflorescence di- or trichotomous, bearing 20-25 rather mediocre, bright yellow flowers, minutely and irregu- larly ferruginous-tomentulose throughout, conspicuously sur- passing the subtending leaves; pedicels 1.0-1.25 em. long; bracts minutely lanceolate, 0.1-0.3 em. long; calyx-lobes ovate- lanceolate, acute to acuminate, 0.35-0.53 em. long, slightly foli- aceous, minutely ferruginous-puberulent to -papillate without, the squamellae obscurely bifid; corolla salverform, minutely ferruginous-tomentulose without, the tube 2.5-2.7 em. long, about 0.2 em. in diameter at the base, slightly dilated toward the orifice, the lobes obliquely obovate-oblong, 1.0-1.1 em. long, ascending or slightly spreading; stamens inserted somewhat below midway within the corolla-tube, the anthers elliptic- sagittate, 0.7-0.75 cm. long, densely villosulous-barbate dor- sally; ovary ovoid-oblongoid, about 0.15 em. long, glabrous; stigma 0.15 cm. long; nectaries separate, slightly surpassing the ovary; follicles unknown. COLOMBIA: BOYACA: on edge of high forest, region of Mt. Chapon, alt. 7000 ft., June 17, 1932, Lawrance 241 (NY, TYPE, MBG, photograph and analytical drawings) 2. Temnadenia stellaris (Lindl.) Miers, Apoc. So. Am. 210. 1878. Echites stellaris Lindl. Bot. Reg. 20: pl. 1664. 1835; A. DC. in DC. Prodr. 8: 457. 1844. Echites varia Stadelm. Flora 24!: Beibl. 17. 1841; A. DC. loc. cit. 455. 1844; Muell.-Arg. in Mart. Fl. Bras. 6t: 157. pl. 47. 1860. Echites Franciscea A. DC. acc. to Lindl. Bot. Reg. n.s. 10: pl. 24. 1847, not A. DC. loc. cit. 452. 1844. (426) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 255 Echites Franciscea A. DC. var. pallidiflora Hook. f. Bot. Mag. 76: pl. 4547. 1850. Echites varia Stadelm. a. purpurea Muell.-Arg. loc. cit. 158. 1860. Echites varia Stadelm. b. rosea Muell.-Arg. loc. cit. 1860. Echites varia Stadelm. c. sulphurea Muell.-Arg. loc. cit. 1860. Temnadenia bicrura Miers, loc. cit. 208. 1878. Temnadema pallidiflora (Hook.) Miers, loc. cit. 211. 1878. Temnadema Franciscea (Lindl.) Miers, loc. cit. 212. 1878, not as to A. DC. loc. cit. 452. 1844. Stems relatively slender, hispidulous to puberulent, eventu- ally glabrate; leaves opposite, petiolate, ovate-elliptie, apex acuminate, base obtuse to rounded, 6-15 em. long, 3-7 em. broad, firmly membranaceous, above minutely bullate-strigillose to glabrate, beneath minutely puberulent; petioles 0.5—0.8 cm. long, puberulent; inflorescence dichotomous, usually equalling or somewhat surpassing the subtending leaves, softly puberu- lent throughout, bearing 10-30 congested, cream-colored flowers suffused with maroon in the tube and orifice; pedicels 0.8-1.0 em. long, somewhat accrescent in fruit; bracts 0.2-1.0 em. long, scarious to more or less foliaceous below; calyx-lobes ovate- lanceolate, acuminate, 0.4-0.6 em. long, puberulent-papillate without, the squamellae deltoid, truncate, minutely lacerate; corolla salverform, glabrous without, the tube 1.5-2.0 em. long, about 0.3 em. in diameter at the base, sharply constricted at the insertion of the stamens, the lobes obliquely obovate, shortly acuminate, 1.0-1.4 cm. long, reflexed or spreading widely; stamens inserted about midway within the corolla-tube, the anthers lanceolate-sagittate, 0.6-0.65 cm. long, sparsely puber- ulent dorsally; ovary ovoid, about 0.2 cm. long, glabrous; stigma 0.15 em. long; nectaries usually somewhat concrescent at the base, equalling or somewhat surpassing the ovary; fol- licles relatively stout, continuous, falcate, 12-16 cm. long, gla- brous ; seeds about 1.5 cm. long, the pale tawny coma about 0.2 em. long. BRAZIL: PERNAMBUCO: data incomplete, Gardner 1060 (Camb.) ; MINAS GERAES: Pico d’Itabira, 1843, Claussen 35 (MP, NY); RIO DE JANEIRO: Mage, date lacking, (427 [Vor. 23 256 ANNALS OF THE MISSOURI BOTANICAL GARDEN Miers 4020 (BM, US); são PAULO: Santos, April 1, 1875, Mosen 3020 (S); Cubatão, alt. 0-50 m., March 2, 1929, Smith 2037 (G, MBG); PARANA: Tacarehy, in fruti- cetis, Febr. 11, 1915, Dusén 16684 (MBG, S); Paranagua, ad marginem silvulae, March, 1914, Jonsson 2a (8) ; Rio Cubatáo, ad marginem silvae fluminalis, Dec. 28, 1911, Dusén 13696 (S) ; Porto Dom Pedro II, in insula in fruticetis, Febr. 25, 1911, Dusén 11449 (MBG, S); Tacarehy, in silva primaeva ad marg. regionis lit., March 18, 1914, Jonsson 91a (S, US). Several color forms are known of this species, which has been cultivated in Europe as a greenhouse or ‘‘stove’’ orna- mental since prior to 1835 when it was illustrated and diseussed in Edwards’ Botanical Register 20: pl. 1664. The various color forms were interpreted subspecifically by Mueller, and later raised to specific rank by Miers. The species having been neglected horticulturally in recent years, it has been impossible to obtain first-hand knowledge of the color variation; and since herbarium specimens retain little by which they may be recog- nized, it has been considered best to retain the integrity of the species. 3. Temnadenia ornata (Hoehne) Woodson, Ann. Mo. Bot. Gard. 19: 383. 1932. Echites ornata Hoehne, Comm. Linh. Telegr. Estrat. Matto Grosso, Annexo 5, Bot. 6: 82. pls. 120; 131, fig. 1, 1915. Stems relatively slender, glabrous; leaves opposite, shortly petiolate, oblong-lanceolate, apex abruptly aeuminate, base obtuse, 8-12 cm. long, 3.5-4.5 em. broad, glabrous, somewhat nitidulous above; petioles 0.4—0.5 em. long; inflorescence re- peatedly and rather irregularly di- or trichotomous, relatively lax, somewhat surpassing the subtending leaves, bearing 15-25 yellowish, erimson-flushed flowers, glabrous throughout; pedi- cels 1.0-1.2 em. long; bracts lanceolate, 0.3-0.5 em. long, some- what foliaceous; calyx-lobes ovate-lanceolate, acuminate, 0.3 em. long, reflexed at the tips, slightly foliaceous, the squamellae deeply lacerate; corolla salverform, glabrous without, the tube 1.0-1.5 em long, about 0.125 cm. in diameter at the base, some- what dilating toward the orifice, the lobes obliquely obovate, 2.0-2.5 em. long, acuminate, widely spreading or somewhat re- flexed ; stamens inserted about midway within the corolla-tube, the anthers 0.5 em. long; carpels ovoid, about 0.125 em. long, (428) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 257 glabrous; stigma 0.1 em. long; nectaries separate, about half equalling the carpels; follicles unknown. BRAZIL: MATTO GROSSO: Procedencia Piruena, May, 1909, Hoehne 1965 (B, 180- TYPE, MBG, photograph and analytical drawings). 4. Temnadenia violacea (Vell. Miers, Apoc. So. Am. 208. 1878. Echites violacea Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 31. 1827; A. DC. in DC. Prodr. 8: 459. 1844; Stadelm. Flora 241: Beibl. 34. 1841; Muell.-Arg. in Mart. Fl. Bras. 6': 158. pl. 50, fig. 3. 1860. Echites Maximilianea Stadelm. loc. cit. 43. 1841. Echites Franciscea A. DC. loc. cit. 452. 1844. Stems relatively stout, densely hispidulous to glabrate; leaves opposite, petiolate, ovate-elliptic, acuminate, base obtuse to rounded, 7-10 em. long, 3-6 em. broad, firmly membrana- ceous, above densely and minutely bullate-hispidulous to gla- brate, beneath densely puberulent; petioles 0.3-0.5 cm. long, puberulent; inflorescence somewhat surpassing the subtending leaves, softly puberulent without, bearing 4-9 congested, crimson-purple to deep rose-colored flowers; peduncle ob- scurely dichotomous to essentially simple; pedicels 1.0-1.2 em. long, puberulent-papillate; bracts lanceolate, 0.5-0.8 em. long, subfoliaceous ; ealyx-lobes ovate-lanceolate, acuminate, 0.5-0.7 em. long, irregularly puberulent without, the squamellae deeply lacerate; corolla infundibuliform, glabrous without, the proper-tube 1.5-1.7 em. long, about 0.3 em. in diameter at the base, slightly constricting toward the insertion of the stamens, the throat conical, 1.5-2.0 cm. long, about 0.8-1.0 cm. in di- ameter at the orifice, the lobes broadly and obliquely obovate, 1.5-2.0 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers 1 em. long, minutely pu- berulent-papillate dorsally; ovary ovoid-oblongoid, about 0.4 em. long; stigma 0.2 cm. long; nectaries essentially separate, about half equalling the ovary; follicles unknown. BRAZIL: CEARA: Campo Grande, in carrascal, March 17, 1910, Löfgren 282 (8); MINAS GERAES: ad Lagóa Santa, Jangada, in silva, date lacking, Warming s.n. (C); data incomplete, Aug.-April, 1840, Claussen s.n. (Camb.) ; RIO DE JANEIRO: ad Rio de Janeiro, June, 1866, Engel s.n. (C); Sio PAULO: Santa Rita do Passa Quatro, Nov. 1, 1897, Hemnendorff 36 (S) ; PARANA: Itarare, ad marg. silvulae, March 19, (429) [Vor. 23 258 ANNALS OF THE MISSOURI BOTANICAL GARDEN 1915, Dusén 16835 (S); Itarare, opp. Monengava, in campo cerrado, alt. 740 m., Jan. 26, 1915, Dusén 16608 (MBG); inter Senges et Fabio Rego, ad marg. silvulae, alt. 770 m., Dec. 11, 1910, Dusén 11038 (MBG, S, US). EXCLUDED SPECIES Temnadenia annularis (L.f.) Miers, Apoc. So. Am. 216. 1878 (Echites annularis L.f. Suppl. 166. 1781) = Prestonia annularis (L.f.) G. Don, Gen. Hist. 4: 84. 1838. Temnadenia cordata (A. DC.) Miers, Apoc. So. Am. 212. 1878 (Echites cordata A. DC. in DC. Prodr. 8: 451. 1844) = Mandevilla oaxacana (A. DC.) Hemsl. Biol. Centr.-Am. Bot. 2: 316. 1882 (Echites Oaxacana A. DC. loc. cit. 451. 1844). Temnadenia corrugulata Miers, Apoc. So. Am. 215. 1878 = Prestonia solanifolia (Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 23: 289. 1936. Temnadenia glaucescens (Mart. € Gal.) Miers, Apoc. So. Am. 214. 1878 (Echites glaucescens Mart. & Gal. Acad. Roy. Brux. 111: 358. 1844) = Mandevilla oaxacana (A. DC.) Hemsl. Biol. Centr.-Am. Bot. 2: 316. 1882 (Echites Oawacana A. DC. in DC. Prodr. 8: 451. 1844). Temnadenia lasiocarpa Miers, Apoc. So. Am. 210. 1878 (Echites lasiocarpa A. DC. in DC. Prodr. 8: 463. 1844) = Mandevilla lasiocarpa (A. DC.) Malme, Bihang till K. Sv. Vet. Akad. Handl. Afd. III. 241°: 25. 1899. Temnadenia leptoloba (Stadelm.) Miers, Apoc. So. Am. 211. 1878 (Echites leptoloba Stadelm. Flora 24': Beibl. 15. 1841) = Prestonia agglutinata (Jacq.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931 (Echites agglutinata Jacq. Enum. Pl. Carib. 13. 1760). Temnadema Lobbiana (A. DC.) Miers, Apoc. So. Am. 209. 1878 (Echites lasiocarpa A. DC. y Lobbiana A. DC. in DC. Prodr. 8: 464. 1844) = Mandevilla lasiocarpa (A. DC.) Malme, Bihang till K. Sv. Vet. Akad. Handl. Afd. III. 241°: 25, 1899 (Echites lasiocarpa A. DC. loc. cit. 463. 1844). Temnadenia palustris (Muell.-Arg.) Miers, Apoc. So. Am. 213. 1878 (Amblyanthera palustris Muell.-Arg. in Mart. Fl. Bras. 6': 145. 1860) = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 171. 1895 (Echites hir- (430) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 259 suta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792, not R. & P.). Temnadenia parviflora (Benth.) Miers, Apoc. So. Am. 215. 1878 (Haemadictyon parviflorum Benth. Pl. Hartw. 355. 1857) — Prestonia parviflora Benth. in Benth. & Hook. Gen. Pl. 2: 709. 1873. Temnadenia quinquangularis (Jacq.) Miers, Apoc. So. Am. 217. 1878 (Echites quinquangularis Jacq. Enum. Pl. Carib. 13. 1760) = Prestonia quinquangularis (Jacq.) Spreng. Syst. 1: 637. 1825. Temnadenia Riedelii (Muell.-Arg.) Miers, Apoc. So. Am. 216. 1878 (Haemadictyon Riedelii Muell.-Arg. in Mart. Fl. Bras. 6*: 170. 1860) = Prestonia Riedelii (Muell.-Arg.) Mgf. in Fedde, Rep. Spec. Nov. 20: 26. 1924. Temnadenia secundiflora (A. DC.) Miers, Apoc. So. Am. 211. 1878 (Echites secundiflora A. DC. in DC. Prodr. 8: 457. 1844) = Mandevilla subsagittata (R. & P.) Woodson, Ann. Mo. Bot. Gard. 19: 69. 1932 (Echites subsagittata R. & P. Fl. Peruv. D ee We AE Temnadenia semidigyna (Berg) Miers, Apoc. So. Am. 213. 1878 (Echites semidigyna Berg, Abh. Zeeuwsch Gen. Wetens. 3: 583. 1773). This appears to be an unidentifiable species of Tabernaemontana. Temnadenia solanifolia (Muell.-Arg.) Miers, Apoc. So. Am. 214. 1878 (Haemadictyon (?) solamifolium Muell.-Arg. in Mart. Fl. Bras. 6': 171. pl. 49. 1860) = Prestonia solanifolia (Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 23: 282. 1936. Temnadenia tenuicula Miers, Apoc. So. Am. 216. 1878 = Prestonia solanifolia (Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 23: 282. 1936 (Haemadictyon (?) solamfolium Muell.- Arg. in Mart. Fl. Bras. 6t: 171. pl. 49. 1860). Temnadenia tomentosa (Vahl) Miers, Apoc. So. Am. 213. 1878 (Echites tomentosa Vahl, Symb. Bot. 3: 44. 1794) = Mandevilla hirsuta (A. Rich.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4%: 171. 1895 (Echites hirsuta A. Rich. Actes Soc. Hist. Nat. Paris 1: 107. 1792, not R. & P.). Temnadenia xanthostoma (Stadelm.)Miers, Apoc. So. Am. (431) [Vor. 23 260 ANNALS OF THE MISSOURI BOTANICAL GARDEN 212. 1878 (Echites xanthostoma Stadelm. Flora 24*: Beibl. 55. 1841) - Mandevilla coccinea (Hook. & Arn.) Woodson, Ann. Mo. Bot. Gard. 20: 734. 1933 (Echites coccinea Hook. & Arn. in Hook. Jour. Bot. 1: 286. 1834). XXII. FEeRNALDIA Woodson Fernaldia Woodson, Ann. Mo. Bot. Gard. 19: 48. 1932. Lactescent, fruticose or suffruticose lianas. Stems volubile, terete; branches alternate, or opposite below. Leaves op- posite, petiolate, entire, penninerved, eglandular; petioles subtended by several minute, pectinate, adaxial stipular ap- pendages. Inflorescence lateral, alternate, simply scorpioid, bearing several rather showy, white flowers. Calyx 5-parted, the lobes equal to subequal, cleft nearly to the receptable, im- bricated, scarious or only very slightly foliaceous, bearing solitary opposite squamellae within at the base. Corolla in- fundibuliform, slightly gibbous to essentially straight, the tube exappendiculate within, the limb aetinomorphie, 5-parted, dex- trorsely convolute, more or less arachnoid-villous within. Stamens 5, included, the anthers connivent and agglutinated to the stigma, consisting of 2 parallel, basally protuberant sporangia borne ventrally near the apex of an enlarged ob- tusely 2-auriculate, peltate connective; pollen granular. Car- pels 2, united at the apex by a slender stylar shaft surmounted by the fusiform-capitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries concrescent, unequally 4-lobed. Follicles 2, apocarpous, terete, acuminate, dehiscing along the ventral suture, containing many dry, truncate, apically comose seeds. Type species: Fernaldia E (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 48. 1932. KEY TO THE SPECIES a. Corolla glabrous without, or essentially so, or the lobes occasionally ciliate without, the proper-tube 2.0—2.2 em. long, about 0.15 em. in diameter at the base, somewhat gibbous at the insertion of the stamens, the throat oadly eampanulate-conieal, 0.9-1.2 em. long.............. 1. F. pandurata bro aa. Corolla generally pilosulose without, the proper-tube 1.8-2.0 em. long, about em. in diameter at the base, scarcely gibbous, the Sa rather rowly conical, 1.6-1.8 em. long................. LLL... 2 E e 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 261 1. Fernaldia pandurata (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 48. 1932. Echites pandurata A. DC. in DC. Prodr. 8: 458. 1844. Urechites Karwinskii Muell.-Arg. Linnaea 30: 440. 1860; Miers, Apoc. So. Am. 125. 1878. Amblyanthera ? pandurata (A. DC.) Muell.-Arg. loc. cit. 448. 1860. Angadenia pandurata (A. DC.) Miers, loc. cit. 182. 1878. Mandevilla velutina K. Sch. in Engl. & Prantl, Nat. Pflan- zenfam. 4?: 171. 1895, not Woodson. Mandevillea potosina T. S. Brandg. Univ. Calif. Publ. Bot. 4: 216. 1912. Echites pinguifolia Standl. Field Mus. Publ. Bot. 8: 35. 1930. Plate 5. Stems relatively slender, densely puberulent to essentially glabrate ; leaves opposite, oblong-elliptic to broadly ovate, apex shortly acuminate, the lower usually broadly cordate, the upper merely obtuse to subtruncate at the base, 4-13 em. long, 1.5-8.0 em. broad, membranaceous, either surface densely puberulent to velutinous, or the upper essentially glabrate ; petioles 1-2 em. long, minutely puberulent to essentially glabrate ; inflorescence lateral, simply scorpioid, bearing 6-18 rather showy, white to pale greenish-yellow flowers, the peduncle usually about half equalling the subtending leaves, minutely puberulent ; pedicels 0.4—0.6 cm. long, minutely puberulent, congested toward the upper half of the peduncle; bracts ovate, acute to acuminate, scarious or only slightly foliaceous, 0.1-0.2 em. long; calyx- lobes ovate, acute to obtuse, 0.2-0.3 em. long, minutely puberu- lent to puberulent-papillate without, the squamellae very mi- nutely crenulate or erose; corolla infundibuliform, essentially glabrous without, or the lobes more or less ciliate, the proper- tube 2.0-2.2 em. long, about 0.15 em. in diameter at the base, somewhat gibbous or arcuate at the insertion of the stamens, the throat broadly campanulate-conical, 0.9-1.2 em. long, 0.7— 0.9 em. in diameter at the orifice, the lobes obliquely obovate, acute to acuminate, 1.0-1.3 cm. long, densely arachnoid-villous (433) [Vor. 23 262 ANNALS OF THE MISSOURI BOTANICAL GARDEN at the base within, the margins occasionally ciliate, patulous; stamens inserted at the base of the corolla-throat, the anthers broadly elliptic-sagittate, shortly and rather obtusely auricu- late, 0.6-0.65 em. long, glabrous; ovary ovoid-oblongoid, about 0.2 em. long, glabrous; stigma 0.2 em. long; nectaries unequally 4-lobed, only slightly fleshy, about half equalling the ovary; mature follicles unknown. MEXICO: TAMAULIPAS: El Rosario, vicinity of Marmolejo, alt. 2100 ft., Aug. 10, 1930, Bartlett 10873 (US); SAN LUIS POTOSI: Rascon, Aug., 1911, Purpus 5408 (MBG, NY); VERA CRUZ: Baños del Carrizal, Aug., 1912, Purpus 6232 (MBG, US); GUERRERO: Acapulco and vicinity, Oct., 1894—March, 1895, Palmer 259 (G, US); oaxaca: Cafetal Las Pitas, alt. 400 m., Oct. 10, 1917, Reko 3511 (US); YUCATAN: Izamal, 1895, Gawmer 815 (MBG). EL SALVADOR: San Salvador, alt. 657 m., date lacking, Laboratorio Quimico 2 (MBG, US); cultivated, vicinity of San Salvador, alt. 650—850 m., March 30-April 24, 1922, Standley 23542 (NY, US). Also reported from Costa Rica. The lowermost leaves are said to be more or less pandurate in outline. The popular name of the species is reported as ‘‘loroco,’’ or ‘‘floroco,’’ and the flowers are said by Standley to be used as a flavoring for rice. Notes on the label of the specimen from the chemical laboratory of the argicultural department of El Salvador state: “Flores en la estacion lluviosa. Flores comestibles despues de coci- miento. Muy apreciadas por su olor y sabor en la comida. Abunda en los mercudos.”’ 2. Fernaldia brachypharynx Woodson, Ann. Mo. Bot. Gard. 19: 380. 1932. Stems relatively slender, minutely puberulent to glabrate when fully mature; leaves opposite, petiolate, broadly ovate, shortly and abruptly aeuminate, base broadly rounded, 7—10 em. long, 5-7 em. broad, membranaceous, above densely his- pidulous to very minutely puberulent-papillate, beneath velu- tinous to rather minutely and sparsely puberulent; petioles 0.9-2.0 em. long, pilosulose; inflorescence lateral, simply scor- pioid, the peduncle much shorter than the subtending leaves, minutely puberulent; pedicels 0.4—0.5 em. long, minutely puber- ulent, congested toward the upper half of the peduncle; bracts ovate, 0.1-0.2 em. long; calyx-lobes ovate, acuminate, 0.2-0.3 (434) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 263 em. long, minutely and sparsely pilosulose without, the squa- mellae deltoid, minutely crenulate; corolla infundibuliform, generally pilosulose without, the proper-tube 1.8-2.0 cm. long, about 0.3 em. in diameter at the base, the throat rather narrowly conical, 1.6-1.8 em. long, about 0.7 em. in diameter at the ori- fice, the lobes obliquely obovate, acuminate, 1.2-1.4 cm. long, arachnoid-villous at the base within, patulous; stamens in- serted at the base of the corolla-throat, the anthers elliptic- sagittate, shortly and rather obtusely auriculate, 0.6 cm. long, glabrous; ovary ovoid-oblongoid, about 0.15 cm. long, glabrous; stigma 0.2 cm. long; nectary irregularly 4-lobed, only slightly fleshy, about half equalling the ovary; follicles unknown. GUATEMALA: ESCUINTLA: along the road from Escuintla to the port of San Jose de Guatemala, Aug. 23, 1860, Hayes s.n. (G, TYPE, MBG, photograph and analytical drawings) ; Villa Nueva, Sept., 1914, Tejada 249 (US). XXIII. AskeTANTHERA Woodson Asketanthera Woodson, Ann. Mo. Bot. Gard. 19: 46. 1932. Echites of authors, in part, not P. Br. Lactescent, suffruticose or suffrutescent lianas. Stems volu- bile, terete; branches mostly opposite below, becoming alter- nate above. Leaves opposite, petiolate, entire, penninerved, eglandular; petioles subtended by several minute, pectinate, adaxial, stipular appendages. Inflorescence lateral, alternate, simply scorpioid, bearing several showy, pedicellate flowers subtended by conspicuously foliaceous bracts. Calyx 5-parted, the lobes subequal, cleft nearly to the receptacle, imbricated, strikingly foliaceous, bearing solitary, opposite squamellae within at the base. Corolla salverform, the tube straight, ex- appendieulate within, the limb actinomorphie, 5-parted, dex- trorsely convolute. Stamens 5, included, the anthers connivent and agglutinated to the stigma, consisting of 2 parallel, basally protuberant sporangia borne ventrally near the apex of an en- larged, sagittate, narrowly 2-aurieulate connective; pollen granular. Carpels 2, united at the apex by a slender stylar shaft surmounted by the fusiform-capitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. (435) [Vor. 23 264 ANNALS OF THE MISSOURI BOTANICAL GARDEN Nectaries 5, separate or somewhat concrescent at the base. Fol- licles 2, apocarpous, terete, dehiscing along the ventral suture, containing many dry, rostrate, apically comose seeds. Type species: Asketanthera calycosa (A. Rich.) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932 KEY TO THE SPECIES M re inserted about midway within the corolla-tube, the anthers barely ER Of: CUDA o ee E 1. A. calycosa aa, IE inserted near the base of the corolla-tube, the anthers deeply in- cluded; species of Hispaniola. b. Corolla 2-5 em. Se inflorescence 8-20-flowered. e. Corolla 2.0-2.5 em. long, the tube somewhat shorter than bai ealyx- lobes, aie glabrous. without sa ies opie a's . A. Picardae ec. Corolla 4-5 em. long, the tube conspicuously surpassing the aranea hispidulous without 22 jane EE 3. A. dolichopetala bb. Corolla 13-16 em. long; inflorescence 2-8-flowered......... 4. A. Ekmaniana 1. Asketanthera calycosa (A. Rich.) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932. Echites calycosa A. Rich. in Sagra, Hist. Cuba 11: 94. 1850. Rhodocalyz calycosus (A. Rich.) Miers, Apoc. So. Am. 140. 1878. Echites Rugeliana Urb. Symb. Ant. 5: 465. 1908. Plate 6. Stems relatively stout, densely ferruginous-hispidulous to glabrate. Leaves opposite, petiolate, broadly ovate-elliptic, apex acuminate, base obtuse to rounded, 8-15 em. long, 5-9 em. broad, membranaceous, above minutely strigillose to glabrate, beneath densely puberulent-pilosulose; petioles 0.7-1.0 em. long, densely puberulent-pilosulose ; inflorescence corymbose, bearing 5-12 congested, pale greenish-white or yellowish flowers ; peduncle somewhat shorter than the subtending leaves, hispidulous; pedicels 1.0-1.5 em. long, densely hispidulous; bracts lanceolate, conspicuously foliaceous, 1.2-2.0 em. long; calyx-lobes elliptic-lanceolate, acuminate, 1-2 cm. long, con- spicuously foliaceous, rather sparsely pilose, the squamellae eltoid, truncate, essentially entire; corolla salverform, rather sparsely pilosulose without, the tube 1.0-1.3 em. long, about 0.2 (436) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 265 cm. in diameter at the base, the lobes obliquely elliptic-oblance- olate, acute, 3.0-4.5 em. long, patulous; stamens inserted about midway within the corolla-tube, the anthers narrowly lance- olate-sagittate, 0.8 em. long, glabrous; ovary ovoid, about 0.3 em. long, glabrous; stigma 0.2 em. long; nectaries about half equalling the ovary ; follicles relatively stout, continuous, acu- minate, 25-30 em. long, densely ferruginous-hispid; seeds 1.6-1.9 em. long, the pale tawny coma 3.0—3.5 em. long. CUBA: PINAR DEL RIO: Sierra de los Organos, grupo del Rosario, valley of Rio Santa Cruz, March 31, 1923, Ekman 16390 (S) ; HABANA: Sierra Anafe, in the part of the Sierra called ‘‘Loma Esperon,’’ in one of the deep clefts, July 11, 1921, Ekman 13037 (S); ORIENTE: Sierra de Nipe, in marracales at Rio Barigua, Sept. 30, 1922, Ekman 15311 (S); Baracoa, in vall. flumin. Rio Macaguanigua, Jan. 19, 1915, Ekman 4340 (S) ; Sierra Maestra, El Peru, rocky guleh of Arroyo del Peru, Aug. 8, 1922, Ekman 16409 (B, S); Sierra de Nipe, ** El Taller," ad Rio Piloto, Febr. 18, 1918, Ekman 9051 (S) ; Bayamon, alt. 500 m., Febr., 1889, Eggers 4707 (B, US); data incomplete, Wright 1377 (B, Bx, K, MBG); MATANZAS [?]: ad fluv San Juan, date lacking, Rugel 397 (B). Echites Rugeliana Urb. differs from E. calycosa A. Rich. only in the somewhat larger calyx and corolla, a character which is rendered insignificant by great variability. Extreme specimens of either are found on the same herbarium sheet of Wright 1377 in the herbarium of the Missouri Botanical Garden. 2. - 7 Picardae (Urb.) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932. Echites SC Urb. Symb. Ant. 5: 466. 1908. Stems relatively slender, minutely ferruginous-puberulent to glabrate; leaves opposite, petiolate, ovate-elliptic, apex acumi- nate, base obtuse, 8-15 cm. long, 3-9 cm. broad, delicately mem- branaceous, above minutely strigillose to glabrate, beneath very minutely puberulent to glabrate; petioles 0.7—1.0 cm. long, minutely puberulent; inflorescence corymbose, bearing 8-20 greenish-yellow flowers; peduncle about half equalling the sub- tending leaves, minutely ferruginous-puberulent; pedicels 0.5-0.8 cm. long, minutely ferruginous-puberulent; bracts elliptic-lanceolate, acute to acuminate, 1.5-2.0 cm. long, con- spicuously foliaceous ; corolla salverform, essentially glabrous without, the tube 1.2-1.5 em. long, about 0.1 cm. in diameter at (437) [Vor. 23 266 ANNALS OF THE MISSOURI BOTANICAL GARDEN the base, the lobes obliquely obovate-oblong, acute, 0.8-1.0 cm. long, patulous; stamens inserted near the base of the corolla- tube, the anthers narrowly lanceolate, sagittate, 0.7 cm. long, glabrous; ovary oblongoid, about 0.2 em. long, glabrous; nec- taries about half equalling the ovary; follicles continuous, acu- minate, 30-45 em. long, glabrous; seeds 1.7 em. long, the pale tawny coma 4 cm. long. HISPANIOLA: HAITI: in montibus prope Pétionville, alt. 400 m., July 17, 1891, Picarda 804 (B); Pte. ph des Nippes, hillside at Bellevue, alt. 500 m., July 16, 7, Ekman 8590 (B, S, US); Presqu'ile de Nord- aem. ee de Terre- neuve, Gros Morne, EE precipices, alt. 400 m., Oct. 925, Ekman 5053 (B, S); Massif de la Hotte, eastern group, Acquin, Ia Colline, in Morne des Abricots, alt. 500 m., Nov. 9, 1926, Ekman 7191 (B); Massif ké ER Hotte, eastern group, e limestone cliffs south of Étang-Miragoane, July 28, 1926, Ekman 6536 B, 8); Presqu'ile du Nord-ouest, Les Gonaives, between Bassin and Mémé, alt. 350 m., Dec. 24, 1927, Ekman 9460 (B, S) ; Basse Vallée de Siburon, Sept. 2, 1910, Christ 2279 (B); prope Bel Endroit, in fruticosis densis, Aug. 9, 1917, SE 659 (B, 8) 3. Asketanthera dolichopetala (Urb.) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932. Echites dolichopetala Urb. Symb. Ant. 7: 335. 1912. Stems relatively stout, densely and minutely ferruginous- hispidulous to glabrate; leaves opposite, petiolate, ovate- elliptie, apex abruptly acuminate, base rather abruptly obtuse, 6—10 em. long, 4—6 em. broad, above densely strigillose-hispidu- lous, beneath densely and minutely pilosulose; petioles 1.5-2.0 em. long, densely and minutely hispidulous; inflorescence corymbose, bearing 8-12 congested greenish cream-colored flowers; peduncle about equalling or somewhat surpassing the subtending leaves, minutely and densely ferruginous-puberu- lent; pedicels 0.5-0.8 cm. long, minutely ferruginous-puberu- lent; bracts broadly lanceolate, 1.0-1.5 em. long, conspicuously foliaceous; calyx-lobes broadly lanceolate, acuminate, 1.2—1.7 em. long, conspicuously foliaceous, pilosulose, the squamellae deltoid, deeply lacerate; corolla salverform, densely fer- ruginous-hispidulous without, the tube 1.5-2.2 em. long, about 0.2 em. in diameter at the base, the lobes obliquely obovate- lanceolate, acute, 2.5-2.7 cm. long, patulous; stamens inserted near the paee of the corolla-tube, the anthers broadly elliptic- (438) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 267 sagittate, 0.7 em. long, glabrous; ovary ovoid, about 0.2 cm. long; nectaries somewhat shorter than the ovary; follicles unknown. HISPANIOLA: SANTO DOMINGO: Barahona, in via ad El Marviel, alt. 100 m., July, 1910, Fuertes 453 (B, BM, K, MBG, S); data incomplete, Martin s.n. (DL). A species offering a striking similarity in superficial aspect to A. calycosa of Cuba, but differing in the deeper insertion of the stamens, the harsher indument, and the deeply lacerate squamellae. 4. Asketanthera Ekmaniana Woodson, nom. no Echites longiflora Ekm. & Helwig, Arkiv f. Bot. 29400: 45. 1929, not Desf. Asketanthera longiflora (Ekm. & Helwig) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932. Stems relatively slender, very minutely puberulent to gla- brate; leaves opposite, petiolate, ovate to broadly elliptic- lanceolate, apex acuminate, base obtuse, 9-14 cm. long, 4-8 cm. broad, delicately membranaceous, above essentially glabrous, beneath very minutely puberulent-papillate to glabrate; pet- ioles 0.5-0.8 cm. long, minutely puberulent-papillate; inflores- cence corymbose, bearing 2-8 very showy, greenish-white or cream-colored flowers; peduncle about half equalling the sub- tending leaves, glabrate; pedicels 1.0-1.2 cm. long, essentially glabrous; bracts broadly elliptic-lanceolate, 1.5-2.0 em. long, conspicuously foliaceous; calyx-lobes broadly elliptic-lanceo- late, acuminate, 2.0-2.5 em. long, conspicuously foliaceous; corolla salverform, glabrous without, the tube 7-8 cm. long, about 0.2 em. in diameter at the base, the lobes obliquely elliptic- laneeolate, 6-8 em. long, patulous; stamens inserted near the base of the corolla-tube, the anthers narrowly oblong-sagittate, 0.8—1.0 em. long, glabrous; ovary oblongoid, about 0.3 cm. long, glabrous; nectaries somewhat concrescent at the base, about one-third equalling the ovary; mature follicles unknown. HISPANIOLA: HAITI: eastern La Hotte, Chapelle Mont Carmel, in a guleh, Nov. 8, 1924, Ekman 2426 (S); Massif de la Hotte, eastern group, Pte. Gonave, near Chap. St.-Michel, deep limestone hill, date lacking, Ekman 6598 (B, 8); SANTO (439) [Vor. 23 268 ANNALS OF THE MISSOURI BOTANICAL GARDEN DOMINGO: Cordillera de Barahona, Sierra de los Comisarios, between Banane and Gros Figuier, limestone, alt. 500 m., Aug. 28, 1926, Ekman 6762 (B, 8 A most remarkable plant, which seems to be a fairly free bloomer, and certainly appears to be a worthy subject for culti- vation in greenhouses. XXIV. MacropHarynx Rusby Macropharynx Rusby, Mem. N. Y. Bot. Gard. 7: 327. pl. 6. 1927. Lactescent, fruticose lianas. Stems volubile, terete; branches opposite. Leaves opposite, petiolate, entire, penninerved, eglandular; petioles subtended by several minute, adaxial, pectinate, stipular appendages. Inflorescence lateral, opposite or infrequently alternate, an extremely condensed, subumbel- late cincinnus, conspicuously bracteate, bearing few to several showy, greenish-white flowers. Calyx 7-9-parted (or 5-parted ?), the lobes somewhat unequal, more or less foliaceous, imbri- cated, cleft nearly to the receptacle, bearing within at the base a solitary, opposite squamella. Corolla infundibuliform (or salverform ?), the tube straight, exappendiculate within, the limb aetinomorphie, 5-parted, dextrorsely convolute. Sta- mens, 5, included, the anthers connivent and agglutinated to the stigma, consisting of 2 parallel, basally protuberant spor- angia borne ventrally near the apex of an enlarged, narrowly sagittate, acutely 2-auriculate connective; pollen granular. Carpels 2, united at the apex by a slender stylar shaft sur- mounted by the fusiform stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, separate or somewhat concrescent at the base. Follieles 2, apocarpous, terete, dehiscing along the ventral suture, containing many dry, comose seeds. Type species: Macropharynx spectabilis (Stadelm.) Wood- son, Ann. Mo. Bot. Gard. 18: 552. 1931. KEY TO THE SPECIES a. Corolla infundibuliform to subsalverform; calyx-lobes 7-9, immediatel subtended: by, similar bracts... see ets cree ieee eee TUM: DS aa. Corolla salverform; calyx-lobes 5, not immediately subtended by bracts. e ca Zeg Té SE 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 269 1. Macropharynx spectabilis (Stadelm.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931. Echites spectabilis Stadelm. Flora 241: Beibl. 44. 1841; A. DC. in DC. Prodr. 8: 462. 1844; Muell.-Arg. in Mart. Fl. Bras. 6': 153. 1860. Elytropus spectabilis (Stadelm.) Miers, Apoc. So. Am. 116. 1878. Macropharynx fistulosa Rusby, Mem. N. Y. Bot. Gard. 7: 329. pl. 6. 1927. Stems stout, densely and minutely ferruginous-tomentulose when young, eventually becoming glabrate; leaves opposite, petiolate, broadly ovate, apex very shortly and abruptly acumi- nate, base rounded to broadly and rather obscurely cordate, 9-30 em. long, 5-17 em. broad, firmly membranaceous to subcoriaceous, above minutely ferruginous-puberulent when young, glabrate and rather lustrous when fully mature, beneath minutely ferruginous-pilosulose to glabrate; petioles 3-5 cm. long, indument as upon the stem; inflorescence lateral, usually opposite, subumbellate, bearing 1-5 showy, greenish-white flowers; peduncle extremely condensed, about half equalling the subtending petioles, minutely ferruginous-puberulent ; bracts linear, 0.4—1.2 em. long, subfoliaceous, usually ferrugi- nous-pilosulose ; calyx-lobes 7-9, linear, long-acuminate, 0.9—1.3 cm. long, ferruginous-puberulent or -pilosulose without, sub- foliaceous, the squamellae minutely lacerate; corolla infundi- buliform or subsalverform, glabrous without, or essentially so, the proper-tube 0.7-1.0 em. long, about 0.2-0.3 em. in diameter at the base, the throat subtubular, 1.3-4.0 em. long, about 0.4—0.8 em. in diameter at the orifice, the lobes obliquely ob- ovate, 1.2-1.7 em. long, patulous; stamens inserted at the base of the corolla-throat, the anthers narrowly sagittate, 0.9—1.0 em. long, glabrous; ovary ovoid-oblongoid, about 0.3 em. long, minutely puberulent-papillate to essentially glabrous; stigma 0.2 em. long; nectaries equalling to somewhat surpassing the ovary; follicles relatively stout, 22-25 em. long, ferruginous- hispidulous when young, eventually glabrate; seeds unknown. BRAZIL: PARA: Belem, ad marginem silvae prope laeum Catu, Jan. 1, 1926, Ducke 21586 (B, US); AMAZONAS: in silvis ad Manacuru, ditionis Japurensis, 1920, (441) [Vor. 23 270 ANNALS OF THE MISSOURI BOTANICAL GARDEN Martius s.n. (B, TYPE); schlingend in den Hügelwaldehen von Porongo bei Santa Cruz, Jan. 1911, Herzog 1512 (B) PERU: LORETO: Stromgebiet des Maranon von Jquitos aufwürts bis zur Santi- ago-Mündung am Pongo de Manseriche, June 23, 1924, Tessmann 3607 (B) ; Iquitos, woods, alt. about 100 m., Aug. 3-11, 1929, Killip ds Smith 27493 (MBG, NY, US); Mishuyacu, near Io. forest, alt. 100 m., Oct.-Nov., 1929, Klug 541 (NY, US). BOLIVIA: BENI: Rurrenabaque, alt. 1000 ft, Dec. E Cardenas 1894 (K, NY, US); LA PAz: San Carlos bei discs Ag. Se 600 m., Dec. 21, 1926, Buchtien 1743 ae ; SANTA CRUZ: Rio Surutu, Prov. Sara, Dee. 97, 1924, SR 6813 (B, K, MBG); Banados del Piray, Buenaventura, alt. 450 m. , May, 1915, Lillo 1314 (B); Pamparegion de Buenavista, Prov. del Zara, alt. 400 m., Dec. 23, 1916, Lillo 3200 (B). Steinbach reports ‘‘No hace randales, porque ramifiea poco." It is a great temptation to segregate the specimens cited above into two, or perhaps three species or varieties based upon discrepancies in the size of the corolla, and on the in- dument of the inflorescence. The flora of the region from which they come is so inadequately understood, however, that it is believed more practical to preserve the integrity of the species for the present. In the herbarium at Berlin-Dahlem there is a specimen, supposedly of this same species, bearing the data ““in silvis pr. Paramaribo," without date, collected by Kegel (174). The data has appeared too dubious to incorporate with the better substantiated range records in the Amazon valley. 2. Macropharynx [?] anomala Woodson, Ann. Mo. Bot. Gard. 21: 614. 1934. Stems relatively slender, minutely and densely ferruginous- tomentulose to glabrate; leaves opposite, petiolate, broadly ovate, apex shortly and narrowly subcaudate-acuminate, base broadly obtuse to rounded, frequently rather obscurely cor- date, 10-18 em. long, 5-9 em. broad, membranaceous, above rather irregularly ferruginous-puberulent to -papillate, be- neath densely and very minutely ferruginous-tomentulose; petioles 2.5-3.0 em. long, minutely ferruginous-tomentulose; inflorescence lateral, opposite, subumbellate-fasciculate, rela- tively few-flowered; peduncle minutely ferruginous-tomentu- lose, somewhat shorter than the subtending petioles; pedicels 0.15-0.2 cm. long, minutely ferruginous-tomentulose; bracts linear, 1.0-1.4 em. long, somewhat foliaceous, minutely pilosu- (442) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 271 lose to puberulent-papillate; calyx-lobes 5, narrowly elliptic- lanceolate, 1.7-1.8 em. long, somewhat foliaceous, minutely puberulent-papillate, the squamellae deltoid, minutely dentic- ulate; corolla salverform, minutely puberulent-papillate with- out, the tube (in well-formed buds) 1.4-1.6 cm. long, about 0.2 em. in diameter at the base, somewhat inflated at the insertion of the stamens thence gradually constricting toward the orifice, the lobes rather narrowly ovate-dolabriform, acuminate, 0.9— 1.0 em. long, patulous; stamens inserted near the base of the eorolla-tube, the anthers ineluded, narrowly sagittate, 0.7—0.72 em. long, minutely papillate to essentially glabrous dorsally; ovary oblongoid, about 0.15 em. long, glabrous; stigma 0.15 em. long; nectaries slightly surpassing the ovary; follicles unknown. ADOR: in silv. tropic. fl. Pilaton, Oct., 1902, Sodiro 107/16 (B, TYPE, MBG, uev and analytieal drawings). A diseussion of reasons for and against relegating this plant to Macropharynx will be found at the place of original publieation. XXV. TuHenarpia HBK. Thenardia HBK. Nov. Gen. 3: 209. 1819; A. DC. in DC. Prodr. 8: 425. 1844; Benth. & Hook. Gen. Pl. 2: 710. 1876; Miers, Apoc. So. Am. 242. 1878; K. Sch. in Engl. € Prantl, Nat. Pflanzenfam. 4?: 187. 1895, not Besse E Mociño. Laetescent, suffrutieose or suffrutescent lianas. Stems terete; branches alternate, or occasionally opposite, particu- larly below. Leaves opposite, petiolate, entire, penninerved, eglandular, the petioles subtended by a solitary, adaxial, stipu- lar appendage. Inflorescence a lateral, alternate, pedunculate, trichotomous, umbellate cyme, the ultimate branches ex- tremely compressed, inconspicuously bracteate. Calyx 5- parted, the lobes equal or essentially so, cleft nearly to the receptacle imbricated, subtended by solitary, opposite squa- mellae. Corolla rotate to shortly salverform, the tube straight, exappendiculate within, the limb actinomorphic, 5-parted, dex- trorsely convolute. Stamens 5, the anthers connivent and ag- (443) [Vor. 23 212 ANNALS OF THE MISSOURI BOTANICAL GARDEN glutinated to the stigma, consisting of 2 parallel sporangia borne ventrally near the apex of an enlarged, narrowly sagit- tate, peltate connective, widely exserted; filaments connate above; pollen granular. Carpels 2, united at the apex by a com- mon stylar shaft surmounted by the fusiform-capitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, essentially separate, or somewhat con- crescent at the base. Follicles 2, apocarpous, terete, dehiscing along the ventral suture, containing many dry, apically comose seeds (fide A. de Candolle). Type species: Thenardia floribunda HBK. Nov. Gen. 3: 210. pl. 240. 1819. KEY TO THE SPECIES a. Corolla relatively large and showy, the limb 2.0-2.8 em. in E the tube not sharply constricted at the insertion of the stan b. Eege shortly salverform, the tube 0.5—0.6 em. long, ob Ca? 0.7—0.8 em Ong" IE E EE TERR: EE bb. dee rotate, the tube 0.2—0.35 em. long, the lobes 1.0-1.3 em. nA i Boone Si Ee . Th. floribunda aa. Corolla relatively small, the limb 1.2—1.5 em. in diameter, the ARN sharply constrieted at the insertion of the stamens. b. Corolla-lobes obtuse; squamellae narrowly oblong-trigonal, entire; nec- taries about half equalling the ovary................. 3. Th. gonoloboides bb. Corolla-lobes aeute; squamellae broadly oblong-dentiform, minutely erose; nectaries eR M or slightly surpassing the ovary........... SOS OO EE 4. Th. Galeottiana 1. Thenardia tubulifera Woodson, Ann. Mo. Bot. Gard. 19: 381. 1932 Stems relatively slender, glabrous; leaves opposite, shortly petiolate, elliptic-lanceolate, apex subeaudate-acuminate, base obtuse, 4-9 em. long, 1.5-3.0 em. broad, membranaceous, gla- brous above, sparsely pilosulose along the midrib and veins be- neath; petioles 0.5-1.0 em. long, sparsely pilosulose to glabrate; inflorescence equalling or somewhat surpassing the leaves, multiflorous; primary peduncle 2.5-3.0 em. long, the secondary 0.5—0.8 em. long; pedicels 2.5-2.7 em. long, puberulent-papillate to essentially glabrous; bracts lanceolate, about 0.1—0.2 em. long; calyx-lobes ovate-lanceolate, acuminate, 0.2-0.3 em. long, essentially glabrous without, the squamellae deltiform, es- (444) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 273 sentially entire; corolla shortly salverform, glabrous without, the tube broadly cylindrical, 0.5-0.6 em. long, about 0.2 cm. in diameter at the base, about 0.3 em. in diameter at the orifice, slightly and gradually constricted somewhat above midway at the insertion of the stamens, the lobes obliquely obovate- orbicular, minutely apiculate, 0.7-0.8 cm. long, spreading; stamens barely exserted, the anthers 0.7—0.8 em. long, the fila- ments 0.35-0.4 em. long, minutely puberulent-papillate above; ovary ovoid, about 0.15 em. long, glabrous; nectaries somewhat shorter than the ovary, essentially separate; stigma 0.125 cm. long; follicles unknown. MEXICO: JALISCO: data incomplete, Diquet s.n. (NY, TYPE, US, MBG, photo- graph and analytical drawings) ; Tonila, alt. 1000 m., Sept., 1923, Reko 4823 (US). Reko reports his specimen as bearing pink, fragrant flowers, which evidently applies to the following species as well: 2. Thenardia floribunda HBK. Nov. Gen. 3: 210. pl. 240. 1819; A. DC. in DC. Prodr. 8: 425. 1844; Miers, Apoc. So. Am. 242. 1878; K. Sch. in Engl. € Prantl, Nat. Pflanzenfam. 4*: 188. 1895. Thenardia ? suaveolens Mart. & Gal. Bull. Acad. Roy. Brux. 11!: 359. 1844. Stems relatively slender, glabrous ; leaves opposite, or rarely ternate, petiolate, oblong-elliptic, apex acutely subcaudate- acuminate, base obtuse, 5.5-13.0 em. long, 2-5 em. broad, deli- eately membranaceous, wholly glabrous or merely sparsely pilosulose at the base of the midrib beneath ; petioles 0.8-1.5 cm. long, glabrous; inflorescence about equalling or somewhat sur- passing the subtending leaves, multiflorous; primary peduncle 2.5-8.0 em. long, essentially glabrous, the secondary 0.4—1.0 em. long, minutely and sparsely pilosulose ; pedicels 2.0—2.2 em. long, essentially glabrous; bracts lanceolate, 0.05—0.1 cm. long ; calyx-lobes ovate-lanceolate, acuminate, 0.25-0.3 cm. long, mi- nutely papillate without, the squamellae deltiform, minutely erose; corolla rotate, glabrous or very indefinitely papillate without, the tube 0.2-0.35 em. long, about 0.15 cm. in diameter at the base and 0.35-0.4 cm. in diameter at the orifice, barely (445) [Vor. 23 274 ANNALS OF THE MISSOURI BOTANICAL GARDEN constricted just below the insertion of the stamens, the lobes obliquely obovate-dolabriform, subcaudate-acuminate, 1.1-1.3 em. long, spreading; stamens widely exserted, the anthers 0.6-0.7 cm. long, the filaments 0.5-0.6 em. long, minutely puber- ulent-papillate above; ovary ovoid, about 0.1 em. long, gla- brous; nectaries essentially separate, somewhat shorter than the ovary; stigma 0.15 em. long; follicles unknown. MEXICO: MORELOS: barranca near Cuernavaca, Aug. 2, 1896, Pringle 7242 (US); bluffs of barranea near Cuernavaca, alt. 5000 ft., Aug. 10, 1898, Pringle 6966 (MBG, NY, US). 3. Thenardia gonoloboides Woodson, spec. nov. Suffrutices volubiles; ramulis teretibus tenuibus glabris; foliis oppositis breviter petiolatis lanceolatis apice anguste acuminatis basi obtusis tenuiter membranaceis glabris vel nervo medio subtus basi minutissime puberulis 4-8 em. longis 1.3-2.5 em. latis petiolo 0.5-1.0 em. longo glabro; inflorescentiis foliis multo brevioribus multifloris; peduneulis primariis 0.5— 1.2 em. longis minute puberulo-papillatis seeundariis vix mani- festis; pedicellis 0.7-0.8 em. longis minute puberulo-papillatis ; bracteis anguste lanceolatis 0.1-0.3 em. longis; calycis laciniis lineari-lanceolatis anguste acuminatis 0.5-0.7 em. longis squa- mellis anguste oblongo-trigonalibus integris; corollae rotatae extus glabrae vel indistinctissime papillatae tubo 0.3-0.35 cm. longo basi ca. 0.125 em. diametro metiente ostio ca. 0.25 em. diametro metiente paulo supra medium abrupte constricto ubique staminigero lobis oblique obovatis obtusis 0.6-0.65 em. longis patulis; staminum antheris valde exsertis 0.3-0.35 em. longis filamentis 0.4 em. longis indistinctissime papillatis; ovario ovoideo ea. 0.1 em. longo glabro; neetariis haud con- crescentibus ovario ca. dimidio brevioribus; stigmate 0.1 cm. longo; folliculis ignotis. Stems relatively slender, glabrous; leaves opposite, shortly petiolate, lanceolate, apex narrowly acuminate, base obtuse, 4-8 em. long, 1.3-2.5 em. broad, delicately membranaceous, gla- brous, or very minutely puberulent at the base of the midrib be- neath; petioles 0.5-1.0 em. long, glabrous; inflorescence much shorter than the subtending leaves, multiflorous; primary (446) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 275 peduncle 0.5-1.2 em. long, minutely puberulent-papillate, the secondary scarcely manifest; pedicels 0.7-0.8 cm. long, mi- nutely puberulent-papillate; bracts narrowly lanceolate, 0.1-0.3 em. long; calyx-lobes linear-lanceolate, narrowly acuminate, 0.5-0.7 em. long, the squamellae narrowly oblong-trigonal, en- tire; eorolla rotate, glabrous or very indistinetly papillate without, the tube 0.3-0.35 cm. long, base about 0.125 cm. in diameter, orifice about 0.25 em. in diameter, abruptly con- stricted slightly above midway at the insertion of the stamens, the lobes obliquely obovate, obtuse, 0.6-0.65 em. long, spread- ing; stamens conspicuously exserted, the anthers 0.3-0.35 cm. long, the filaments 0.4 em. long, indistinctly papillate; ovary ovoid, 0.1 em. long, glabrous; nectaries essentially separate, about half equalling the ovary; stigma 0.1 cm. long; follicles unknown. MEXICO: OAXACA: canyons of mountains near Oaxaca, alt. 6500 ft., Aug. 20, 1894, Pringle 4822 (MBG, TYPE, NY, US). Closely related to the following species, from which it may be distinguished, in addition to the key characters, by the some- what larger corolla in general, with tube constricted slightly above the middle, and by the larger anthers with filaments very indistinetly papillate. 4. Thenardia Galeottiana Baill. Bull. Soc. Linn. Paris 2: 819. 1890. Stems relatively slender, glabrous; leaves opposite, peti- olate, elliptic-lanceolate, apex acuminate, base obtuse, 3-9 em. long, 0.6-2.5 em. broad, membranaceous, glabrous above, be- neath minutely puberulent in the lower axils of the midrib to glabrate; petioles 0.4—0.9 cm. long, glabrous; inflorescence much shorter than the subtending leaves, multiflorous; primary pe- dunele 0.4-0.6 em. long, minutely puberulent-papillate, the secondary scarcely manifest; bracts lanceolate, 0.05-0.2 cm. long; ealyx-lobes 0.4—0.6 cm. long, narrowly acuminate, mi- nutely puberulent-papillate below, the squamellae oblong- dentiform, minutely erose; corolla rotate, glabrous or very in- distinctly papillate without, the tube 0.2-0.25 cm. long, base (447) [Vor. 23 216 ANNALS OF THE MISSOURI BOTANICAL GARDEN about 0.1 em. in diameter, orifice about 0.2 cm. in diameter, abruptly constricted slightly below midway and there stami- nigerous, the lobes obliquely ovate, acute, 0.5-0.55 cm. long, spreading; stamens widely exserted, the anthers 0.25 em. long, the filaments 0.2-0.25 cm. long, puberulent-papillate; ovary ovoid, about 0.07 cm. long, glabrous; nectaries equalling or slightly surpassing the ovary, essentially separate; stigma 0.1 em. long; follicles unknown. MEXICO: OAXACA: Talea, alt. 3000 ft., Aug., 1844, Galeotti 1565 (MBG, NY, US, ISOTYPES ) ; GUERRERO: Iguala Canyon, alt. 2500 ft., Sept. 21, 1905, Pringle 13585 (G, US). Galeotti reported the flowers of his specimens as “albidis,” which would appear to apply to the preceding species as well. However, unlike those of Th. gonoloboides, the tips of the corolla-lobes of Th. Galeottiana appear to be lined with pink upon the basis of herbarium specimens. EXCLUDED SPECIES Thenardia (?) corymbosa Benth. in Hook. Jour. Bot. 3: 246. 1841 = Forsteronia umbellata (Aubl.) Woodson, Ann. Mo. Bot. Gard. 22: 208. 1935 (Apocynum umbellatum Aubl. Hist. Pl. Gui. Fr. 1: 275; 3: pl. 108. 1775). Thenardia (?) laurifolia Benth. Hook. Jour. Bot. 3: 246. 1841 = Forsteronia laurifolia (Benth.) A. DC. in DC. Prodr. 8: 438. 1844. Thenardia umbellata (Aubl.) Spreng. Syst. 1: 636. 1825 (Apocynum umbellatum Aubl. Hist. Pl. Gui. Fr. 1: 275; 3: pl. 108. 1775) = Forsteronia umbellata (Aubl) Woodson, Ann. Mo. Bot. Gard. 22: 208. 1935. XXVI. PnEsTOoN1IA R. Br. Prestonia R. Br. Mem. Wern. Soc. 1: 69. 1811; A. DC. in DC. Prodr. 8: 428. 1844; Muell.-Arg. in Mart. Fl. Bras. 6': 161. 1860; Benth. & Hook. Gen. Pl. 2: 709. 1876; Miers, Apoc. So. Am. 143. 1878; K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 188. 1895. (448) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 277 Haemadictyon Lindl. Trans. Hort. Soc. 6: 70. 1826; A. DC. loc. cit. 425. 1844; Muell.-Arg. loc. cit. 165. 1860; Miers, loc. cit. 254. 1878. Rhaptocarpus Miers, loc. cit. 151. 1878. Temnadema Miers, loc. cit. 207. 1878, in part. Mitozus Miers, loc. cit. 217. 1878, in part. Exothostemon G. Don, according to Miers, loc. cit. 238. 1878, in part. Belandra S. F. Blake, Contr. Gray Herb. 52: 78. 1917. Echites of authors, in part, not P. Br. Laetescent, suffrutieose or suffrutescent lianas. Stems volubile, terete, the branches opposite, or opposite below be- coming alternate above. Leaves opposite, entire, penninerved, eglandular; nodes stipulate. Inflorescence lateral, rarely sub- terminal, bostrychoidally racemose, frequently corymbose, bracteate. Calyx 5-parted, the lobes essentially equal, cleft nearly to the receptacle, imbricated, bearing within a solitary, opposite squamella. Corolla salverform, rarely infundibuli- form, the tube straight or very rarely slightly gibbous, usually bearing 5 epistaminal appendages within, or exappendiculate, or the appendages replaced by vertical, callous ridges, the ori- fice bearing a callous, faucal annulus, the limb actinomorphic, 5-parted, dextrorsely convolute. Stamens 5, the anthers con- nivent and agglutinated to the stigma, consisting of 2 parallel sporangia borne ventrally near the apex of an enlarged, nar- rowly sagittate, peltate connective; pollen granular; filament short, subeylindrieal, variously pubescent. Carpels 2, united at the apex by a common stylar shaft surmounted by the fusi- form or subeapitate, occasionally maniculate or digitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, separate or completely concres- cent. Follicles 2, apocarpous or more or less agglutinated, dehiscing along the ventral suture, containing many dry, truncate or inconspicuously rostrate, apically comose seeds. Type species: Prestonia tomentosa R. Br. Mem. Wern. Soc. 1: 10. ISIE (449) [Vor. 23 278 ANNALS OF THE MISSOURI BOTANICAL GARDEN KEY TO THE SECTIONS A. Corolla exappendiculate within, the faucal annulus relatively tenuous and inconspicuous, not greatly thickened or tubereulate; anthers wholly in cluded Pan meen ene a EE ect. 1. COALITAE AA. oe EE within (exeept in certain species of $ Tomentosae), e fau s conspicuously thickened or tuberculate; anther-tips Haie. often barely so (ineluded in certain species of $ Tomentosae). B. gei lobes pu small and inconspicuous, only uc foliaceous, sually v less refiprod e EE Seet. 2. ACUTIFOLIAE BB. ps -lobes ES large and conspicuous, strikingly us or coriaceous, not reflexed. C. Corolla glabrous or merely papillate without.......... Sect. 3. ANNULARES CC. Corolla densely pubescent without, at least the lobes. .Sect. 4. TOMENTOSAE Sect. 1. coaurran Woodson. Relatively slender, glabrous (pubescent in P. solanifolia) lianas; leaves membranaceous ; inflorescence simple, typically bostrychoid-racemose; calyx- lobes relatively inconspicuous, slightly foliaceous; corolla glabrous without, exappendiculate within, the faucal annulus relatively tenuous and inconspicuous, not greatly thickened and tuberculate; anthers wholly included. Spp. 1-4. KEY TO THE SPECIES a. BA relatively elongate, longer than the subtending leaves; corolla- owly oblong-lanceolate; plants of the Antilles... .1. agglutinata aa. Piden corymbose, shorter than the subtending leaves; ERE lobes obovate; species of South America. b. Plants glabrous, or only the lee surface of the leaves inconspicuously puberulent-papillate. e. Calyx-lobes lanceolate to oblong-lanceolate, acuminate; corolla-tube not inflated at the EE 2. P. coalita ee. Calyx-lobes broadly obovate, rounded or obtuse; corolla-tube conspicu- ousliyantated+at ebe 3. P. Dusenii bb. Plants softly and densely ferruginous-puberulent, particularly the TT PLACE EU UR dd uU E 4. P. solanifolia 1. Prestonia Sse ERE (Jaeq.) Woodson, Ann. Mo. Bot. Gard. 18: 552 Echites E Jacq. Enum. Pl. Carib. 13. 1760. Echites adglutinata Jacq. Stirp. Amer. 31. pl. 23. 1763; A. DC. in DC. Prodr. 8: 448. 1844, not Burm., sphalm. Echites circinalis Sw. Prodr. 52. 1788; A. DC. loc. cit. 466. 1844. (450) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 279 Echites sanguinolenta Tussac, Fl. Ant. 95. pl. 11. 1808. Haemadictyon circinalis (Sw.) G. Don, Gen. Hist. 4: 83. 1838. Echites leptoloba Stadelm. Flora 24': Beibl.15. 1841. Haemadictyon nutans (Anders.) A. DC. D sanguinolenta (Tussac) A. DC. loc. cit. 426. 1844. Echites revoluta A. DC. loc. cit. 457. 1844. Echites circinalis Sw. B Thomasiana A. DC. loc. cit. 466. 1844. Temnadenia leptoloba (Stadelm.) Miers, Apoc. So. Am. 211. 1878. Anechites adglutinata (Jacq.) Miers, loc. cit. 236. 1878. Anechites circinalis (Sw.) Miers, loc. cit. 1878. Anechites Thomasiana (A. DC.) Miers, loc. cit. 237. 1878. Plants completely glabrous; stems terete, relatively slender ; leaves elliptic-oblong to broadly oval, apex shortly acuminate, base obtuse or rounded, 4-12 em. long, 2-8 em. broad, firmly membranaceous, opaque; petioles 1.2-2.5 em. long; stipular ap- pendages numerous, interpetiolar, minutely dentiform-flagelli- form; racemes relatively elongate and lax, bearing 10-20 pale, greenish-white flowers; peduncle more or less flexuous, usually somewhat longer than the subtending leaves; pedicels 0.1—0.2 em. long, somewhat accrescent after maturity; bracts ovate- lanceolate, scarious or somewhat subfoliaceous, about equalling the pedicels; calyx-lobes ovate-lanceolate, acuminate, essen- tially equal, 0.2—0.3 em. long, scarious or slightly subfoliaceous, the internal squamellae deeply laciniate; corolla salverform, glabrous without, the tube 0.7-0.8 em. long, about 0.1 em. in diameter at the base, exappendiculate within, the faucal an- nulus tenuous, relatively inconspicuous, the lobes obliquely oblong-lanceolate, acuminate, about 0.5 cm. long, reflexed; stamens inserted about midway within the corolla-tube, the anthers wholly included, narrowly lanceolate-sagittate, 0.5 cm. long, glabrous; ovary oblongoid, about 0.2 cm. long, abruptly produced into the style, glabrous; stigma 0.15 em. long; nec- taries narrowly compressed-oblongoid, separate, about half as long as the ovary; follicles unknown. (451) [Vor. 23 280 ANNALS OF THE MISSOURI BOTANICAL GARDEN HISPANIOLA: HAITI: eastern La Hotte, Chapelle Mont Carmel, on the descent to Rio Bras Gauche, Nov. 8, 1924, Ekman 2431 (S); Isle La Tortue, Basse-Terre, in coastal thickets, Oct. 28, 1925, Ekman 5131 (S) ; Dept. du Nord, Bayeux, on the road to Aux Borgnes, Nov. 25, 1924, Ekman 2664 (S) ; data incomplete, Swartz s.n. (S, MBG, photograph). Porto Rico: prope Cayey ad Pedro Avila, inter arbores Coffeae Arabicae, Sept. 22, 1895, Sintenis 2457 (S). It is possible that the specimen from Porto Rico may repre- sent a chance introduction. 2. Prestonia coalita (Vell.) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931. ? Echites sulphurea Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 26. 1827; A. DC. in DC. Prodr. 8: 468. 1844, Echites coalita Vell. Fl. Flum. 112. 1830; Icon. 3: pl. 40. 1827; A. DC. loc. cit. 458. 1844; Muell.-Arg. in Mart. Fl. Bras. 6!: 155. 1860. Echites Vauthieri A. DC. loc. cit. 457. 1844; Miers, Apoc. So. Am. 201. 1878. Echites Blanchetu A. DC. loc. cit. 448. 1844; Muell.-Arg. loc. cit. 157. 1860. Echites Martii Muell.-Arg. loc. cit. 155. 1860. Rhaptocarpus coalitus (Vell.) Miers, loc. cit. 152. 1878. Rhaptocarpus apiculatus Miers, loc. cit. 153. 1878. Rhaptocarpus Martii (Muell.-Arg.) Miers, loc. cit. 1878. Mitozus Blanchetii (A. DC.) Miers, loc. cit. 219. 1878. Mitozus concinnus Miers, loc. cit. 223. 1878. Stems terete, relatively slender, minutely and sparsely pu- berulent-papillate when young, soon becoming glabrate and conspicuously lenticellate; leaves oblong-lanceolate to rather broadly elliptic, apex acuminate, base obtuse to rounded, 5-15 em. long, 2-6 em. broad, firmly membranaceous, above glabrous, beneath puberulent-papillate to glabrate; petioles 0.5—1.0 em. long; stipular appendages intrapetiolar, 2-4, minutely denti- form-flagelliform; racemes simple, more or less corymbose, bearing 8-20 pale greenish-yellow flowers; peduncle 14-14 as long as the subtending leaves; pedicels 0.5-0.7 em. long, some- what accrescent after maturity, minutely puberulent-papillate (452) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 281 to glabrate; bracts ovate-lanceolate, 0.2-0.3 em. long, subfoli- aceous ; calyx-lobes lanceolate to oblong-lanceolate, acuminate, 0.3-0.5 cm. long, foliaceous or subfoliaceous, minutely papii- late to glabrate without, the internal squamellae deltoid, entire or somewhat erose ; corolla salverform, glabrous or indistinetly papillate without, the tube 1.0-1.7 em. long, about 0.15 cm. in diameter at the base, exappendiculate within, the faucal an- nulus tenuous, relatively inconspicuous, the lobes obliquely obovate, obtuse to shortly acuminate, 0.5-0.8 em. long, reflexed ; stamens inserted about midway within the corolla-tube, the anthers wholly included, narrowly lanceolate-sagittate, 0.5-0.6 em. long, glabrous; ovary ovoid, abruptly produced into the style, about 0.1 em. long, glabrous; stigma 0.15 cm. long; nec- taries ovoid, separate, somewhat shorter than the ovary ; fol- lieles relatively slender, conspicuously moniliform, usually united at the tips, 20-45 em. long, glabrous; seeds about 1 cm. long, the pale yellowish coma about 2.5 em. long. BRAZIL: CEARA: Crato, in ‘‘caapuerao,’’ March 20, 1910, Lofgren 620 (S); data incomplete, T1 1755 (V); BAHIA, Jacobina, weg apo Blanchet 3513 (V); MINAS GERAES: ELS , Wawra $ Maly 279 (V); Praesidide de St. J. Bapt., date cado e 5 (V); Caldas adt di Jan. 26, 1826, Widgren s.n. (S); Caldas, 1857, Regnell II 358 (S); Lagoa Santa, date lacking, Warming s.n. (C, S); RIO DE JANEIRO: peus date laeking, Mikan s.n. (V); cu il date lacking, Pohl 5397 (V); dece date lacking, Pohl 2452 (V); d in- complete: Widgren s.n. T e 168 Ger SAO PAULO: prope Penha in bein urbis S. Paulo, alt. 750 m., 1902, Wacke . (V); Mooca, buschwald, Febr. 23 1913, Brade 5696 (S); Serra de Coral Ge: 25, 1875, m 4269 (8); PARANA: in silvula ad flumen Rio Tibagy pr. Ponta Grossa, Jan. 7, 1904, Dusén s.n. (S); same locality, pis 17, 1909, Wem 8038 (S); Jaguariahyva, ad marginem silvae primaevae, alt. m., May 8, 1914, Jonsson 286a (S, G); Capáo Grande, in (8); MATTO bo Gg za d rage Sach d ` 1894, move 1472 (8); Santa Cruz da Barra, in ripa dumetosa fluvii Paraguay, Mareh 2 894, Lindmann 3163 (S); DATA INCOMPLETE: Widgren 1325 (S); Riedel s.n. E 3. Prestonia Dusenii (Malme) Woodson, Ann. Mo. Bot. Gard. 18: 552. 1931. Echites Dusenii Malme, Arkiv f. Bot. 22A?: 9. 1928. Stems terete, relatively slender, minutely papillate when (453) [Vor. 23 282 ANNALS OF THE MISSOURI BOTANICAL GARDEN young, glabrate and conspicuously lenticellate when fully ma- ture; leaves broadly elliptic-oblong, apex subcaudate-acumi- nate, base obtuse or rounded, 5-15 em. long, 2.5-6.0 em. broad, above minutely and sparsely papillate to glabrate, minutely puberulent along the midrib, beneath minutely puberulent- papillate, firmly membranaceous; petioles 1.0-1.3 em. long; stipular appendages interpetiolar, numerous, minutely denti- form; racemes corymbose, simple, bearing 5-12 greenish- yellow flowers; peduncle about as long as the subtending petioles; pedicels 0.7-0.8 em. long, somewhat accrescent after maturity, minutely puberulent-papillate; bracts ovate, 0.1-0.3 em. long, foliaceous; calyx-lobes obovate, minutely mucronu- late, somewhat unequal, 0.5-0.7 cm. long, foliaceous, minutely papillate without at the base, the internal squamellae deltoid- trigonal, minutely lacerate ; corolla salverform, minutely papil- late without, the tube 1.0-1.3 em. long, about 0.25 em. in di- ameter at the base, conspicuously narrowing toward the orifice, exappendiculate within, the faucal annulus tenuous, relatively inconspicuous, the lobes obovate, shortly acuminate, 0.7-0.8 em. long, reflexed; stamens inserted somewhat below midway within the corolla-tube, the anthers wholly included, narrowly sagittate, about 0.5 em. long, glabrous; ovary ovoid, 0.1 em. long, glabrous; stigma 0.1 em. long; nectaries compressed- ovoid, somewhat concrescent at the base, about half as long as the ovary; follicles relatively stout and rigid, united at the apex, 18-23 em. long, glabrous; seeds not seen. BRAZIL: PARANA: Alexandra, Serra da Prata, ad marg. silvae prim., March 5, 1911, Dusén 11486 (S, TYPE, MBG, photograph and analytical drawings); são PAULO: Santos, Jan. 20, 1875, Mosen 3433 (S). From slight intergradations observed, it is suspected that P. Dusenu and P. coalita may hybridize in the field. 4. Prestonia solanifolia (Muell.-Arg.) Woodson, comb. nov. Haemadictyon (?) solanifolium Muell.-Arg. in Mart. Fl. Bras. Di: 171. pl. 49. 1860. Temnadenia solanifolia (Muell.-Arg.) Miers, Apoc. So. Am. 214. 1878. (454) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 283 Temnadenia corrugulata Miers, loc. cit. 215. 1878. T emnadenia tenuicula Miers, loc. cit. 216. 1878. Stems terete, relatively stout, densely ferruginous-hirtellous, conspicuously lenticellate when fully mature; leaves elliptic to oval, apex acuminate, base obtuse or rounded, 7-13 em. long, 3.5-7.0 em. broad, firmly membranaceous, above minutely and relatively sparsely hirtellous-strigillose to glabrate, beneath densely ferruginous-hirtellous ; petioles 0.9-1.3 em. long; stip- ular appendages interpetiolar, numerous, minutely dentiform; racemes densely corymbose, bearing 15-40 bright yellow flowers; peduncle somewhat longer than the subtending peti- oles, minutely ferruginous-hirtellous ; pedicels 0.7—1.0 em. long, somewhat accrescent after maturity, minutely ferruginous- hirtellous; bracts lanceolate, 0.2-0.3 cm. long, subfoliaceous ; ealyx-lobes oblong-lanceolate, acuminate, 0.3—0.35 cm. long, subfoliaceous, rather sparsely and laxly ferruginous-pilosulose without, the internal squamellae deeply lacerate; corolla salverform, minutely papillate to glabrate without, the tube 0.8-1.0 em. long, about 0.2 em. in diameter at the base, ex- appendiculate within, the faucal annulus tenuous and relatively inconspicuous, the lobes obliquely obovate, shortly acuminate, 0.6-0.7 em. long, widely spreading ; stamens inserted somewhat below midway within the corolla-tube, the anthers wholly in- cluded, oblong-sagittate, 0.3-0.35 em. long, glabrous; ovary broadly ovoid, about 0.1 em. long, glabrous or very minutely papillate ; stigma fusiform-subcapitate, 0.05 em. long; nectaries compressed-ovoid, separate, nearly equalling the ovary; im- mature follicles relatively slender, continuous, united at the apex, densely ferruginous-hirtellous ; seeds unknown. BRAZIL: RIO DE JANEIRO (?): data incomplete, Schiich s.n. (V, TYPE, MBG, photo- graph and analytical drawings) ; MINAS GERAES: original forest, Matto Virgem, alt. 730 m., Nov. 21, 1930, Mexia 5337 (MBG); data incomplete, 1831, Ackermann s.n. (M, corvPE, MBG, photograph and analytical drawings); SAO PAULO: Campinas, Nov., 1894, Novaes 11202 (B). The collection from Sao Paulo appears to differ from the more northerly specimen in having somewhat smaller flowers, but is too fragmentary to warrant segregation. (455) [Vor. 23 284 ANNALS OF THE MISSOURI BOTANICAL GARDEN Sect. 2. AcuTIFOLIAE Woodson. Relatively slender, glabrous lianas; leaves membranaceous (coriaceous in P. marginata, pachyphylla, and simulans), usually veined with red or purple when young; inflorescence simple, typieally bostrychoid- racemose ; calyx-lobes relatively small and inconspicuous, only slightly foliaceous, usually more or less reflexed; corolla gla- brous without, appendiculate within, the faucal annulus con- spicuously thickened; anther-tips exserted (except in P. cyamphylla and P. Hassleri). Spp. 5-13. KEY TO THE SPECIES a. Anthers rather deeply included; calyx-lobes broadly oblong, not sharply reflexed. b. Leaves DE SE corolla-tube not inflated at the sah plants of Orther Bolivia OCCUPE siete sine TERES . P. cyaniphylla bb. Leaves SE BCS to suborbicular; corolla-tube po at the base; apecios-0f Paraguay o uos puoi Aa E E EE EE RU 6. P. Hassleri aa. Anther- ag barely exserted, or at least attaining the orifice of the corolla- tube; calyx-lobes lanceolate, the tips reflexed. b. Epistaminal appendages wholly included. e. Leaves membranaceous; anthers pubescent dorsally, or very rarely glabrate. d. Nectaries shorter than the ovary; follicles relatively stout, continu- ; plants of Martinique, Guadeloupe, and Trinidad (including Obapgo) EE 7. P. quinquangularis dd. Neetaries as long as the ovary or somewhat longer; follicles ge tively slender, slightly articulated; plants of Panama and Sou GELEET DP. io ec. Leaves coriaceous; anthers glabrous. es inate, not heavily coriaceous, the inii verrueose ADOYO in desiccation A O ue E . P. marginata dd. Leaves obtuse or rounded, frequently abruptly EE heavily ceous, the venation not verrucose above....... 10. P. pachyphylla eor bb. S appendages slightly exserted, or at least attaining the orifice of the corolla-tube. e. Leaves coriaceous or subcoriaceous. ooo... 11. P. simulans ec. Leaves rather delicately membranaceo d. Leaves broadly rhombic-ovate to ira Care append- TE ages slightly ae EECH . Lindmanüi dd. Leaves rather narrowly oblong-elliptie; epistaminal Ee at- taining the orifice of the corolla-tube................ 13. P. lagoensis 5. Prestonia cyaniphylla (Rusby) Woodson, comb. nov. Echites cyaniphylla Rusby, Bull. N. Y. Bot. Gard. 4: 409. 1907 (456) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 285 Stems relatively slender, minutely papillate when very young, soon becoming glabrate and conspicuously lenticellate when fully mature; leaves oblong-elliptic, apex acuminate, base obtuse to rounded, 6-8 em. long, 2.5-4.0 em. broad, membrana- ceous, above glabrous, or sparsely papillate at the base, opaque, beneath glabrous, subglaucous; petioles 1-2 em. long; stipular appendages intrapetiolar, numerous, narrowly and minutely dentiform; racemes subcorymbose, simple, bearing 12-20 greenish-yellow flowers toward the end of the peduncle; pe- duncle usually somewhat longer than the subtending leaves, glabrous; pedicels 0.8-1.0 em. long, somewhat accrescent after maturity, minutely papillate to glabrate; bracts narrowly ob- long-lanceolate, 0.1-0.25 em. long, subfoliaceous; calyx-lobes broadly oblong, acute to abruptly acuminate, 0.45-0.55 cm. long, foliaceous, ascending or slightly spreading, not sharply reflexed, glabrous without, the internal squamellae deltoid, slightly lacerate; corolla salverform, minutely papillate with- out, the proper-tube 1.2-1.5 em. long, about 0.3 em. in diameter at the base, slightly narrowing toward the middle, then again gradually expanding toward the orifice, epistaminal append- ages very minute, wholly included, the faucal annulus conspic- uously thickened, the lobes obliquely obovate, inconspicuously acuminate, 0.6-0.8 cm. long, reflexed or widely spreading; stamens inserted somewhat below midway within the corolla- tube, the anthers narrowly sagittate, about 0.5 cm. long, gla- brous, rather deeply included; ovary ovoid-oblongoid, about 0.2 em. long, glabrous; stigma 0.125 em. long; nectaries com- pressed-ovoid, separate or somewhat concrescent at the base, somewhat shorter than the ovary; follicles relatively long and slender, continuous, agglutinated and united at the tip, 30-45 em. long, glabrous; seeds about 0.4 em. long, the pale tawny coma about 2 cm. long. BOLIVIA: LA PAZ: Milluhuaya in Nord-Yungas, alt. 1200 m., Dec., 1917, Buchtien 4372 (G, B) ; exact locality and date lacking, Bang 2267 (MBG, NY, TYPE). 6. Prestonia Hassleri Woodson, spec. nov. Suffrutieosa volubilis omnino glabra; ramulis teretibus gracilibus maturitate conspicue lenticellatis; foliis rhomboi- (457) [Vor. 23 286 ANNALS OF THE MISSOURI BOTANICAL GARDEN deo-ovatis vel suborbicularibus 4-10 em. longis 3-7 em. latis membranaceis supra olivaceis opacis subtus subglaucescent- ibus; petiolis 0.8-2.0 em. longis; appendiculis stipulaceis intrapetiolaribus multis minute dentiformibus; inflorescentiis racemosis simplicibus plurifloris ; pedunculis folia aequantibus vel paulo superantibus; pedicellis 0.7-1.0 em. longis post maturitatem paulo accrescentibus; bracteis anguste ovatis 0.2-0.3 em. longis subfoliaceis; calycis laciniis late oblongis acutis vel abrupte acuminatis 0.4-0.5 em. longis foliaceis ascendentibus patulisve haud reflexis intus basi squamellam deltiformem subintegram gerentibus; corollae salverformis virido-flavae tubo proprio 1.0-1.5 em. longo basi ca. 0.35 em. diametro metiente prope apicem sensim angustato appendic- ulis epistaminalibus minutis profunde inclusis annulo faucale manifeste incrassato lobis oblique obovatis breviter acumi- natis 0.5-0.8 em. longis patentibus; antheris anguste sagittatis 0.5-0.6 em. longis glabris apice paulo exsertis; ovario ovoideo ca. 0.15 em. longo glabro; stigmate 0.15 em. longo; nectariis com- presse ovoideis integris ovario paulo brevioribus; folliculis juventate gracillimis paulo artieulatis glabris agglutinatis; seminibus ignotis. Plants completely glabrous; stems terete, relatively slender, conspieuously lenticellate when fully mature; leaves rhomboid- ovate or suborbicular, apex abruptly and shortly acuminate, base obtuse or rounded, 4-10 em. long, 3-7 em. broad, mem- branaceous, above olivaceous, opaque, beneath subglaucescent; petioles 0.8-2.0 em. long; stipular appendages intrapetiolar, numerous, minutely dentiform; inflorescence racemose, simple, bearing 20-40 greenish-yellow flowers; peduncle about as long as the subtending leaves or somewhat longer; pedicels 0.7—1.0 cm. long, somewhat accrescent after maturity; bracts narrowly ovate, 0.2-0.3 em. long, subfoliaceous; calyx-lobes broadly ob- long, acute or abruptly acuminate, 0.4—0.5 em. long, foliaceous, ascending or somewhat spreading, not reflexed, the internal squamellae deltoid, subentire; corolla salverform, glabrous without, the proper-tube 1.0-1.5 em. long, about 0.35 em. in diameter at the base, conspicuously narrowed toward the ori- (458) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 287 fice, epistaminal appendages minute, deeply included, the faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.5—0.8 em. long, reflexed or widely spreading; stamens inserted somewhat below midway within the corolla-tube, the anthers narrowly sagittate, 0.5—0.6 cm. long, the tips slightly exserted or at least attaining the orifice of the corolla-tube, glabrous; ovary ovoid, about 0.15 cm. long, glabrous; stigma 0.15 em. long; nectaries compressed-ovoid, separate, somewhat shorter than the ovary; immature follicles relatively long and slender, slightly articulated, agglutinated and united at the apex, glabrous; seeds unknown. PARAGUAY: in regione lacus Ypacaray, March, 1913, Hassler 12527 (B, C, MBG, TYPE); common in hedges, Villa-Rica, March 12, 1929, Jørgensen 4182 (MBG); in regione collium, Cordillera de Villa-Riea, Jan., 1905, Hassler 8765 (B); in sylva, San Bernardino, Dec., year lacking, Hassler 3583 (B, BB, V); Cordillera de Altos, Oct., 1902, Fiebrig 64a (B, DL, G, M). This species is very close to the preceding, of which it may eventually be proved a variety. 7. Prestonia quinquangularis (Jacq.) Spreng. Syst. 1: 637. 1825. Echites quinquangularis Jacq. Enum. Pl. Carib. 13. 1760; Stirp. Amer. 32. pl. 25. 1763; A. DC. in DC. Prodr. 8: 468. 1844. Echites nutans Anders. Trans. Soc. Arts Lond. 25: 203. 1807; Sims, Bot. Mag. 51: pl.2473. 1824. Haemadictyon venosum Lindl. Trans. Hort. Soc. 6: 70. 1826; Miers, Apoc. So. Am. 255. 1878. Haemadictyon nutans (Anders.) A. DC. loc. cit. 426. 1844. Temnadenia quinquangularis (Jacq.) Miers, loc. cit. 217. 1878. Stems relatively slender, glabrous, inconspicuously lenti- cellate when fully mature; leaves oblong- to ovate-elliptic, acute to shortly acuminate, base obtuse to rounded, 6-14 cm. long, 2-6 em. broad, membranaceous, conspicuously veined with red or purple when young, above glabrous, beneath minutely papillate to glabrous; petioles 1-2 em. long; stipular append- ages interpetiolar, numerous, minutely dentiform; inflores- [ Vor. 23 288 ANNALS OF THE MISSOURI BOTANICAL GARDEN cence racemose, simple, bearing 6-20 greenish-yellow flowers; peduncle about as long as the subtending leaves or somewhat shorter, glabrous; pedicels 0.7-1.5 em. long, somewhat accres- cent after maturity, glabrous; bracts ovate-lanceolate, 0.1—0.15 em. long, subfoliaceous; calyx-lobes ovate-lanceolate, acu- minate, 0.15-0.2 cm. long, subfoliaceous, sharply reflexed, glabrous without, the internal squamellae deltoid-ligulate, mi- nutely erose; corolla salverform, minutely papillate without, 1.2-1.5 em. long, about 0.3 cm. in diameter at the base, the epi- staminal appendages about 0.1 em. long, wholly ineluded, the faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate to obtuse, 0.7-1.0 em. long, sharply reflexed; stamens inserted slightly above midway within the corolla-tube, the anthers narrowly sagittate, 0.55-0.6 em. long, pubescent dorsally; ovary ovoid, about 0.2 em. long, glabrous; nectaries compressed-ovoid, separate or more or less concres- cent at the base, somewhat shorter than the ovary; follicles relatively stout, continuous, agglutinated and united at the apex, 20-35 em. long, glabrous; seeds about 1 em. long, the tawny coma 2.0-2.5 em. long. MARTINIQUE: Pelée, date lacking, Kunth s.n. (B). GUADELOUPE: prope Capesterre, 1892, Duss 2841 (B) ; route de Pigeon à la Ponte Noire, Dee, 28, 1895, Duss 3713 (B). TRINIDAD: Caparo, April 27, 1908, Broadway 2707 (B); St. Clair Experiment Station, wild on fences, May 22, 1907, Broadway 2582 (B, M) ; O'Meara Savannah, Arima, April 16, 1908, Broadway 2799 (B); Montpelier, Tobago, running over shrubs and grasses, Oct. 3, 1909, Broadway 3122 (B) ; Tobago ad Wellbeck, Nov. 23, 1912, Broadway 4350 (B); exact locality and date lacking, von Rohr 23 (C). The purple venation of the leaves is said by collectors to be replaced by white not infrequently. This species has also been reported locally from Venezuela and the Guianas. 8. Prestonia acutifolia (Benth.) K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 188. 1895. Haemadictyon acutifolium Benth. ex Muell.-Arg. in Mart. Fl. Bras. 61: 167. 1860; Miers, Apoc. So. Am. 260. 1878. Haemadictyon acutifolium Benth. B latifolium Muell.-Arg. loc. cit. 1860. (460) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 289 Haemadictyon calignosum Miers, loc. cit. 1878. Echites Bangii Rusby, Bull. N. Y. Bot. Gard. 4: 409. 1907. Echites Hulkiana Pulle, Rec. Trav. Bot. Néerl. 9: 160. 1912. Echites Laurentiae-disca Rusby, Descr. So. Am. Pl. 85. 1920. Prestonia acutifolia (Benth.) K. Sch. var. latissima Mgf. Notizblatt 9: 982. 1926. Stems relatively slender, minutely puberulent when young, glabrate and conspicuously lenticellate when fully mature; leaves elliptic to oblong- or ovate-elliptic, apex acuminate, base obtuse or rounded, firmly membranaceous, usually veined with red or purple when young, above dark green, opaque, glabrous, beneath somewhat paler, minutely and irregularly papillate to glabrous, 6-16 cm. long, 2-8 em. broad; petioles 0.6-2.0 cm. long; stipular appendages intrapetiolar, numerous, minutely dentiform ; inflorescence racemose, simple, bearing 6-40 green- ish-yellow flowers; peduncle usually somewhat shorter than the subtending leaves, occasionally somewhat longer, glabrous; pedicels 0.5-1.2 em. long, somewhat acerescent after maturity, glabrous or very minutely papillate; bracts ovate-lanceolate, 0.1-0.2 em. long, subfoliaceous; calyx-lobes narrowly ovate- lanceolate, acuminate, 0.15-0.2 em. long, sharply reflexed, sub- foliaceous, minutely papillate to glabrous without, the internal squamellae deltoid-liguliform, minutely lacerate or erose; corolla salverform, glabrous or very minutely papillate with- out, the tube 1.5-2.0 em. long, about 0.4 cm. in diameter at the base, epistaminal appendages 0.05-0.2 em. long, wholly in- cluded, the faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.7—1.0 em. long, sharply reflexed ; stamens inserted at about the upper 14 of the corolla- tube, the anthers rather broadly sagittate, 0.4—0.55 em. long, pubescent dorsally, rarely glabrate; ovary ovoid, about 0.2 em. long, glabrous; stigma 0.1 em. long; nectaries compressed- oblongoid, separate or more or less conerescent, somewhat sur- passing, or about equalling the ovary; follicles relatively long and slender, obscurely and rather distantly articulated, ag- (461) [Vor. 23 290 ANNALS OF THE MISSOURI BOTANICAL GARDEN glutinated and united at the apex, usually rather tortuous, 20— 40 em. long, glabrous, or very indistinctly papillate; seeds 0.1 em. long, the coma pale tawny, 3-4 em. long. PANAMA: CANAL ZONE: shore s. of PES Trail, Barro Colorado Island, Febr. 12, aia sig ces $ Vestal 498 (FN BIA: SANTANDER: alluvial m ait 80-90 m., Badillo, Jan. 16, 1918, Pen- SH PES (NY); Rio Surata Valley, between ERR and El Jaboncillo, alt. 800-1500 m., Jan. 2, 1927, Killip $ Smith 16305 (NY, US); upper Rio Lehrija V ley, northwest of Bucaramanga, alt. 400— 700 m., Dec. 29, 1926, Killip $ Smith 16308 (US, NY, B); MAGDALENA: open grass and thiekets, rare, near Bonda, alt. 250 m., Nov. 9, Cienaga, Sept. 10, 1898, H. H. Smith 1645 (NY); ANTIOQUIA: Pto. no near CHE June 20, 1928, Toro 1105 (NY). VENEZUELA: DISTRITO FEDERAL: Caracas, date lacking, collector unknown (8); DATA INCOMPLETE: May, 1896, Rusby $ Squires 302 (B, M, MBG, NY); AMAZONAS: ad fl. Casiquiari, 1853-54, Spruce 3430 (V DUTCH GUIANA: on river banks, fluv. Doct. Nov. 26, 1910, Hulk 303 (U). RAZIL: AMAZONAS: in vicinibus Barra, Prov. Rio Negro, Dee., 1850—March, 1851, Spruce 1002 (B, Camb., M, TYPE, V) ; Bóa Vista, Rio Branco super., in silvis ripariis, July, 1913, Kuhlmann 3650 (B); MATTO GROSSO: Cuyaba, July 2, 1903, Malme 3122 (S); in ‘‘cerrado’’ pr. oppidium Cuyaba, March 22, 1904, Malme 1478 ÉIS DER CR ad villam Cuyabensis, date lacking, Manso $ Lhotzky 37 (B); DATA INCOMPLETE: AUN (M); Glaziou 20412 (B). PERU: LoRETO: Puerto Arturo, lower Rio Huallaga below ERE dense forest, AR about 135 m., Aug. 24-25, 1929, Killip $ Smith 27840 (M G, US); Amaqueria, Stromgebiet do Ueayali von 10? S. bis zur Mündung, Ge 24, 1923, Tessmann 3368 (B, DL); Mündung des Apaga, Stromgebiet des Maranon von Jquitos aufwarts bis zur Santiago-Mündung am Pongo de Manseriche, ca 77? 30' West., alt. 145 m., Dee. 23, 1924, Tessmann 4816 (B); Mishuyaeu, near Iquitos, alt. 00 m., forest, b 1930, Klug 1119 (NY, US); HUANUCO: zwischen Monzon und den INS alt. 600-700 m., Aug. 28, 1903, Weberbauer 3605 (B, DL); JUNIN: thickets, La Merced, alt. SS 700 m., May 29-June 4, 1929, Killip $ Smith 23387 (MBG, US); in sunny montana along Se same locality, Aug. 10—24, 1923, Mac- bride 5245 (B, FM); SAN MARTIN: Alto Rio Huallaga, alt. 360-900 m., Dec., 1929, Williams 5550 (FM, MBG). BOLIVIA: SANTA CRUZ: bosques de Buenavista, alt. 450 m. , April 22, 1917, Stein- ach 3331 (B); quintas de la ciudad de Santa ise alt. 450 m., Febr. 20, 1917, Steinbach 3256 (B) ; BENI: Rurrenabaque, alt. 1 0 ft., Dee. 1, 1921, Cardenas 1748 (NY); DATA INCOMPLETE: Bang 2053 (B, MBG, ue PARAGUAY: in reg. cursus superioris fl, Apa, wo hte Hassler 8172 (B, BB, V); in reg. vieine San Estanislao , Jan., year lacking, Hassler 6001 (B, BB); Cen- turion, Jan. 5, 1909, Fiebrig 4541 (B); in ee S EE ““Sieera de Amambay,’’ March, 1908, Rojas 10270 ARGENTINA: FORMOSA: Villa, Febr., 1918, Jørgensen 3119 (B, BA, MBG). It is not improbable that an accumulation of additional specimens will necessitate the segregation of P. acutifolia as (462) 1936] WOODSON—-STUDIES IN THE APOCYNACEAE. IV 291 interpreted in this revision. At the present moment the num- ber of specimens appears too scanty, considering the vast geo- graphical distribution of the species, to merit division into subsidiary units. 9. Prestonia marginata (Benth.) Woodson, comb. nov., not Mef. Haemadictyon marginatum Benth. in Hook. Jour. Bot. 3: 250. 1841; A. DC. in DC. Prodr. 8: 426. 1844; Miers, A poc. So. Am. 256. 1878. Haemadictyon Cayennense A. DC. loc. cit. 427. 1844; Miers, loc. cit. 1878. Haemadictyon papillosum Muell.-Arg. Linnaea 30: 432. 1860. Stems terete, relatively slender, minutely and sparsely his- pidulous when young, glabrate, or seabridulous in the vicinity of the nodes, and rather inconspicuously lenticellate when fully mature; leaves oblong-elliptie, apex acuminate to subeaudate- acuminate, base acute to obtuse, 6-12 em. long, 2-5 em. broad, glabrous, coriaceous, but not very thickly so, either surface somewhat nitidulous, the venation more or less verrucose throughout; petioles 0.3-0.4 em. long; stipular appendages interpetiolar, 6-8, minutely dentiform; inflorescence racemose, simple, or rarely dichotomous, bearing 10-40 pinkish-yellow flowers; peduncle about equalling or somewhat surpassing the subtending leaves; pedicels 0.8-1.5 em. long, somewhat accres- cent after maturity, glabrous or very minutely and sparsely papillate; bracts ovate-lanceolate, 0.8—1.0 em. long, searious; calyx-lobes ovate-lanceolate, acuminate, 0.2-0.3 em. long, some- what reflexed or spreading, glabrous to very irregularly papil- late without, the internal squamellae deltoid, somewhat cleft or divided; corolla salverform, glabrous without, 1.3-1.7 cm. long, about 0.3 em. in diameter at the base, epistaminal append- ages 0.05—0.07 em. long, wholly included, the faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.8-1.0 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the (463) [Vor. 23 292 ANNALS OF THE MISSOURI BOTANICAL GARDEN anthers oblong-sagittate, 0.5 cm. long, glabrous; ovary ovoid, about 0.2 em. long, glabrous; stigma 0.1 em. long; nectaries compressed-ovoid, separate or somewhat concrescent at the base, somewhat shorter than the ovary; follicles relatively slender, continuous or very obscurely articulated, usually somewhat faleate and united at the apex, 11-16 em. long, gla- brous; seeds about 1.1 em. long, shortly rostrate, the pale yel- lowish coma about 2.5 em. long. BRITISH GUIANA: Pirara, date uncertain, Schomburgk 713 (K, TYPE); banks of the Potaro River, Tumatumari, July 4—6, 1921, Gleason 336 (NY, US); Malali, Demerara River, Do about 5° 35' N., Oct. 30-Nov. 5, 1922, Cruz 2631 (G). FRENCH GUIANA: Cayenne, date JAEN Vargas s.n. (DC). BRAZIL: PARA: overflow bank, Rio Thome Assu, Distrieto Acara, alt. 35 m., Aug. 1, 1931, Mexia 6029 (MBG 10. Prestonia pachyphylla Woodson, nom. nov. Prestonia marginata Mgf. Notizblatt 9: 88. 1924, not Haemadictyon marginatum Benth. (Prestonia margin- ata (Benth.) Woodson). Stems relatively stout, minutely and rather sparsely pilosu- lose when very young, glabrate, or minutely scabridulous in the vicinity of the nodes, and rather inconspicuously lenticel- late when fully mature; leaves broadly oval, apex obtuse to rounded, frequently abruptly mucronulate, base broadly ob- tuse to rounded, 6-12 em. long, 2-4 em. broad, glabrous, heavily coriaceous, either surface somewhat nitidulous, below some- what glaucescent, the venation not verrucose above; petioles 0.5-0.8 em. long; stipular appendages intrapetiolar, numerous, minutely dentiform; inflorescence racemose, simple, bearing 20-40 greenish-yellow flowers; peduncle glabrous to very in- conspicuously papillate, somewhat surpassing the subtending leaves; pedicels 0.7-1.0 em. long, somewhat accrescent after maturity, glabrous to very inconspicuously papillate; calyx- lobes ovate-lanceolate, acuminate, 0.3-0.4 em. long, somewhat reflexed or spreading, glabrous without, the internal squa- mellae somewhat incised or divided; corolla salverform, gla- brous or very inconspicuously papillate without, 1.2-1.3 em. long, about 0.3 em. in diameter at the base, epistaminal ap- pendages 0.2-0.25 em. long, wholly included, the faucal an- (464) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 293 nulus conspicuously thickened, the lobes obliquely obovate, rounded, 0.8-1.0 cm. long, reflexed; stamens inserted at about the upper 14 of the corolla-tube, the anthers oblong-sagittate, about 0.5 em. long, glabrous; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.1 em. long; nectaries compressed-oblongoid, somewhat surpassing the ovary, separate or somewhat con- crescent at the base; follicles unknown. RAZIL: AMAZONAS: Serra de Mairary, Surumu, Rio Branco, Nov., 1909, Ule $451 (B, TYPE, MBG, photograph and analytical drawings). 11. Prestonia simulans Woodson, spec. nov. Suffruticosa vel fruticosa volubilis; ramulis gracilibus gla- bris maturitate rimulosis inconspicueque lenticellatis; foliis oblongo-elliptieis apice obtusis minute mucronatis basi obtusis 8-10 cm. longis 3-4 em. latis coriaceis subcoriaceisve glaber- rimis supra plus minusve illustribus subtus opacis margine post exsiccationem leviter revolutis; petiolis 0.2-0.3 em. longis glabris; inflorescentiis lateralibus simplicibus foliis sub- aequantibus flores gilvos 8-10 laxe gerentibus; pedunculo glaberrimo parte tertia superiore florigero; pedicellis 1.3-1.5 cm. longis glabris; bracteis minute ovato-lanceolatis vix bene visis; calycis laciniis ovato-lanceolatis acuminatis apice plus minusve reflexis 0.4—0.45 cm. longis subfoliaceis glaberrimis intus basi squamellam deltoideo-ligulatam integram vel irreg- ulariter bifidem gerentibus; corollae salverformis tubo 0.9-1.0 em. longo basi ea. 0.125 cm. diametro metiente supra basi aliquid ampliato deinde ad insertionem staminum gradatim angustato faucibus ca. 0.15 em. diametro metientibus extus glabris intus prope insertionem staminum minute retrorseque barbatis appendicibus epistaminalibus linearibus ca. 0.2 em. longis paulo exsertis annulo faucium conspicue incrassato, lobis oblique obovato-dolabriformibus breviter acuminatis 1.0-1.2 em. longis patentibus; antheris lanceolato-sagittatis 0.5 em. longis glabris apice exsertis; ovario ovoideo ca. 0.2 cm. longo glabro; stigmate 0.1—0.125 cm. longo; nectariis compresse ovoideis haud concrescentibus ca. 0.15 em. longis; folliculis ignotis. (465) [Vor. 23 294 ANNALS OF THE MISSOURI BOTANICAL GARDEN Stems relatively slender, glabrous, rimulous and inconspicu- ously lentieellate at maturity; leaves opposite, shortly peti- olate, oblong-elliptic, apex obtuse, minutely mucronate, base obtuse, 8-10 em. long, 3-4 em. broad, coriaceous or subcori- aceous, above more or less lustrous, beneath opaque, the margin somewhat revolute in desiccation; petioles 0.2-0.3 em. long, gla- brous; inflorescence simple, relatively lax, about equalling the subtending leaves, bearing 8-10 yellowish flowers; peduncle glabrous, the upper third florigerous; pedicels 1.3-1.5 em. long, glabrous; bracts minutely ovate-lanceolate, very minute; calyx- lobes ovate—lanceolate, acuminate, the tips more or less re- flexed, 0.4—0.45 em. long, subfoliaceous, glabrous, the internal squamellae deltoid-ligular, entire or irregularly bifid; corolla salverform, the tube 0.9-1.0 em. long, about 0.125 em. in di- ameter at the base, somewhat inflated below midway, thence gradually eonstrieted toward the insertion of the stamens, about 0.15 em. in diameter at the orifice, glabrous without, re- trorsly barbate above the insertion of the stamens within, epi- staminal appendages linear, about 0.2 em. long, somewhat ex- serted, faucal annulus conspicuous, the lobes obliquely obovate- dolabriform, shortly acuminate, 1.0-1.2 cm. long, reflexed; anthers lanceolate-sagittate, 0.5 em. long, glabrous, the tips exserted; ovary ovoid, about 0.2 em. long, glabrous; stigma 0.1- 0.125 em. long; nectaries compressed-ovoid, not concrescent, about 0.15 em. long; follicles unknown. COLOMBIA: CUNDINAMARCA: Toeaima, Dec., 1932, Arbelaez 2140 (US, TYPE, MB3G, analytieal drawings). Closely approaching the habit of P. pachyphylla, from which it differs in the exsertion of the epistaminal appendages. 12. Prestonia Lindmanii (Malme) Hoehne, Comm. Linh. Telegr. Estrat. Matto Grosso, Annexo 5, Bot. 6: 88. 1915. Haemadictyon Lindmanii Malme, Bihang till K. Sv. Vet. Akad. Handi Afd. III. 24**: 31. pl. 3. fig. 10. 1899. Stems relatively slender, minutely puberulent-papillate when very young, soon becoming glabrate and rather incon- spicuously lenticellate; leaves broadly rhombic-ovate to sub- (466) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 295 orbicular, apex shortly and frequently very abruptly acumi- nate, base obtuse to rounded, 5-10 em. long, 3-7 em. broad, membranaceous, glabrous, or very minutely puberulent-papil- late when very young, either surface opaque, the lower somewhat paler; petioles 1-3 em. long; stipular appendages intrapetiolar, numerous, minutely dentiform-flagelliform; in- florescence racemose, simple, bearing 10-40 greenish-yellow flowers; peduncle glabrous, usually equalling or somewhat surpassing the subtending petioles; pedicels 1.0-1.25 em. long, somewhat accrescent after maturity, glabrous; bracts ovate to ovate-lanceolate, 0.15-0.2 em. long, subfoliaceous; calyx-lobes ovate-lanceolate, acute to acuminate, 0.25-0.35 em. long, sub- foliaceous, glabrous or very inconspicuously papillate without, the internal squamellae deltoid, somewhat incised or lacerate; corolla salverform, glabrous or very minutely papillate with- out, the tube 1.7-2.0 em. long, about 0.4 cm. in diameter at the base, epistaminal appendages 0.35-0.45 em. long, slightly ex- serted, the faucal annulus conspicuously thickened, the lobes obliquely obovate, inconspicuously acuminate, 0.8-1.0 cm. long, reflexed; stamens inserted at about the upper 14 of the corolla- tube, the anthers oblong-sagittate, 0.5 cm. long, glabrous, the tips slightly exserted; ovary ovoid, about 0.25 cm. long, gla- brous; stigma 0.15-0.2 em. long; nectaries compressed-ovoid, separate or somewhat concrescent at the base, about equalling the ovary; follicles relatively long and slender, continuous or essentially so, separate and parallel, or slightly agglutinated when immature, 30-35 em. long, glabrous; seeds 0.8-1.0 em. long, shortly rostrate, the pale tawny coma about 2.5 em. long. BRAZIL: RIO DE JANEIRO: exact locality and date lacking, Glaziou 12943, 14072 (B); MATTO GROSSO: Cuyaba, Febr., 1832, Manso $ Lhotzky 33 (B, DL); Corumba, April 7, 1903, Malme s.n. (S); in ripa dumetosa fluvii Paraguay, Santa Cruz da Barra, March 25, 1894, Lindmann A3161 (S, TYPE, MBG, photograph and analytical drawings) ; Toscania, Caceres, Jan., 1909, Hoehne 1200 (B). ARAGUAY: in reg. cursus superioris fl. Apa, Dec., 1901, Hassler 8065 (B, BB, V); exact locality and date lacking, Hassler 3061 (BB). The collections of Glaziou represented as from Rio de Janeiro are open to suspicion. The inaccuracy of Glaziou’s labels is notorious. (467) [Vor. 23 296 ANNALS OF THE MISSOURI BOTANICAL GARDEN 13. Prestonia lagoensis (Muell.-Arg.) Woodson, comb. nov. Haemadictyon Lagoense Muell.-Arg. in Warming, Viden- skab. Meddel. Nat. Foren. Kjoeb. 115. 1869. Haemadictyon Warmingii Muell.-Arg. loc. cit. 116. 1869. Plants completely glabrous; stems relatively slender, incon- spicuously lenticellate when fully mature; leaves rather nar- rowly oblong-elliptic, apex acuminate, base acute, 6.5-10.5 em. long, 2.3—4.2 em. broad, membranaceous, either surface opaque, the lower somewhat paler; petioles 0.7-1.7 em. long; stipular appendages intrapetiolar, numerous, minutely dentiform; in- florescence racemose, simple, bearing 10-30 greenish-yellow flowers; peduncle somewhat lax and flexuous, somewhat sur- passing the subtending leaves; pedicels 1.5-1.8 em. long; bracts lanceolate, 0.1-0.15 cm. long, subfoliaceous; calyx-lobes ovate- lanceolate, acuminate, 0.3-0.36 em. long, subfoliaceous, the internal squamellae deltoid-ligular, inconspicuously incised or divided; corolla salverform, glabrous without, the tube 1.3-1.5 cm. long, about 0.35 cm. in diameter at the base, the epistaminal appendages about 0.2 cm. long, attaining the orifice, the faucal annulus conspicuously thickened, the lobes obliquely obovate, acuminate, 0.7-0.8 cm. long, reflexed; stamens inserted at about the upper Y, of the corolla-tube, the anthers oblong-sagittate, 0.5 cm. long, glabrous, the tip barely exserted or at least at- taining the orifice of the corolla-tube; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.125 em. long; nectaries com- pressed-ovoid, about equalling or slightly surpassing the ovary, separate or somewhat concrescent at the base; follicles unknown. BRAZIL: MINAS GERAES: Lagoa Santa, 1870, Warming s.n. (C, TYPE, S, MBG, photograph and analytical drawings). It is quite possible that this is merely a variety of the preceding. Sect. 3. ANNULARES Woodson. Relatively stout (except in P. exserta, P. velutina, and P. parvifolia), glabrous or softly pubescent lianas; leaves membranaceous to coriaceous; inflo- rescence simple or variously compound, typically bostrychoid- (468) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 297 racemose to subumbellate ; calyx-lobes relatively large and con- spicuous, foliaceous, petalaceous, or coriaceous; corolla gla- brous or minutely papillate without, appendiculate within, the faucal annulus conspicuously thickened; anther-tips exserted or at least attaining the faucal annulus. Spp. 14-45. KEY TO THE SPECIES A. Neetaries thick and fleshy (see also 37—38), separate, or only united at the base; species of South America (including Trinidad and Tobago), and Panama. B. eigen lianas; ealyx-lobes foliaceous, uniformly green (see also 39). C. ta glabrous; leaves not cordate D. Epistaminal ndages surpassing the ae inserted at the base of the faucal annulus, or slightly below. E. Bracts Pid ineonspieuous, linear Sé linear-lanceolate, 0.1— 0 ong, only slightly foliaceous...............1 4. P. exserta EE. ES very conspicuous, LE to ind lanceolate, 0.7—1.0 em. long; manifestly EIER, eebe bce ee bass ela 15. P. velutina DD. Epistaminal appendages dix surpassed by the anthers, inserted well below the faueal E. Braets conspicuousl us. foliaceous, nearly as large as the calyx- lobes; leaves finely puberulent beneath; plants of ripe c e 99 069€ vv 6 9 v v b PEN M B EE xr ve RUATR 0 b» eX KEE WOEN P svi EE. Bracts relatively inconspicuous, only slightly cons much smaller than the cal eaves E ges or es- ntially so; plants of (southern ?) Brazil........ perplexa CO EEN Bee pubescent dorsally; leaves uid et e b i eordate, at least in part, oecasionally only somewhat trun- cavo or rounded" Bb Die DANG as raser EE EE T8: P. LT BB. Frutieose or suffruticose lianas; calyx-lobes petalaceous or coriaceou usually more or less suffused or tinted with purple, at least the nee so colored. C. cn appendages I. included. D. Inflorescence simple: A ria 19. P. didyma DD. Inflorescence com E ‘ner minutely pubescent dorsally. E. Calyx-lobes 0.7—1.0 e E: Se lobes e SCH as long as the tube; leaves acute oo A ` pr. FF. ane lobe about as long as the tube, or nearly so; leaves NEE, A AEE co en EE EE. ied -lobes 1.0—1.8 em. long. F. orescence diehotom p or 9 wianensis mous or rarely trichotomous; calyx green- somewhat suffused with pur G. Calyx-lobes about as long as the Soe delicately mem- branaceous. (469) [Vor. 23 298 ANNALS OF THE MISSOURI BOTANICAL GARDEN H. Calyx-lobes 1.7-1.8 em. long, somewhat suffused with purple; corolla ‘‘raisin purple," the tube 1.7-1.9 em. long, a lobes 1.0-1.2 cm. long; anthers minutel EE dE EECH 22. P. purpurissata HH. Calyx-lobes 1.2-1.3 em. long, greenish; corolla yellowish- pink, the tube 1.2-1.3 em. long, the lobes 1.4—1.5 cm. long; anthers pilomnlose cade nese. oe sce ees P. discolor See Addenda) GG. Calyx-lobes manifestly shorter than the corolla-tube, cori- aceous or subcoriaceous. H. Leaves delicately membranaceous; corolla-tube gradually constricting toward the orifice; plants of northeastern Cd E EE 28. P. finitima HH. Leaves subcoriaceous; corolla-tube not constricting toward the orifice; plants of eastern Peru and EE EE CN E € 4 a T ORO Ps 9 E P 39.» s EE EE O OR OT AAO ORHAN ORI MANO OO TODO NOD MANO ne OdO naar 5 P. Brittonü CC. IM appendages exserted, or at least attaining ve faucal nnulus ID; Plante glabrous (very minutely a papillate dëi in 34), or only the inflorescence minutely puberulent to glabra E. AR EE compound, or SH simple; corolla-tube ium em. long. F. jid dem greatly EE the subtending leaves; calyx- ely membranaceous. serge 6. P. laxa FF. AE a much shorter e SS subtending leaves; calyx- lobes coriaceous or subcoriaceous. G. Calyx-lobes 0.9—1.6 em. long; inflorescence di- or trichot- omous H. Braets very conspicuous, more or less foliaceous or petal- aceous, caducous (except in 29). L lee appendages barely ons the faucal an- lus; flowers yellow; plants of Ecuador............ II. Epistaminal appendages exserted. J. Inflorescence corymbose, manifestly dichotomous; ucous ; flowers greenish-white or -yellow; GE of eastern Bolivia. ee e e a: 28. P. robusta JJ. Inflorescence densely ies -capitate; bracts persistent; flowers reddish, tinted with uu yel- low, and white; plants of POETE d EE HH. Bracts very inconspicuous, scarious, SE I. Corolla-tube essentially glabrous without; ae gla- DrOus- Cet EE 0. P. trifida ile SE tube minutely velutinous without; e mi- tely puberulent dorsally................... 31. P. vana (470) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 299 GG. Calyx-lobes 0.5—0.8 em. long; inflorescence dichotomous or simple. H. Leaves obovate to obovate-lanceolate, apex obtuse or inflorescence dichotomous; squamellae entire Ces GREEN e da . P. plumierifolia HH. Leaves broadly Mu shortly acuminate; inflorescence simple or obscurely dichotomous; Ru deeply Un Se dew cd Ter dr ER CRI 3. P. amazonica EE. Infloreseence simple; eorolla-tube 0.7—1.5 em. long. F. Leaves e lete labrous; calyx- ps 0.9-1. x em. long; plants of northern and central Brazil......... . P. Lindleyana FF. Leaves puberulent when young, EE Ge beneath ; ealyx-lobes 0.4—0.65 em. long; plants of southeastern Brazil Pees A A een EEN 35. P. denticulata DD. Plants more or less densely ferruginous-pubescent throughout..... do DS Er RU RISES. EE EE 36. P. Meg'agros AA. Neetaries thin and more or less diaphanous (except in 37 38), concrescent fie pom irregularly lacerate); species of Central America (inelud- ing Panama) and Ecuador B. Nectaries more or less thick and fleshy throughout, completely con- crescent. 6 TE di- or trichotomous, much shorter than the subtending ves; calyx-lobes 1.0-1.2 em. long................... 37. P. obovata CC. or dispo: simple, about as long as the subtending leaves or nearly calyx-lobes 0.4-0.5 em. Jong. ....... eere 88. P. concolor BB. ore thin and more or less Re at least the margins, con- erescent (frequently irregularly lacerate C. Epistaminal a deeply included. D. Anthers glabro E. Inflorescence gem shorter than the subtending leaves; calyx- lobes 0.8-1.0 em. long, only slightly suffused with purple at base and tip; corolla-tube gradually constricting I. the orifles; planta of Panne. dee 9 D ee ero 9. P. versicolor EE. PN about as "uc as the subtending leaves, or GC 80; calyx-lobes 1.1-1.5 cm. long, conspicuously and generally suf- fused with purple; ee Am tube not constricting I the los; planta of Ecuador. ...:.. EE chu eo 0. P. peregrina DD. iban minutely deër em EE nflorescence dichotomous; plants of Colombia......... 41. P. vallis EE. Inflorescence simple; plants of British Honduras..... 42. P. Schippii CC. Epistaminal appendages exserted, or at least attaining the orifice. D. Calyx-lobes coriaceous to subcoriaceous; peduncle glabrous or very indefinitely papillate; P of Central Ameriea. C E. Calyx-lobes 0.7—0.9 cm. long, obtusish to acute, scarcely if at all eis with Ei e. wegder) inflorescence to thrice diehotomous........<........ . P. guatemalensis EE. E ie 1.1-1.6 e deese minate, usually gine suffused with purple; BCEE hëtzen usually minutely erose; in- florescence once dichotomous to simple....... 44, P. portobellensis (471) [Vor. 23 300 ANNALS OF THE MISSOURI BOTANICAL GARDEN DD. Calyx-lobes membranaceous; peduncle densely and minutely fer- ruginous-puberulent; plants of eastern Peru......... 45. P. lacerata 14. Prestonia exserta (A. DC.) Standl. Jour. Wash. Acad. Sci. 15: 460. 1925. Haemadictyon exsertum A. DC. in DC. Prodr. 8: 426. 1844; Miers, Apoc. So. Am. 255. 1878. Prestonia tobagensis Urb. Symb. Ant. 5: 467. 1908. Prestoma gracilis Rusby, Deser. So. Am. Pl. 91. 1920. Stems relatively slender, softly puberulent when young, be- coming glabrate and inconspicuously lenticellate when fully mature; leaves elliptic to broadly oval, acute to shortly acumi- nate, base obtuse to rounded, 6-12 em. long, 2-6 cm. broad, membranaceous, either surface opaque, above glabrous, be- neath minutely puberulent or glabrate to glabrous; petioles 0.6-1.5 em. long; stipular appendages intrapetiolar, numerous, minutely dentiform; inflorescence corymbose, simple, bearing 10-40 yellow flowers; peduncle somewhat shorter than the sub- tending leaves, minutely puberulent to glabrate; pedicels 0.8— 1.5 cm. long, somewhat accrescent after maturity, minutely puberulent ; bracts linear to linear-lanceolate, 0.1-0.5 em. long, only slightly foliaceous ; calyx-lobes linear to linear-lanceolate, rarely oblong-elliptic, 0.6-0.7 em. long, foliaceous, minutely puberulent without, the internal squamellae minutely lacerate ; corolla salverform, glabrous or minutely papillate without, the tube 1.5-1.8 em. long, about 0.3 em. in diameter at the base, epistaminal appendages inserted at the base of the faucal an- nulus or slightly below, surpassing the anthers, 0.35-0.55 cm. long, the faucal annulus conspicuously thickened, the lobes broadly dolabriform, shortly acuminate to obtuse, 0.9-1.2 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers oblong-sagittate, 0.45—0.55 em. long, glabrous; ovary ovoid, about 0.2 em. long, glabrous; stigma 0.12-0.2 em. long; nectaries somewhat con- crescent at the base, compressed ovoid-oblongoid, about equal- ling the ovary; follicles relatively long and slender, somewhat agglutinated when young, eventually somewhat tortuous, 23- 29 em. long, glabrous; seeds 1.2-1.3 em. long, the pale yellowish coma 2.2-2.7 em. long. (472) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 301 PANAMA: COCLE: Penonome and vicinity, alt. 50-1000 ft., Febr. 23-Mareh 22, 1908, Williams 362 (NY); PANAMA: vine on shrubs on more or less forested slopes, Taboga Isl., Febr. 26-7, 1923, Macbride 2785 (FM). COLOMBIA: MAGDALENA: rare in forest, near Masinga, Nov. 18, 1898, H. H. Smith 1644 (NY). VENEZUELA: MERIDA: La Victoria, alt. 2000 ft., Nov. 21, 1856, Fendler 2111 (G, MBG); ARAGUA: La Trinidad de Maracay, alt. 440 m., April 18, 1913, Pittier 6047 (B); DISTRITO FEDERAL: Caracas, 1830, Vargas 54 (DC, TYPE); in dry bushes, Puerto Ur t March + 1930, Pittier 13412 (US). TRINIDAD: 13, 1908, Broadway 2730 (B); Belmont, Valley Road, Nov. à pn iners pis: (B); a elimber on hedges, Botanie Station, Tobago, Dec. 6, 1909, Broadway 3373 (B) ; Spring Road, Scarborough, Nov. 13, 1911, Broad- way 4150 (B, $, Gg Welbeck, Tobago, Nov. 23, 1912, Broadway 4350 (B, M); Knagg’s Hill, Port of Spain, Nov. 20, 1921, Williams 10966 (NY); Tobago prope ‘í Frenehfield?" in e interior., alt. 400 m., Oct. 23, 1889, Eggers 5568 (B). This species is manifestly not as uniform as others of the genus, but segregation has appeared inadvisable upon the basis of our present knowledge. Walliams 362, from the northern- most known station of the species, is somewhat distinctive in its smaller size in general, and in the somewhat narrower leaves particularly. Macbride 2785, collected from not far distant, however, is typical in every obvious respect. The species as recognized above also varies in the pubescence of the leaves, the glabrous extreme being found, with obvious intergrada- tions, upon the small island of Tobago, while the more pubes- cent are found upon the mainland in Venezuela. Although the type specimen of P. gracilis Rusby (H. H. Smith 1644) will fall into relationship with P. exserta in the key to species, its some- what broader calyx-lobes suggest that it may represent a nat- ural hybrid with the following species. 15. Prestonia velutina Woodson, Ann. Mo. Bot. Gard. 18: 504. 1931. Stems relatively slender, persistently puberulent or pilosu- lose, apparently not lenticellate; leaves elliptic-lanceolate to broadly oval, apex acuminate, base obtuse, 5-7 em. long, 2-4 em. broad, membranaceous, above pilosulose to glabrate, beneath, densely and somewhat yellowish-tomentulose to minutely velu- tinous; petioles 0.5-1.5 em. long; stipular appendages intra- petiolar, numerous, minutely dentiform-flagelliform ; inflores- cence corymbose, simple, bearing 10-15 greenish-yellow (473) [Vor. 23 302 ANNALS OF THE MISSOURI BOTANICAL GARDEN flowers; peduncle somewhat shorter than the subtending leaves, shortly hirtellous or pilosulose; pedicels 1.0-1.4 em. long, somewhat accrescent after maturity, pubescence as on the peduncle; bracts elliptic to obovate-lanceolate, 0.7—1.0 em. long, conspicuously foliaceous, minutely puberulent; calyx-lobes oblong-lanceolate, acute to acuminate, 1.0-1.2 em. long, folia- ceous, minutely puberulent without, the internal squamellae deltoid-ligular, minutely erose; corolla salverform, glabrous without, the tube 1.6-1.8 cm. long, about 0.3 em. in diameter at the base, epistaminal appendages inserted at the base of the faucal annulus or slightly below, surpassing the anthers, 0.45— 0.55 cm. long, the faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.8-0.9 em. long, re- flexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers oblong-sagittate, 0.5 em. long, glabrous ; ovary ovoid, about 0.2 em. long ; stigma 0.15 em. long; nectaries fleshy, more or less concrescent, somewhat shorter than the ovary; follicles unknown. COLOMBIA: MAGDALENA: Aracataca, alt. 30 m., Dec. 25, 1925, Schultze 245 (B); CUNDINAMARCA: moist open loam, alt. 350—400 m., Girardot, July 19, 1917, Rusby $ Pennell 121 (G, NY); Girardot, Febr. 15, 1929, Toro 59 (NY); TOLIMA: Honda, Aug., 1919, Ariste-Joseph s.n. (NY, TYPE, MBG, photograph and analytical drawings). This species is somewhat intermediate between P. exserta and P. parvifolia, of which it may conceivably be a hybrid, although it is evidently rather stable in its morphology, and extends without the known range of the former species, except to the north, and is even further removed from the latter to the south in Ecuador. 16. Prestonia parvifolia K. Sch. spec. nov. in herb. Suffrutescens volubilis; ramulis gracilibus juventate pilosu- lis hirtellisve tandem glabratis haud conspicue lenticellatis ; foliis ovalibus obovato-elliptieisve apice breviter acuminatis basi obtusis 5-7 em. longis 2-4 em. latis membranaceis opacis supra minute sparseque strigilloso-hirtellis subtus minute puberulis; petiolis 0.4—0.6 cm. longis puberulis; appendicibus stipulaceis intrapetiolaribus numerosis minute dentiformibus ; (474) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 303 inflorescentiis corymbosis simplicibus 15-30-floris ; pedunculo foliis breviore hirtello; pedicellis 0.7-1.0 cm. longis puberulis ; bracteis obovato-ellipticis 0.7-1.0 em. longis foliaceis sparse minuteque puberulo-papillatis; calycis laciniis obovato-ellip- ticis brevissime acuminatis 1.1—1.4 cm. longis foliaceis minute sparseque puberulo-papillatis squamellis deltoideis minute erosis; corollae salverformis virido-luteae extus sparse mi- nuteque puberulo-papillatae tubo 1.6-1.8 em. longo basi ca. 0.4 em. diametro metiente appendicibus epistaminalibus paulo exsertis 0.25-0.3 em. longis multo brevioribus quam antheris annulo faucium eonspieue inerassato lobis oblique obovatis breviter aeuminatis 0.9-1.0 cm. longis reflexis patulisve; antheris oblongo-sagittatis 0.5-0.55 em. longis glabris; ovario ovoideo ca. 0.2 em. longo glabro; stigmate 0.15-0.16 em. longo; nectariis compresse oblongoideis basi plus minusve concres- centibus ovario paululo brevioribus ; folliculis ignotis. Stems relatively slender, pilosulose or hirtellous when young, glabrate but not conspieuously lenticellate when fully mature; leaves oval to obovate-elliptic, apex shortly acuminate, base obtuse, 5-7 cm. long, 2-4 cm. broad, membranceous, opaque, above minutely and sparsely strigillose-hirtellous, be- neath minutely puberulent; petioles 0.4—0.6 em. long, puberu- lent; stipular appendages intrapetiolar, numerous, minutely dentiform ; inflorescence corymbose, simple, bearing 15—30 yel- low flowers; peduncle somewhat shorter than the subtending leaves, hirtellous; pedicels 0.7-1.0 em. long, somewhat accres- cent after maturity, puberulent; braets obovate-elliptie, 0.7— 1.0 em. long, conspicuously foliaceous, minutely and sparsely puberulent-papillate ; calyx-lobes obovate-elliptie, very shortly acuminate, 1.1—1.4 em. long, foliaceous, minutely and sparsely puberulent-papillate, the internal squamellae deltoid, minutely erose ; corolla salverform, minutely puberulent-papillate with- out, the tube 1.6-1.8 em. long, about 0.4 em. in diameter at the base, epistaminal appendages slightly exserted, much sur- passed by the anthers, 0.25-0.3 cm. long, the faucal annulus con- spicuously thickened, the lobes obliquely obovate, shortly acu- minate, 0.9-1.0 em. long, reflexed or widely spreading ; stamens inserted at about the upper 14 of the corolla-tube, the anthers (475) [Vor. 23 304 ANNALS OF THE MISSOURI BOTANICAL GARDEN oblong-sagittate, 0.5-0.55 em. long, glabrous; ovary ovoid, about 0.2 em. long, glabrous; stigma 0.15-0.16 em. long; nec- taries compressed-oblongoid, somewhat concrescent at the base, slightly shorter than the ovary; follicles unknown. ECUADOR: MANABI: prope Hacienda El Reereo, pa 6, 1895, Eggers 15430 (B, TYPE, MBG, photograph and analytical drawings) ; in locis siccis, El Recreo, Ge 1893, Eggers 15430 bis (M); GUAYAS: ''FI. i date lacking, Pavon s.n. (BB). Similar to P. velutina in the approximate shape and size of the bracts and calyx-lobes, but differing notably in the shorter, barely exserted epistaminal appendages, more floriferous in- florescences, and puberulent-papillate corolla. 17. Prestonia perplexa Woodson, spec. nov. Suffruticosa vel suffrutescens volubilis; ramulis gracilibus juventate minute hirtellis tandem glabratis inconspicue lenti- cellatis cortice rubro-brunneo; foliis ovato-ellipticis apice acu- minatis basi obtusis 6-10 em. longis 3-6 em. latis rigide mem- branaceis supra juventate minutissime puberulis mox glabratis subtus glabris; petiolis 0.8-1.0 cm. longis glabris; appendicibus stipulaceis interpetiolaribus numerosis minute angusteque dentiformibus; inflorescentiis subeorymbosis simplicibus 12- 20-floris; pedunculo foliis subaequante vel paulo superante minute hirtello; pedicellis 1.2-1.5 em. longis post maturitatem conspicue accrescentibus minute hirtellis; bracteis linearibus 0.1-0.4 em. longis paululo foliaceis; calycis laciniis late obo- vato-elliptieis acutis acuminatisve 0.8-1.1 em. longis foliaceis sparse indistincteque puberulo-papillatis squamellis late del- toideis minute emarginatis vel erosis; corollae salverformis dilute virido-luteae extus sparse minutissimeque papillatae tubo 1.3-1.5 em. longo basi ca. 0.3 em. diametro metiente appen- dicibus epistaminalibus haud vel vix exsertis 0.2-0.25 em. longis multo brevioribus quam antheris annulo faucium con- spicue inerassato lobis obovato-dolabriformibus haud acumi- natis 0.6-0.8 cm. longis patentibus ; antheris exsertis sagittatis 0.45 em. longis glabris ; ovario ovoideo ca. 0.1 em. longo minutis- sime papillato; stigmate 0.175-0.2 em. longo; nectariis com- (476) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 305 presse ovoideis basi connatis ovarium aequantibus; folliculis ignotis. Stems relatively slender, minutely hirtellous when young, becoming glabrate and very inconspicuously lenticellate with a reddish-brown bark; leaves ovate-elliptic, apex acuminate, base obtuse, 6-10 em. long, 3-6 cm. broad, firmly membrana- ceous, above very minutely puberulent when young, soon be- coming glabrate, beneath glabrous; petioles 0.8-1.0 cm. long, glabrous; stipular appendages interpetiolar, minutely and nar- rowly dentiform; inflorescence subcorymbose, simple, bearing 12-20 pale greenish-yellow flowers; peduncle about equalling or slightly surpassing the subtending leaves, minutely hirtel- lous; pedicels 1.2-1.5 em. long, conspicuously accrescent after maturity, minutely hirtellous; bracts linear, 0.1-0.4 em. long, only slightly foliaceous; calyx-lobes broadly obovate-elliptie, acute to acuminate, 0.8-1.1 em. long, conspicuously foliaceous, membranaceous, sparsely and indistinctly puberulent-papil- late, the squamellae broadly deltoid, minutely emarginate or erose; corolla salverform, sparsely and very minutely papil- late without, the tube 1.3-1.5 em. long, about 0.3 em. in diameter at the base, epistaminal appendages barely exserted or es- sentially included, much surpassed by the anthers, 0.2-0.25 em. long, faucal annulus conspicuously thickened, the lobes obo- vate-dolabriform, not acuminate, 0.6-0.8 em. long, reflexed or widely spreading; anthers inserted at about the upper Y of the corolla-tube, sagittate, 0.45 em. long, glabrous, the tips con- spieuously exserted; ovary ovoid, about 0.1 cm. long, very minutely papillate; stigma 0.175-0.2 em. long; nectaries com- pressed-ovoid, united at the base, fleshy, about equalling the ovary ; follieles unknown. BRAZIL: DATA INCOMPLETE: Lund s.n. (C, TYPE, MBG, photograph and analytieal drawings). It is obviously dangerous, as well as exasperating, to base a species upon a single collection bearing no data but the name of the collector and merely the notation **Brasilia." Neverthe- less, a taxonomie monograph is obliged to consider all material available however scanty and dubious as to origin. As the (477) . [ Vor. 23 306 ANNALS OF THE MISSOURI BOTANICAL GARDEN majority of Brazilian Apocynaceae collected by Lund and bearing more exact data have been from the southeastern coastal states, it appears that the provenience of the type speci- men of P. perplexa may rather safely be assigned to that gen- eral region, the states of Rio de Janeiro or Sáo Paulo in particular. The morphological characters of P. perplexa indicate a strong relationship to P. parvifolia of Ecuador, and to no other known species of its own regional flora. From the latter spe- cies, however, P. perplexa is amply distinct in such characters as the dimensions of the flower as well as those employed in the key. 18. Prestonia mollis HBK. Nov. Gen. 3: 221. 1819. Prestoma glabrata HBK. loc. cit. 222. 1819; Miers, Apoc. So. Am. 145. 1878, not K. Sch. Haemadictyon molle (HBK.) A. DC. in DC. Prodr. 8: 427. 1844. Haemadictyon glabratum (HBK.) A. DC. loc. cit. 1844. Haemadictyon pallidum A. DC. loc. cit. 428. 1844; Miers, loc. cit. 259. 1878. Haemadictyon tomentellum Benth. Bot. Voy. Sulph. 126. 1844 ; Miers, loc. cit. 1878. Prestonia ecuadorensis K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 4?: 188. 1895. Prestonia Weberbaueri Mgf. Notizblatt 9: 89. 1994. Stems relatively slender, usually minutely puberulent or puberulent-papillate when young, soon becoming glabrate and rather conspicuously lenticellate; leaves broadly ovate to oblong-elliptic, apex shortly and abruptly acuminate to obtuse or rounded, base usually broadly and rather obscurely cordate to rounded or somewhat truncate, 4-15 em. long, 2-9 em. broad, membranaceous, above minutely and rather sparsely puberu- lent-papillate to glabrate or glabrous, beneath minutely tomen- tulose to glabrate or glabrous, usually paler and somewhat glaucescent; petioles 0.5-3.5 cm. long, minutely and rather sparsely puberulent-papillate to glabrate or glabrous; stipular appendages intrapetiolar, numerous, minutely dentiform-fla- (478) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 307 gelliform ; inflorescence corymbose, simple, bearing 10-40 pale yellow flowers; peduncle somewhat shorter than the subtend- ing leaves, very minutely puberulent to glabrous; pedicels 0.6-1.2 em. long, somewhat accrescent after maturity, minutely puberulent to glabrous; bracts narrowly oblong-lanceolate, acuminate, 0.5-2.6 em. long, conspicuously foliaceous, minutely papillate to glabrous; calyx-lobes narrowly oblong-lanceolate to linear, acuminate, 0.8-2.1 em. long, conspicuously foliaceous, minutely papillate to glabrous without, the internal squamellae deltoid, minutely erose or lacerate; corolla salverform, gla- brous or minutely papillate without, 2.2-3.4 em. long, 0.4—0.5 cm. in diameter at the base, epistaminal appendages slightly exserted or at least attaining the orifice, 0.2-0.55 cm. long, the faucal annulus conspicuously thickened, the lobes obliquely obovate, usually shortly acuminate, occasionally obtuse, 1.7— 2.5 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers narrowly sagittate, 0.6-0.8 em. long, minutely hirtellous, rarely nearly elabrate, slightly exserted; ovary ovoid, about 0.2 cm. long, glabrous; stigma 0.15-0.2 em. long; nectaries compressed- ovoid, separate or essentially so, about half equalling the ovary; follieles relatively stout, continuous, usually aggluti- nated and joined at the tips, 15-30 em. long, glabrous; seeds 0.8-1.0 em. e the pale yellowish coma 3.5-4.0 cm. long. CUADOR : prope Hacienda El Reereo, Aug. 4, 1893, Eggers 14956 (B, ); GUAYAS: imm Puna, May, 1892, Eggers 14782 (B, M); prope Guayaquil, date lacking, Sodiro s.n. (B) ; on Puna, 1852, Andersson 79 (8); Guayaquil, 1862, Andersson 79 (S); Guayaquil, date lacking, SEN 670 (C EY V); ad fluvi Daule prope bigis date lacking, Spruce 6484 (B, BB L, V) ; oil camp E tween Guayaquil and Salinas, alt. 0—100 m., ia 21—24, uM Hitohoook 20069 (G, NY, US); Panigon Plantation, 8 miles en of Milagro, alt. 50 m., July 11-13, 1923, Hitchcock 20593 (G, NY, US); DATA INCOMPLETE: Warszewicz s.n. (B); i p. dea eomm 1857-9, Spruce 6019 (BB, DC, DL, G, hs LE im Matthews s.n. (BB, NY); am Marañon a Balsas, ce Mii gemi ümen, Straüchen, ut d Rohrgrüssern, alt. 920 m., June 24, 1904, Ned. pes (B); CAJAMARCA: Tal des Marañon bei Bellavista, Regengrünes Gebiisch, gebildet aud hohen Bteiehera und kleinen Baü- men, alt. 500—600 m., Mai 2, 1912, Weberbauer 6231 (B, G) The specimens from northern Peru, representing roughly P. Weberbaueri Mef., appear to be distinguished superficially (479) [Vor. 23 308 ANNALS OF THE MISSOURI BOTANICAL GARDEN from the more typical specimens from farther north in having somewhat smaller leaves with less indication of the obscurely cordate base. However, enlisting all the specimens cited above in a rather uniform series of intergradations, it will be found that the extremes have been collected in the province of Guayas, Ecuador, alone. The type specimen of P. Weber- baueri (Weberbauer 6231 in herb. Berol.) is also distinguished by anthers approaching glabrescenee. "This character, how- ever, is not shared by other specimens from Peru. Seen sepa- rately, the specimens cited appear to indicate need of segre- gation: when examined altogether, the unity of the species is apparent. 19. Prestonia didyma (Vell.) Woodson, comb. nov. Echites didyma Vell. Fl. Flum. 109. 1830; Icon. 3: pl. 27. 1827; A. DC. in DC. Prodr. 8: 468. 1844; Muell.-Arg. in Mart. Fl. Bras. 6!: 155. 1860. Haemadictyon membranaceum Muell.-Arg. loc. cit. 167. 1860; Miers, Apoc. So. Am. 260. 1878. Rhaptocarpus didymus (Vell.) Miers, loc. cit. 152. 1878. Stems relatively stout, puberulent when young, glabrate and inconspicuously lenticellate when fully mature; leaves ovate- elliptic to rather narrowly oval, apex acuminate, base obtuse, 9-16 em. long, 3.0-8.5 em. broad, membranaceous, glabrous; petioles 1-2 em. long; stipular appendages intrapetiolar, nu- merous, very minutely dentiform; inflorescence racemose, simple, bearing 10-20 rather livid greenish-yellow flowers; peduncle somewhat shorter than the subtending leaves, mi- nutely puberulent ; pedicels 1.5-2.1 em. long, somewhat accres- cent after maturity, minutely puberulent; bracts linear, 0.3- 0.4 cm. long, very slightly foliaceous; calyx-lobes elliptie- lanceolate, acuminate, 1.2-1.8 cm. long, green, faintly flushed with purple toward the base, glabrous, the internal squamellae deltoid-trigonal, minutely emarginate or erose; corolla salver- form, glabrous or minutely papillate without, the tube 0.8-1.5 cm. long, about 0.3 cm. in diameter at the base, epistaminal ap- pendages wholly included, about 0.2 em. long, the faucal an- nulus conspicuously thickened, the lobes obliquely obovate, (480) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 309 0.75-1.0 em. long, reflexed or widely spreading; stamens in- serted at about the upper 1% of the corolla-tube, the anthers very narrowly sagittate, about 0.6 cm. long, glabrous, slightly exserted; ovary ovoid, about 0.1 cm. long, glabrous; stigma 0.15 em. long; nectaries compressed-ovoid, conspicuously con- crescent at the base, somewhat surpassing the ovary; follicles relatively stout, continuous, usually united at the tips and somewhat agglutinated, 20-25 em. long, glabrous ; seeds 1.0—1.1 em. long, the pale yellowish coma 3.0-3.5 cm. long. BRAZIL: RIO GRANDE DO NORTE: Taipu, date lacking, Schott 5389 (V, TYPE, MBG, photograph and analytical drawings); RIO DE JANEIRO: ad urbem loco Fortaleza Sao João, March 25, 1916, Frazão 7144 (B); DATA INCOMPLETE: Sellow s.n. (B). The foundation of this transfer appears to be well estab- lished through a technieal examination of Velloso's plate of a plant collected ‘‘in sylvis maritimis"! in the vicinity of Rio de Janeiro. Velloso’s illustrations, although often almost com- ically wooden and lifeless, frequently portray with surprising aptitude basic characteristics of the species depicted. In this respect they may be comparable to modern ‘‘cartoons.’’ 20. Prestonia annularis (L. f.) G. Don, Gen. Hist. 4: 84. 1838. Echites annularis L. f. Suppl. 166. 1781. Haemadictyon ? annulare (L. f.) A. DC. in DC. Prodr. 8: 428. 1844. Temnadenia annularis (L. f.) Miers, Apoc. So. Am. 216. 1878. Plants completely glabrous; stems relatively stout, conspic- uously lenticellate when fully mature; leaves broadly ovate to oblong-elliptic, apex obtuse to acute, base obtuse to rounded, 14-27 cm. long, 4.5-12.0 em. broad, subcoriaceous, opaque; petioles 0.7-2.0 em. long; stipular appendages intra- or inter- petiolar, numerous, minutely dentiform-flagelliform; inflores- cence racemose, dichotomous, bearing 10-30 purplish-yellow flowers; peduncle much shorter than the subtending leaves; pedicels 1.1-1.3 em. long, somewhat accrescent after maturity ; bracts linear, about 0.1 em. long, scarious or only slightly folia- ceous; calyx-lobes oblong-elliptic, acute to acuminate, 0.7—1.0 em. long, subcoriaceous, pale green suffused with purple, gla- (481) [Vor. 23 310 ANNALS OF THE MISSOURI BOTANICAL GARDEN brous without, the internal squamellae deltoid-ligular, mi- nutely erose or truncate; corolla salverform, glabrous or very minutely papillate without, the tube 1.3-1.5 em. long, about 0.25 em. in diameter at the base, epistaminal appendages wholly included, about 0.15 em. long, the faucal annulus conspicuously thickened, the lobes obliquely obovate-dolabriform, 0.8-0.9 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers oblong-sagittate, 0.45-0.5 em. long, minutely puberulent-papillate dorsally, slightly exserted; ovary ovoid, about 0.1 em. long, glabrous; stigma 0.1-0.15 em. long; nectaries compressed-ovoid, sepa- rate or somewhat conerescent at the base, about equalling the ovary; follicles relatively long and stout, continuous, usually somewhat faleate and frequently united at the tips, 40-45 em. long, glabrous; seeds 1.2-1.4 cm. long, the pale yellowish coma 3.9—4.0 em. long. DUTCH GUIANA: e regione Para, 1851, Wullschlágel 1028 (Bx); Jaglust, fluv. Suriname, in silva, June 26, 1913, Alprato 40E (U); Paramaribo, date lacking, Focke 1056 (U); Paramaribo, June 25, 1850, Wullschlügel 1028 (U, V); data in- complete, Hostmann $ Kappler s.n. (S). 21. Prestonia guianensis Gleason, Bull. Torrey Bot. Club 53: 209. 1926. Plants completely glabrous; stems relatively stout, conspic- uously lenticellate when fully mature; leaves oval to oblong- elliptic, apex acuminate, base obtuse to rounded, 10-13 em. long, 4.5-6.0 em. broad, subcoriaceous, opaque; petioles 0.7- 1.0 em. long; stipular appendages intrapetiolar, numerous, mi- nutely dentiform; inflorescence racemose, dichotomous, bear- ing 30-40 cream-colored flowers; peduncle equalling or some- what surpassing the subtending leaves; pedicels 0.8-1.0 em. long, somewhat accrescent after maturity; bracts linear, sca- rious, less than 0.1 em. long; calyx-lobes oblong-elliptic, acute to acuminate, membranaceous or nearly subcoriaceous, 0.9— 1.0 em. long, green suffused with purple, the internal squamel- lae broadly deltoid-dentiform, truncate or very minutely erose; corolla salverform, glabrous or very minutely papillate with- out, the tube 1.2-1.4 cm. long, about 0.3 em. in diameter at the base, epistaminal appendages wholly included, about 0.07—0.1 (482) 1936] . WOODSON—STUDIES IN THE APOCYNACEAE. IV ah? em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 1.0-1.25 cm. long; sta- mens inserted at about the upper 14 of the corolla-tube, the anthers elliptic-sagittate, 0.5-0.55 em. long, puberulent-papil- late dorsally, slightly exserted; ovary ovoid, about 0.1 cm. long, glabrous ; stigma 0.08—0.1 cm. long; nectaries compressed- ovoid, separate or slightly concrescent at the base, equalling the ovary ; follicles unknown. BRITISH GUIANA: Pomeroon River, Pomeroon District, Dee. 17-24, 1922, Cruz 3097 (G, NY, TYPE, MBG, photograph and analytical drawings). Although this species is closely related to P. annularis of ad- joining Dutch Guiana, it seems distinct because of its thicker foliage, as well as its longer corolla-lobes which appear nearly white, and longer inflorescence with longer floriferous branches and shorter sterile primary peduncle. 22. Prestonia purpurissata Woodson, spec. nov. Suffruticosa volubilis ; ramulis erassiusculis maturitate gla- bris conspicue lenticellatis ; foliis ovalibus oblongo-ellipticisve apice breviter acuminatis obtusisve basi obtusis vel rotundatis 9-15 em. longis 3.0-6.5 cm. latis firme membranaceis opacis omnino glabris; petiolis 0.5-1.0 em. longis ; appendicibus stipu- laceis interpetiolaribus numerosis minute dentiforme-flagelli- formibus; inflorescentiis subeorymbosis di- vel trichotome di- visis flores 20-40 purpurissatos gerentibus; pedunculo foliis subaequante; pedicellis 1.3-1.5 em. longis post maturitatem paulo accrescentibus minutissime papillatis; bracteis lineari- bus 0.1-0.4 em. longis scariaceis; calycis laciniis oblongo-lance- olatis acuminatis 1.7—1.8 cm. longis membranaceis sat purpuris- satis squamellis dentiformibus emarginatis vel paulo laceratis; corollae salverformis extus glabrae vel minutissime papillatae tubo 1.7-1.9 em. longo basi ca. 0.35 em. diametro metiente ap- pendicibus epistaminalibus omnino inclusis ca. 0.1 cm. longis annulo faucium conspicue incrassato lobis oblique obovatis acu- minatis 1.0-1.2 cm. longis patentibus; antheris pandurate sagittatis 0.6 em. longis dorso minute papillatis paulo exsertis ; ovario ovoideo ca. 0.2 em. longo glabro; stigmate ca. 0.15 em. longo; nectariis compresse oblongoideis integris vel basi paul- (483) (Vor. 23 312 ANNALS OF THE MISSOURI BOTANICAL GARDEN lulo concrescentibus ovarium paulo superantibus; folliculis ignotis. Plants completely glabrous; stems relatively stout, glabrous and conspicuously lenticellate when fully mature; leaves oval to oblong-elliptic, apex obtuse or shortly acuminate, base ob- tuse or rounded, 9-15 cm. long, 3.0-6.5 em. broad, firmly mem- branaceous, opaque; petioles 0.5-1.0 em. long; stipular ap- pendages interpetiolar, numerous, minutely dentiform-flagelli- form; inflorescence subcorymbose, di- or trichotomous, bearing 20—40 *'raisin-purple" flowers; peduncle about equalling the subtending leaves; pedicels 1.3-1.5 cm. long, somewhat accres- cent after maturity, very minutely papillate ; bracts linear, 0.1— 0.4 em. long, scarious; calyx-lobes oblong-lanceolate, acumi- nate, 1.7-1.8 em. long, membranaceous, conspicuously flushed with purple, the internal squamellae dentiform, emarginate or slightly erose; corolla salverform, glabrous or very minutely papillate without, the tube 1.7-1.9 cm. long, about 0.35 em. in diameter at the base, epistaminal appendages wholly included, about 0.1 cm. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, acuminate, 1.0-1.2 em. long, reflexed or widely spreading; stamens inserted at about the upper Y, of the corolla-tube, the anthers sagittate, somewhat pandurate, 0.6 em. long, minutely papillate dorsally, slightly exserted; ovary ovoid, about 0.2 em. long, glabrous; stigma about 0.15 em. long; nectaries compressed-oblongoid, separate or somewhat concrescent at the base, slightly surpassing the ovary; follicles unknown. COLOMBIA: EL VALLE: edge of forest, La Cumbre, alt. 1650-1850 m., May 14- 19, 1922, PR 5719 (US, TYPE, MBG, photograph and analytical EE Differs from its immediate relatives, P. finitima and P. Phenax superficially in its longer, delicately petalaceous calyx- lobes. 23. Prestonia finitima Woodson, spec. nov. Suffruticosa volubilis; ramulis erassiuseulis juventate mi- nute puberulis maturitate glabratis inconspicueque lenticel- latis; foliis ellipticis apice acuminatis basi obtusis 9-25 em. longis 3.0-11.0 em. latis tenuiter membranaceis opacis juven- (484) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 319 tate minute sparseque puberulis mox glabratis; petiolis 1.3- 2.5 em. longis; appendicibus stipulaceis interpetiolaribus mi- nute dentiforme-flagelliformibus; inflorescentiis subeorym- bosis dichotome divisis flores 15-20 lutescentes gerentibus; peduneulo foliis paulo breviore; pedicellis 1.4-1.5 cm. longis, post maturitatem paulo accrescentibus indistincte papillatis ; braeteis minutissimis vix bene visis; calycis laciniis late ob- longo-elliptieis breviter acuminatis 1.2-1.4 em. longis subcori- aceis purpurissatis glabris vel indistinctissime papillatis squamellis latissime deltoideis erosis; corollae salverformis extus indistincte papillatae tubo 1.8-1.9 em. longo basi ca. 0.4 em. diametro metiente prope fauces sensim angustato appen- dicibus epistaminalibus omnino inclusis ca. 0.1 cm. longis an- nulo faucium conspicue inerassato lobis oblique-obovatis bre- viter acuminatis 0.7—1.0 em. longis patentibus ; antheris pandu- rate sagittatis 0.5 em. longis dorso puberulo-papillatis paulo exsertis; ovario ovoideo ca. 0.15 em. longo glabro; stigmate 0.15—0.2 em. longo; nectariis compresse-oblongoideis ovarium paulo superantibus prope basem connatis apice truncatis vel leviter undulatis ; folliculis ignotis. Stems relatively stout, minutely puberulent when young, glabrate and inconspicuously lenticellate when fully mature; leaves elliptic, apex acuminate, base obtuse, 9-25 em. long, 3.0- 11.0 em. broad, delicately membranaceous, minutely and sparsely puberulent when young, soon becoming glabrate; petioles 1.3-2.5 em. long, glabrous; stipular appendages inter- petiolar, numerous, minutely dentiform-flagelliform ; inflores- cence subcorymbose, dichotomous, bearing 15-20 brownish- yellow flowers ; peduncle somewhat shorter than the subtending leaves, glabrous; pedicels 1.4-1.5 em. long, somewhat accres- cent after maturity, indistinctly papillate; bracts very minute, scarcely visible; calyx-lobes broadly oblong-elliptic, shortly acuminate, 1.2-1.4 cm. long, subcoriaceous, suffused with purple, glabrous or very indistinctly Tanini, the internal squamellae very broadly deltoid, erose; corolla salverform, in- distinctly papillate without, the tube 1.8-1.9 cm. long, about 0.4 cm. in diameter at the TE gradually constricting toward the orifiee, epistaminal appendages wholly ineluded, about 0.1 (485) [ Vor. 23 314 ANNALS OF THE MISSOURI BOTANICAL GARDEN em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.7-1.0 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers pandurately sagittate, 0.5 cm. long, puberulent-papillate dorsally, slightly exserted; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.15—0.2 em. long; nectaries compressed-oblongoid, slightly surpassing the ovary, connate toward the base, the apices truncate or slightly undu- late; follicles unknown. IL: AMAZONAS: Suhinya, Sept. 1, 1912, Koch-Gruenberg 30 (B, TYPE, MBG, RAZ photograph and analytieal drawings); Municipality Humayta, near Tres Casas, on varzea land, Sept. 14—Oct. 11, 1934, Krukoff 6190 (NY). There is a very obvious danger in depending too trustingly upon collectors! notes concerning flower color in the majority of eases, without doubt. In the instance of the species center- ing about P. Phenax and P. trifida, however, the collector's re- port of the flower color has been found to coincide with certain morphological characters used in the delimitation of species. Furthermore in the case of species represented by several spe- cimens with flower color noted, sufficient unanimity has been found to use such observations in descriptions, and to a less extent in keys. It is thought that such notes may be of use in identification, particularly when fresh material may be available. 24. Prestonia Phenax Woodson, spec. nov. Suffrutieosa volubilis; ramulis crassiusculis maturitate gla- bris conspicue lenticellatis; foliis late ovatis vel late oblongo- ellipticis apice acuminatis rariusve obtusiusculis basi obtusis vel rotundatis 10-28 em. longis 4-14 cm. latis subcoriaceis opacis glabris ; petiolis 0.8-2.0 em. longis glabris ; appendicibus stipulaceis interpetiolaribus numerosis minute dentiformibus ; inflorescentiis corymbosis di- vel trichotome divisis flores viridi-luteos 10-30 gerentibus; peduneulo foliis multo brev- iore; pedicellis 1.0-1.3 em. longis glabris post maturitatem paulo accrescentibus; bracteis linearibus minutis vix bene visis ; calycis laciniis oblongo-elliptieis apice acutis vel breviter acuminatis 1.2-1.3 em. longis membranaceis vel subcoriaceis (486) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 315 viridibus basi apiceque purpurissatis glabris squamellis late deltoideis minute emarginatis integrisve; corollae salver- formis extus glabrae vel minutissime papillatae tubo 1.5-1.7 em. longo basi ca. 0.3-0.4 cm. diametro metiente appendicibus epistaminalibus omnino inclusis ea. 0.1-0.2 em. longis annulo faucium conspicue inerassato lobis oblique obovatis breviter acuminatis 0.8-1.2 em. longis patentibus; antheris sagittatis paulo panduratis 0.5 em. longis dorso minute puberulo-papil- latis paulo exsertis; ovario ovoideo ca. 0.15 em. longo glabro; stigmate 0.15-0.2 em. longo; nectariis compresse oblongo- ovoideis integris vel basi concrescentibus ovarium paulo super- antibus ; folliculis juventate gracilibus continuis falcatis apice connatis ; seminibus ignotis. Plants completely glabrous; stems relatively stout, conspic- uously lenticellate when fully mature; leaves broadly ovate to oblong-elliptie, apex acuminate to somewhat obtuse, base ob- tuse or rounded, 10-28 cm. long, 4-14 em. broad, subcoriaceous, opaque; petioles 0.8-2.0 em. long; stipular appendages inter- petiolar, numerous, minutely dentiform; inflorescence corym- bose, di- or trichotomous, bearing 10-30 greenish-yellow, brown-tinted flowers; peduncle much shorter than the subtend- ing leaves ; pedicels 1.0-1.3 em. long, somewhat accrescent after maturity; bracts linear, minute, very inconspicuous; calyx-lobes oblong-elliptic, acute or shortly acuminate, 1.2-1.3 cm. long, membranaceous or subcoriaceous, green tinted with purple at base and tip, glabrous, the internal squamellae broadly deltoid, entire or minutely emarginate; corolla salverform, glabrous or minutely papillate without, the tube 1.5-1.7 em. long, about 0.3- 0.4 cm. in diameter at the base, epistaminal appendages wholly included, 0.1-0.2 em. long, faucal annulus conspicuously thick- ened, the lobes obliquely obovate, shortly acuminate, 0.8-1.2 em. long, reflexed or widely spreading; stamens inserted at about the upper 1% of the corolla-tube, the anthers oblong-sagit- tate, somewhat pandurate, 0.5 cm. long, minutely puberulent- papillate dorsally, slightly exserted; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.15-0.2 em. long; nectaries com- pressed oblong-ovoid, separate or somewhat united at the base, (487) [Vor. 23 316 ANNALS OF THE MISSOURI BOTANICAL GARDEN somewhat surpassing the ovary; immature follicles relatively slender, continuous, falcate, united at the apex ; seeds unknown. PERU: LORETO: Urwald, Stromgebiet des Ucayali von 10° S. bis zur Miindung, July 26, 1923, Tessmann 3046 (B, TYPE, MBG, photograph and analytical draw- ings); edge of water, Fortaleza, Yurimaguas, Oct.-Nov., 1929, Williams 4310 (B, FM). BOLIVIA: SANTA CRUZ: barranea en el bosque, Rio Surutu, alt. 400 m., Oct. 8, 1925, Steinbach 7272 (B). Williams 4310 may be found to represent another species or variety, since it differs from the other specimens cited in having somewhat smaller, narrower leaves, and more conspic- uously purplish calyx. Tessmann 3046 ani Steinbach 7272 from widely separate localities, however, show striking uni- formity in all observable characters. 25. Prestonia Brittonii N. E. Br. Bull. Torrey Bot. Club 51: 9. 1924 Plants completely glabrous; stems relatively stout, conspicu- ously lenticellate when fully mature; leaves oval, apex shortly acuminate, base obtuse or rounded, 13-16 em. long, 7-9 em. broad, rigidly membranaceous to subcoriaceous, opaque; peti- oles 1.5-2.5 em. long; stipular appendages interpetiolar, numerous, minutely dentiform-flagelliform; inflorescence co- rymbose, repeatedly dichotomous, bearing 40-50 brownish- yellow flowers; peduncle much surpassing the subtending leaves; pedicels 1.7-2.1 em. long, somewhat accrescent after maturity; bracts minutely ovate-lanceolate, very inconspicu- ous; calyx-lobes oblong-elliptic, apex acute to shortly acumi- nate, 1.0-1.25 em. long, subcoriaceous, deep purple, glabrous or very indistinctly papillate, the internal squamellae broadly deltoid, minutely erose; corolla salverform, glabrous without, the tube 1.5-1.8 em. long, about 0.35 em. in diameter at the base, very slightly constricting toward the orifice, epistaminal ap- pendages deeply included, 0.15 em. long, faucal annulus con- spicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.8-0.9 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers oblong-sagittate, somewhat pandurate, 0.5 em. long, (488) 1936] j WOODSON—STUDIES IN THE APOCYNACEAE. IV 317 minutely puberulent-papillate dorsally, slightly exserted; ovary ovoid, about 0.1 em. long, glabrous; stigma 0.15 em. long; nectaries compressed-oblongoid, separate or slightly concres- cent at the base, apex truncate or minutely emarginate, equal- ling or slightly surpassing the ovary; follicles unknown. TRINIDAD: Heights of Aripo, Jan. 10-26, 1922, Broadway 10009 (NY, ISOTYPE, MBG, photograph and analytical drawings). 26. Prestonia laxa Rusby, spec. nov. in herb. Suffruticosa volubilis; ramulis gracilibus juventate minute sparseque puberulis maturitate glabratis inconspicue lenticel- latis; foliis oblongo-ellipticis apice breviter acuminatis basi obtusis rotundatisve 9-19 em. longis 2.7-7.5 cm. latis rigide membranaceis vel subcoriaceis opacis vel supra paululo nitidulis; petiolis 0.5—0.8 em. longis; appendicibus stipulaceis interpetiolaribus numerosis minute dentiformibus; inflores- centiis corymbosis dichotome subdivisis flores 10-40 albidos carneo-maculatos gerentibus; pedunculo folia multo super- ante; pedicellis 1.5-2.0 em. longis post maturitatem paulo ac- crescentibus minutissime papillatis ; bracteis linearibus 0.1-0.2 em. longis seariaceis; calycis laciniis oblongo-ellipticis apice breviter acuminatis 1.0-1.4 cm. longis tenuiter membranaceis viridi-purpurissatis extus glabris vel minutissime papillatis squamellis dentiforme-trigonalibus; corollae salverformis ex- tus glabrae tubo 1.8-2.3 em. longo basi ca. 0.4 cm. diametro metiente appendicibus epistaminalibus paulo exsertis ca. 0.5 em. longis annulo faucium conspicue incrassato lobis oblique obovatis breviter acuminatis 0.7-0.8 em. longis patentibus; antheris oblongo-sagittatis paulo panduratis 0.5-0.55 cm. longis dorso minute puberulo-papillatis paulo exsertis; ovario ovoideo ea. 0.2 em. longo glabro; stigmate 0.15 cm. longo; nectariis compresse ovoideis integris vel basi plus minusve con- natis ovario brevioribus; follieulus juventate gracilibus con- tinuis faleatis glabris; seminibus ignotis. Stems relatively slender, sparsely and minutely puberulent when young, glabrate and inconspicuously lenticellate when fully mature; leaves oblong-elliptic, apex shortly acuminate, base obtuse to rounded, 9-19 em. long, 2.7—7.5 em. broad, rigidly (489) [Vor. 23 318 ANNALS OF THE MISSOURI BOTANICAL GARDEN membranaceous to subcoriaceous, opaque or somewhat nitidu- lous above, glabrous; petioles 0.5-0.8 em. long; stipular ap- pendages interpetiolar, numerous, minutely dentiform; inflo- rescence corymbose, dichotomously subdivided, bearing 10—40 white, crimson-blotched flowers; peduncle much surpassing the subtended leaves; pedicels 1.5-2.0 em. long, somewhat accres- cent after maturity, minutely papillate; bracts linear, 0.1-0.2 em. long, scarious; calyx-lobes oblong,elliptic, shortly acumi- nate, 1.0-1.4 em. long, delicately membranaceous, greenish- purple, glabrous or very minutely papillate without, the in- ternal squamellae dentiform-trigonal; corolla salverform, gla- brous without, the tube 1.8-2.3 em. long, about 0.4 em. in di- ameter at the base, epistaminal appendages slightly exserted, about 0.5 em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.7-0.8 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers oblong-sagittate, slightly pandurate, 0.5-0.55 cm. long, minutely puberulent- papillate dorsally, slightly exserted; ovary ovoid, about 0.2 em. long, glabrous; stigma about 0.15 em. long; nectaries com- pressed-ovoid, separate or somewhat connate at the base, some- what shorter than the ovary; immature follicles relatively slender, continuous, faleate, glabrous; seeds unknown. COLOMBIA: MAGDALENA: edge of mountain forest and clearing below Valparaiso, alt. 4000 ft., March 26, 1899, H. H. Smith 1647 (NY, Type, G, MBG, photograph and analytical drawings); Bergenwald, Sierra Nevada de Santa Marta, J uly 9, 1926, Schultze 474 (B). The color of the corolla is described by Smith as “whitish blotched with crimson around the throat,” and by Schultze as “rotlich weiss, innen schneeweiss dunkelkarmin rot gezeichnet. Der Ring goldgelb.” 2T. Prestonia rotundifolia K. Sch. spec. nov. in herb. Suffrutieosa volubilis; ramulis juventate minute puberulis maturitate glabratis conspicue lenticellatis; foliis late ovalibus vel late obovato-ovalibus apice obtusis rotundatisve saepius minute mucronatis basi rotundatis 9.5-17.5 em. longis 4.5-10.0 cm. latis rigide membranaceis opacis glabris; petiolis 0.6-1.6 (490) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 319 em. longis glabris; appendicibus stipulaceis intrapetiolaribus multis minute dentiforme-flagelliformibus; inflorescentiis co- rymbosis subumbellatis obscure di- vel trichotome divisis flores 8-20 luteos gerentibus; pedunculo foliis multo breviore; pedicellis 1.1-1.2 em. longis minute appresseque puberulis; bracteis anguste elliptico-oblanceolatis 0.65-1.2 em. longis plus minusve petalaceis purpurissatis caducis; ealyeis laciniis oblongo-elliptieis breviter acuminatis 0.9-1.1 em. longis sub- coriaceis purpurissatis extus minutissime appresseque pu- berulo-papillatis squamellis dentiforme-deltoideis minutissime erosis; corollae salverformis extus minute papillatae tubo 1.5- 1.7 em. longo basi ea. 0.35 em. diametro metiente appendicibus epistaminalibus faucibus paene attingentibus ca. 0.25 em. longis annulo faucium conspicue inerassato lobis oblique ob- ovatis breviter acuminatis 1.0-1.8 em. longis patentibus; antheris anguste oblongo-sagittatis paulo panduratis 0.45—0.5 em. longis glabris paulo exsertis; ovario ovoideo ca. 0.2 em. longo glabro; stigmate 0.15 em. longo; nectariis compresse oblongo-ovoideis integris vel basi plus minusve connatis apice truneatis ovarium aequantibus vel paullulo superantibus; fol- liculis ignotis. Stems relatively stout, minutely puberulent when young, glabrate and conspicuously lenticellate when fully mature; leaves broadly oval to broadly obovate-oval, apex obtuse or rounded, frequently very shortly mueronulate, base rounded, 9.5-17.5 em. long, 4.5-10.0 em. broad, firmly membranaceous, opaque, glabrous; petioles 0.6-1.6 em. long, glabrous; stipular appendages intrapetiolar, numerous, dentiform-flagelliform; inflorescence corymbose or subumbelalte, obscurely di- or tri- chotomous, bearing 8-20 yellow flowers; peduncle much shorter than the subtending leaves; pedicels 1.1-1.2 cm. long, somewhat accrescent after maturity, minutely appressed-puberulent; bracts narrowly elliptic-oblanceolate, 0.65-1.2 cm. long, more or less petalaceous, purplish, caducous; calyx-lobes oblong- elliptic, shortly acuminate, 0.9-1.1 em. long, subcoriaceous, purplish, without minutely appressed puberulent, the internal squamellae dentiform-deltoid, very minutely erose; corolla salverform, minutely papillate without, the tube 1.5-1.7 cm. (491) [Vor. 23 320 ANNALS OF THE MISSOURI BOTANICAL GARDEN long, about 0.35 em. in diameter at the base, epistaminal ap- pendages barely attaining the faucal annulus, about 0.25 cm. long, faueal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 1.0-1.3 em. long, reflexed or spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers narrowly oblong-sagittate, slightly pandurate, 0.45-0.5 em. long, glabrous, slightly exserted; ovary ovoid, about 0.2 em. long, glabrous; stigma about 0.15 em. long; nectaries compressed oblong-ovoid, separate or slightly con- nate at the base, apex truncate, about equalling or very slightly surpassing the ovary; follicles unknown. ECUADOR: MANABI: in fruticetis siccis, El Reereo, Aug., 1893, Eggers 15078 (M, TYPE, MBG, photograph and analytical drawings); GUAYAS: Toachi, alt. 400—860 m., Dec., 1883, Sodiro 106/1 (B). Distinguished from P. trifida by the peculiar, petalaceous braets. 28. Prestonia robusta Rusby, Descr. So. Am. Pl. 91. 1920. Stems relatively stout, glabrous and conspicuously lenticel- late when fully mature; leaves broadly oval to obovate-oval, apex obtuse or rounded to very abruptly and shortly acuminate, base obtuse to rounded, 10-21 em. long, 5.5-12.5 em. broad, sub- coriaceous, opaque, glabrous; petioles 1.4-1.6 cm. long, gla- brous; stipular appendages interpetiolar, numerous, minutely dentiform; inflorescence corymbose or subumbellate, obscurely di- or trichotomously compound, bearing 10-20 greenish-white or yellowish flowers; peduncle much shorter than the subtend- ing leaves, glabrous to very sparsely puberulent; pedicels 0.4—0.9 em. long, somewhat accrescent after maturity, minutely puberulent-papillate; bracts oval to elliptic, 0.3-0.5 em. long, somewhat foliaceous or tinted with purple, caducous; calyx- lobes oval to broadly oblong-elliptic, acute to acuminate, 1.0— 1.5 em. long, coriaceous or subcoriaceous, somewhat suffused with purple, sparsely and minutely puberulent-papillate with- out, the internal squamellae dentiform-ligular, truncate or Ee erose; corolla salverform, glabrous or very minutely papillate without, the tube 1.3-1.5 em. long, about 0.3 em. in diameter at the base, epistaminal appendages slightly exserted, (492) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 321 0.25—0.3 em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.8-0.9 cm. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers narrowly sagittate, 0.55 em. long, glabrous or essentially so; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.12-0.15 cm. long; nectaries com- pressed-ovoid, separate or slightly connate at the base, about equalling the ovary; follicles unknown. BOLIVIA: LA PAZ: Tumupasa, alt. 1800 ft., Dec. 15, 1901, E. S. Williams 571 (NY, TYPE, MBG, photograph and analytical drawings); SANTA CRUZ: bosque, Buenavista, alt. 400 m., Dec. 30, 1925, Steinbach 7376 (B); Montecito de Fuca, Buenavista, alt. 450 m., Sept. 29, 1916, Steinbach 2863 (B). 29. Prestonia macroneura (Muell.-Arg.) Woodson, comb. nov. Haemadictyon macroneurum Muell.-Arg. in Mart. Fl. Bras. 61: 169. 1860; Miers, Apoc. So. Am. 262. 1878. Plants completely glabrous; stems relatively stout, conspicu- ously lenticellate when fully mature; leaves broadly oval, apex obtuse to very shortly acuminate, base obtuse to rounded, 12- 15 em. long, 7-9 em. broad, firmly membranaceous, opaque; petioles 0.5-0.8 em. long; stipular appendages intrapetiolar, numerous, minutely dentiform; inflorescence densely sub- umbellate-capitate, simple or very obscurely compound, bear- ing 15-20 congested, reddish flowers tinged with purple, yellow, and white; peduncle much shorter than the subtending leaves; pedicels 0.3-1.0 em. long, somewhat accrescent after maturity; bracts oval to oblong-lanceolate, acute to acuminate, 0.4—1.0 em. long, foliaceous, persistent; calyx-lobes oval to broadly elliptic, acuminate, 0.9-1.1 em. long, green slightly tinged with purple at the base, the internal squamellae deltoid, minutely lacerate; corolla salverform, glabrous or very minutely papil- late without, the tube 1.3-1.5 em. long, about 0.3 cm. in diameter at the base, epistaminal appendages slightly exserted, about 0.4 em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.7-0.9 cm. long, reflexed or widely spreading; stamens inserted at about the upper 1 of the corolla-tube, the anthers sagittate, about 0.5 cm. long, (493) [VoL. 23 322 ANNALS OF THE MISSOURI BOTANICAL GARDEN glabrous, slightly exserted; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.1-0.15 cm. long; nectaries compressed oblong-ovoid, separate or somewhat concrescent at the base, equalling or slightly surpassing the ovary; follicles unknown. BRAZIL: AMAZONAS: in sepibus ad Porto dos Juris prope Cataract. Capatenses, Dec., year lacking, Martius 3029 (M, TYPE, V, MBG, photograph and analytical drawings). An examination of Vellozo’s illustration of Echites denticu- lata (Icon. 3: pl. 30. 1830), a species with elongate inflores- cence and inconspicuous bracts, does not appear to ally that plant with Martius’ specimen, as indicated in Mueller’s syn- onymy. The description of the color of the corolla has been adapted from Mueller. 30. Prestonia trifida (Poepp.) Woodson, in Gleason & A. C. Smith, Bull. Torrey Bot. Club 60: 392. 1933. H aemadictyon trifidum Poepp. Nov. Gen. 3: 67. pl. 275. 1845; Miers, Apoc. So. Am. 261. 1878. NEC. (Haemadictyon) Evansii S. Moore, Trans. Linn. Soc. Bot. IT. 4: 395. 1895. Prestoma glabrata K. Sch. Verhandl. Bot. Ver. Branden- burg 47: 189. 1905, not HBK. Stems relatively stout, inconspicuously puberulent when very young, soon becoming glabrate and conspicuously lenticel- late; leaves broadly ovate to broadly oval, apex very abruptly and Boni. acuminate to obtuse, base broadly obtuse-rounded, 9-31 em. long, 4.5-14.5 em. broad, coriaceous or subcoriaceous, opaque or slightly nitidulous above, somewhat glaucescent be- neath, glabrous; petioles 0.8-2.5 em. long, glabrous; stipular appendages interpetiolar, numerous, minutely dentiform; in- florescence corymbose, tri- or rarely dichotomous, bearing 15- 30 yellow flowers; peduncle much shorter than the subtending leaves, very minutely puberulent to glabrate; pedicels 0.6—1.8 em. long, very minutely puberulent-papillate to glabrate, some- what accrescent after maturity; bracts very minute, broadly ovate-dentiform to ovate-lanceolate, scarious, persistent; calyx- lobes oblong- elliptic, acute to very shortly acuminate, 0.9-1.5 em. long, coriaceous, minutely appressed-puberulent to pu- (494) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV O29 berulent-papillate, more or less strongly suffused with purple, the internal squamellae deltoid, minutely erose or lacerate; eorolla salverform, glabrous to minutely puberulent-papillate without, the tube 1.5-1.8 em. long, about 0.35 em. in diameter at the base, epistaminal appendages slightly exserted, 0.2—0.4 em. long, faueal annulus eonspieuously thickened, the lobes obliquely obovate, shortly acuminate, 0.7—1.0 cm. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers oblong-sagittate, slightly pan- durate, 0.5-0.6 em. long, glabrous, slightly exserted; ovary ovoid, about 0.125 em. long, glabrous, stigma 0.1-0.15 cm. long; neetaries compressed ovoid, separate or more or less connate at the base, about equalling the ovary; follieles unknown. COLOMBIA: CHOCO: between La Oveja and Quibdo, April 1-2, 1931, Archer 1749 (US). PERU: LORETO: forest, Mishuyacu, near Iquitos, alt. 100 m., Jan., 1930, Klug 774 (FM, US); same locality, Febr.-March, 1930, Klug 915 (FM, US); Tarapoto, Dee., 1902, Ule 6604 (B, DL); silva non inundata, Rio Huallaga, Yurimaguas, Febr. 11, 1924, Kuhlmann a (B); Tarapoto, alt. 150 m., Dee., 1929, L. Williams 6049 (FM); Maynas, silvae primaevae, Febr., 1831, ta 2161 (V, TYPE, MBG, photo- graph and an a drawings). BRAZIL: MATTO GROSSO: ad ripas fl. Paraguay inter Santa Cruz et Villa Maria, Dec., 1891, Moore 819 (B, NY, MBG, photograph and analytical drawings); near source of the Jatuarana River, Machado River region, Krukoff 1545 (NY). 31. Prestonia vana Woodson, spec. nov. Frutieosa volubilis ut ereditur; ramulis gracilibus matur- itate glabris conspicue lenticellatis; foliis late ovalibus apice rotundatis breviter acuminato-mucronatis basi late obtusis rotundatisve 13-16 em. longis 8.0-9.5 em. latis coriaceis sub- coriaceisve glaberrimis opacis superne paulo lucentibus; pet- iolis 1.0-1.2 em. longis glabris; appendicibus stipulaceis intra- petiolaribus pectinatis numerosis; inflorescentiis corymbosis flores 20-25 speciosos viridi-lutescentes rubicundos gerentibus ; peduneuli dichotomi foliis bis terve brevioris partibus florigeris minute denseque puberulis sterilibus glabris; pedicellis 1.7—2.0 em. longis minute denseque puberulis; braeteis ovato-lance- olatis 0.2—0.3 em. longis scariaceis haud eaducis; calycis laciniis oblongo-elliptieis apice acute acuminatis 1.2-1.3 cm. longis subeoriaceis viridibus vel paulo purpurissatis extus minute (495) [Vor. 23 324 ANNALS OF THE MISSOURI BOTANICAL GARDEN sparseque puberulis intus glabris squamellis profunde irreg- ulariterque laeeratis; eorollae salverformis tubo 1.5-1.6 cm. longo basi ca. 0.25 em. diametro metiente extus minute velu- tino intus prope insertionem staminum molliter puberulo caeterumque glabro appendicibus epistaminalibus linearibus ca. 0.4—0.45 em. longis valde exsertis annulo faucium conspicue inerassato lobis oblique obovato-dolabriformibus conspicue acuminatis 0.7-0.8 cm. longis patentibus; antheris oblongo- sagittatis 0.5 em. longis dorso minute denseque puberulis apice exsertis; ovario ovoideo ca. 0.125 em. longo sparse minuteque pilosulo; stigmate subcapitato-maniculato ca. 0.15 em. longo; nectariis inerassatis compresse ovoideis ovarium paulo superantibus; folliculis ignotis. Stems somewhat slender, glabrous, conspicuously lenticel- late at maturity; leaves broadly oval, apex rounded, usually acuminate-mucronate, base broadly obtuse to rounded, 13-16 em. long, 8.0-9.5 em. broad, coriaceous to subcoriaceous, gla- brous, opaque, or the upper surface somewhat shining; petioles 1.0-1.2 em. long, glabrous; stipular appendages intrapetiolar, pectinate, numerous; inflorescence corymbose, bearing 20-25 showy, greenish-yellow, red-flushed flowers; peduncle twice to thrice shorter than the subtending leaves, the floriferous branches minutely and densely puberulent, the sterile axis glabrous; pedicels 1.7-2.0 cm. long, minutely and densely pu- berulent; bracts ovate-lanceolate, 0.2-0.3 em. long, scarious or very slightly foliaceous when very young, persistent; calyx- lobes oblong-elliptic, apex acutely acuminate, 1.2-1.3 em. long, subcoriaceous, green or very slightly suffused with purple, without minutely and rather sparsely puberulous, within gla- brous, the squamellae deeply and irregularly lacerate; corolla- tube 1.5-1.6 em. long, about 0.25 em. in diameter at the base, minutely velutinous without, softly puberulent within about the insertion of the stamens, epistaminal appendages linear, 0.4—0.45 em. long, conspicuously exserted, faucal annulus con- spicuously thickened, the lobes obliquely obovate-dolabriform, conspicuously acuminate, 0.7-0.8 cm. long, reflexed; anthers oblong-sagittate, 0.5 em. long, minutely and densely puberulent (496) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 325 dorsally, the tips slightly exserted; ovary ovoid, about 0.125 em. long, sparsely and minutely pilosulose; stigma subeapitate- maniculate, about 0.15 em. long; nectaries fleshy, essentially separate, compressed-ovoid, slightly surpassing the ovary; fol- licles unknown. PERU: LORETO: Balsapuerto, alt. about 220 m., forest, May, 1933, Klug 3066 (MBG, TYPE). This species is quite likely to be mistaken for P. trifida upon first sight. From that species, however, it differs in the looser inflorescence, longer pedicels, more delicate texture of the calyx-lobes, and more conspicuously exserted epistaminal ap- pendages, in addition to the pubescent flowers and anthers as indicated in the key to species. 32. Prestonia plumierifolia Mgf. Notizblatt 10: 1038. 1930. Stems relatively stout, minutely appressed-puberulent when young, glabrate and conspicuously lenticellate when fully mature; leaves obovate to obovate-lanceolate, apex obtuse or rounded, infrequently extremely abruptly and shortly acumi- nate to submucronate, base obtuse, cuneate, 8-16 cm. long, 3-6 cm. broad, subcoriaceous, opaque, or the upper surface slightly nitidulous, sparsely and minutely puberulent when very young, soon perfectly glabrate; petioles 0.5-1.5 em. long, minutely and rather sparsely puberulent-papillate to glabrate; stipular ap- pendages intrapetiolar, numerous, minutely and narrowly dentiform; inflorescence corymbose, dichotomous, bearing 8-30 greenish-yellow flowers; peduncle somewhat shorter than the subtending leaves; pedicels 0.4—1.0 em. long, somewhat accres- cent after maturity, minutely ferruginous-puberulent; bracts ovate-lanceolate, 0.1 cm. long or less, scarious; calyx-lobes oblong-elliptic, acute to acuminate, 0.6-0.8 em. long, subcori- aceous, minutely and sparsely puberulent-papillate to glabrate, conspicuously suffused with purple, the internal squamellae deltoid-dentiform, minutely erose; corolla salverform, glabrous or very minutely papillate without, the tube 1.2—1.5 cm. long, about 0.35 cm. in diameter at the base, epistaminal appendages barely exserted, 0.3 cm. long, faucal annulus conspicuously (497) [Vor. 23 326 ANNALS OF THE MISSOURI BOTANICAL GARDEN thickened, the lobes obliquely obovate, shortly acuminate, 0.7— 0.9 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers slightly ex- serted, narrowly sagittate, slightly pandurate, 0.5 em. long, glabrous; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.1-0.15 cm. long; nectaries compressed-ovoid, concrescent at the base, somewhat shorter than the ovary; follicles unknown. BRAZIL: in hortum paraénsem, introductu (prov. Rio Purus), Aug. 9, 1905, Huber 7030 (B, ISOTYPE, MBG, photograph and analytical drawings). This species is founded upon a cultivated specimen said to have been introduced from the region of the Rio Purus, State of Amazonas, Brazil. 33. Prestonia amazonica (Benth.) Maebr. Field Mus. Publ. Bot. 11: 34. 1931. Haemadictyon Amazonicum Benth. ex Muell.-Arg. in Mart. Fl. Bras. 6*: 166. 1860; Miers, Apoc. So. Am. 262. 1878. Stems relatively stout, minutely puberulent when young, be- coming glabrate and conspicuously lenticellate when fully ma- ture; leaves broadly oval, apex abruptly and shortly acuminate, base obtuse or rounded, 7-13 em. long, 3-7 em. broad, firmly membranaceous, opaque, minutely puberulent when young, glabrate when mature; petioles 0.8-1.1 cm. long, minutely puberulent to glabrate; stipular appendages intrapetiolar, nu- merous, minutely denticulate; inflorescence corymbose or sub- umbellate, simple or obscurely dichotomous, bearing 6-15 greenish-yellow flowers; peduncle somewhat shorter than the subtending leaves, minutely and sparsely puberulent to gla- brate; pedicels 1.5-1.7 cm. long, somewhat accrescent after maturity, minutely puberulent-papillate; bracts minutely ovate-lanceolate, 0.1—0.3 em. long, scarious; calyx-lobes ovate, acuminate, 0.5-0.6 em. long, subeoriaceous, suffused with purple, minutely papillate without, the internal squamellae very deeply lacerate; corolla salverform, densely papillate without, the tube 1.3-1.4 em. long, about 0.25 em. in diameter at the base, epistaminal appendages exserted, 0.4 em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, (498) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 327 0.7—0.8 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers slightly exserted, oblong-sagittate, 0.45-0.5 em. long, glabrous; ovary ovoid, about 0.1 em. long, minutely papillate; stigma 0.125 em. long; nectaries compressed-obovoid, concrescent at the base, about equalling the ovary; follicles unknown. BRAZIL: PARA: ad ripas fl. Trombetas et lacus Quiriquiry, Dec., 1849, Spruce 239 (M, TYPE, V, MBG, photograph and analytical drawings). 34. Prestonia Lindleyana Woodson, in Gleason & A. C. Smith, Bull. Torrey Bot. Club 60: 392. 1933. Haemadictyon calycinum Lindl. ex Miers, Apoc. So. Am. 259. 1878, not Muell.-Arg. Stems glabrous, relatively slender, not evidently lenticellate; leaves oblong- to obovate-elliptie, apex shortly acuminate, base obtuse to rounded, 10-16 em. long, 3-7 em. broad, firmly mem- branaceous to subcoriaceous, glabrous, pale green above, some- what glaucescent beneath; petioles 0.3-0.8 cm. long, glabrous; stipular appendages numerous, interpetiolar, minutely and narrowly dentiform-flagelliform; inflorescence densely race- mose, simple, bearing 10-20 yellowish flowers; peduncle some- what shorter than the subtending leaves; pedicels 0.5—1.0 em. long, minutely appressed-puberulent; bracts ovate or ovate- lanceolate, 0.1-0.2 em. long, slightly foliaceous; calyx-lobes ovate to ovate-oblong, acute to acuminate, 0.9—1.0 em. long, membranaceous, green, flushed with purple at the base, gla- brous, the internal squamellae broadly deltoid-trigonal, en- tire or slightly erose-undulate; corolla salverform, glabrous or very minutely papillate without, the tube 0.9-1.5 cm. long, about 0.25 em. in diameter at the base, epistaminal ap- pendages exserted, 0.25 em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.6— 0.9 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers slightly exserted, oblong-sagittate, slightly pandurate, 0.4-0.45 cm. long, glabrous; ovary ovoid, about 0.15 em. long, densely and minutely papillate; stigma 0.125-0.15 em. long; nectaries com- (499) [VoL. 23 328 ANNALS OF THE MISSOURI BOTANICAL GARDEN pressed obovoid, separate or slightly concrescent, equalling or slightly surpassing the ovary; follicles unknown. BRAZIL: AMAZONAS: prope Barra, Prov. Rio Negro, Oct., 1851, Spruce 1882 (B, Bx, Camb., TYPE, V, MBG, photograph and analytical drawings); Municipality Hu- mayta, near Livramento, on Rio Livramento, on varzea land, Oct. 12-Nov. 6, 1934, Krukoff 6763 (NY); MATTO GROSSO: on varzea land near river-shore, near Tabajaa, upper Machado River region, Nov.—Dec., 1931, Krukoff 1427 (MBG, NY). 35. Prestonia denticulata (Vell.) Woodson, comb. nov. Echites denticulata Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 30. 1827; A. DC. in DC. Prodr. 8: 455. 1844. Echites suberosa Vell. Fl. Flum. 111. 1830; Icon. 3: pl. 34. 1827; A. DC. loc. cit. 475. 1844. Haemadictyon Gaudichaudii A. DC. loc. cit. 426. 1844; Muell-Arg. in Mart. Fl. Bras. 6': 168. 1860; Miers, Apoc. So. Am. 256. 1878. Haemadictyon denticulatum (Vell.) Miers, loc. cit. 257. 1878. Haemadictyon ovatum Miers, loc. cit. 258. 1878. Prestonia Gaudichaudu (A. DC.) K. Sch. in Engl. € Prantl, Nat. Pflanzenfam. 4?: 188. 1895. Stems relatively slender, minutely appressed-puberulent when very young, becoming glabrate and rather inconspicu- ously lenticellate when fully mature; leaves oblong-elliptie to broadly oval, apex obtuse to shortly acuminate, base obtuse or rounded, 6-13 em. long, 3-6 em. broad, firmly membranaceous, opaque, puberulent when young, above glabrous, beneath mi- nutely and rather sparsely papillate when fully mature; peti- oles 0.2-1.1 em. long, minutely appressed-puberulent to gla- brate; stipular appendages intrapetiolar, numerous, narrowly dentiform ; inflorescence racemose, simple, bearing 12-30 yel- low flowers; peduncle about equalling or somewhat shorter than the subtending leaves, sparsely appressed-puberulent to glabrate; pedicels 0.4-1.2 cm. long, densely and minutely pu- berulent; bracts minutely ovate-lanceolate, 0.1—0.3 cm. long, searious; calyx-lobes broadly oblong- to ovate-elliptic, acute to shortly acuminate, 0.4—0.65 em. long, subcoriaceous to rigidly membranaceous, more or less suffused with purple, minutely puberulent to puberulent-papillate, the internal squamellae (500) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 329 deltoid-ligular, entire or very minutely erose; corolla salver- form, glabrous or very minutely papillate without, the tube 1.1-1.3 em. long, about 0.25-0.3 cm. in diameter at the base, epistaminal appendages somewhat exserted, 0.25-0.32 cm. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.45-0.65 cm. long, re- flexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers slightly exserted, sag- ittate, 0.4-0.55 em. long, glabrous; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.1-0.125 cm. long; nectaries com- pressed-ovoid, irregularly concrescent at tie base, equalling or slightly surpassing the ovary; follicles relatively long and slender, conspicuously moniliform, 20-39 em. long, glabrous or very sparsely and minutely puberulent-papillate; seeds 1.2— 1.3 em. long, the pale yellowish coma 1.5-1.9 em. long. AZIL: RIO DE JANEIRO: in collibus umbrosis inter frutices, vic. Rio de Janeiro, Li Gaudichaud 533 (B, DC, DL, MBG, photograph and analytical drawings); Rio de Janeiro, date lacking, Glaziou 12954 (B); same data, Glaziou 3728 (Bx); same data, 1872, Glaziow 4881 (B); Barra do Pirahy, April 13, 1926, Hoehne $ Gehrt 17319 (B); Morro de Sao Joao, Jan. 8, 1887, Schenck 1947 (B); Serra da Bica, Febr., 1897, Ule 4285 (B); in silvis prope Hortum Botanicum Rio de Janeiro, Febr. 13, 1916, Constantino 7787 (B); DATA INCOMPLETE: Sellow s.n. (B); Riedel s.n. (B, BB, G, U, V) ; Regel s.n. (Camb.) ; Forrest s.n. (Camb.). P. denticulata is one of the most uniform and distinctive species of the genus. Consequently Vellozo's epithets may be rather safely interpreted as pertaining to this species, par- ticularly since they refer to plants of Rio de Janeiro, where P. denticulata 1s apparently frequent and endemic. 36. Prestonia Meg'agros (Vell. Woodson, Ann. Mo. Bot. Gard. 21: 623. 1934. Echites Meg'agros Vell. Fl. Flum. 110. 1830; Icon. 3: pl. 33. 1827. Haemadictyon asperum Muell.-Arg. in Mart. Fl. Bras. 61: 169. 1860; Miers, Apoc. So. Am. 258. 1878. Haemadictyon megalagrion Muell.-Arg. loc. cit. 170. 1860. Prestoma megalagrion (Muell.-Arg.) Miers, loc. cit. 149. 1878. ? Prestonia laeta Miers, loc. cit. 1878. (501) [Vor. 23 330 ANNALS OF THE MISSOURI BOTANICAL GARDEN Stems relatively stout, densely ferruginous-hirtellous to gla- brate; leaves broadly oval to broadly obovate, apex obtuse or rounded, occasionally very shortly and abruptly acuminate- submucronate, 11-20 cm. long, 5-12 cm. broad, coriaceous or subeoriaceous, beneath opaque, densely ferruginous-puberu- lent to subtomentulose, above somewhat nitidulous, ferrugi- nous-puberulent to glabrate; petioles 1.0-1.4 em. long, ferrugi- nous-hirtellous ; stipular appendages intrapetiolar, numerous, minutely dentiform; inflorescence corymbose, di- or trichoto- mous, bearing 20-40 yellowish flowers; peduncle somewhat shorter than the subtending leaves, ferruginous-hirtellous ; pedicels 0.6-1.2 cm. long, somewhat accrescent after maturity, minutely ferruginous-hirtellous; bracts narrowly lanceolate, 0.4—0.7 em. long, scarious or somewhat petalaceous ; calyx-lobes oblong-lanceolate, acuminate, 1.0-1.6 em. long, minutely fer- ruginous-hirtellous without, subcoriaceous, deeply suffused with purple, the internal squamellae broadly deltoid, minutely emarginate or erose; corolla salverform, glabrous or very mi- nutely papillate without, the tube 1.3-1.4 cm. long, about 0.3 cm. in diameter at the base, epistaminal appendages slightly exserted, 0.3-0.35 cm. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 1.0— 1.2 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers slightly ex- - serted, oblong-sagittate, slightly pandurate, 0.55 em. long, gla- brous; ovary ovoid, about 0.15 em. long, minutely papillate ; stigma 0.15 em. long; nectaries compresse d-ovoid, more or less concrescent at the base, slightly shorter than the ovary; fol- licles unknown. BRAZIL: PARA: Obidos, silva non inundata, July 23, 1927, Ducke 21602 (B, 8); Obidos, silva secundaria, May 11, 1905, Ducke 21634 (B). These specimens compare very well with Vellozo’s illustra- tion, particularly with regard to the peculiar indument. 37. Prestonia obovata Standl. Jour. Wash. Acad. Sci. 15: 459. 1925. Stems relatively stout, glabrous, conspicuously lenticellate when fully mature; leaves obovate to obovate-oblong, apex ob- (502) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 331 tuse or rounded, occasionally abruptly and very shortly acu- minate-submucronate, base obtuse, somewhat cuneate, 9-16 cm. long, 4-8 em. broad, coriaceous or subcoriaceous, glabrous, above dark green and somewhat nitidulous, beneath paler and somewhat glaucescent; petioles 0.7—1.2 cm. long, glabrous, or very sparsely and minutely pilosulose; stipular appendages interpetiolar, numerous, minutely dentiform; inflorescence co- rymbose, dichotomous, bearing 10-30 purplish-yellow flowers; peduncle much shorter than the subtending leaves, glabrous; pedicels 1.1-1.5 em. long, minutely and sparsely pilosulose to glabrate; bracts minutely ovate-lanceolate, 0.15-0.2 cm. long, searious; calyx-lobes oblong-elliptic, acuminate, 1.0-1.2 em. long, coriaceous, minutely papillate to glabrate without, pur- plish, the internal squamellae deltoid-dentiform, minutely erose; corolla salverform, minutely papillate without, 1.0-1.3 em. long, about 0.35 em. in diameter at the base, epistaminal ap- pendages wholly included, 0.12—0.15 em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.7-1.0 em. long, reflexed or widely spreading; sta- mens inserted at about the upper 14 of the corolla-tube, the anthers narrowly sagittate, slightly pandurate, 0.5-0.55 cm. long, glabrous, slightly exserted; ovary ovoid, about 0.15 cm. long, glabrous; stigma 0.15 em. long; nectaries concrescent, rather thick and fleshy, entire or slightly undulate, somewhat surpassing the ovary; follicles unknown. PANAMA: CANAL ZONE: trail between Gamboa and Cruces, July 2, 1911, Pittier 3767 (G, US, TYPE, MBG, photograph and analytical drawings); Barbour-Lathrop Trail, Barro Colorado Island, Oct. 12, 1931, Shattuck 129 (MBG). 38. Prestonia concolor (S. F. Blake) Woodson, in Standl. & Record, Field Mus. Publ. Bot. 12: 327. 1936. Belandra concolor S. F. Blake, Contr. Gray Herb. 52: 78. 1917. Stems relatively slender, minutely and rather sparsely pilos- ulose when very young, glabrate and rather inconspicuously lenticellate when fully mature; leaves oblong-elliptic, apex shortly acuminate, base obtuse to rounded, 9-13 em. long, 3-5 em. broad, subcoriaceous, glabrous, pale green, above slightly (503) [Vor. 23 332 ANNALS OF THE MISSOURI BOTANICAL GARDEN nitidulous, beneath opaque; petioles 0.6-0.9 em. long, glabrous to minutely papillate; stipular appendages interpetiolar, nu- merous, very minutely dentiform-flagelliform; inflorescence racemose, simple, bearing 30-40 greenish-yellow (?) flowers; peduncle about equalling the subtending leaves, glabrous or minutely papillate ; pedicels 0.7-1.0 em. long, somewhat accres- cent after maturity, minutely papillate; bracts linear-lanceo- late, 0.05-0.07 em. long, scarious; calyx-lobes ovate-elliptic, acute to acuminate, 0.4—0.5 em. long, membranaceous or slightly subcoriaceous, deeply suffused with purple, minutely papillate without, the internal squamellae deltoid, minutely erose or lacerate; corolla salverform, glabrous or very minutely papil- late without, the tube 1.5-1.8 em. long, about 0.3 em. in diameter at the base, epistaminal appendages wholly included, 0.17—0.2 em. long, faueal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.8-1.0 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers slightly exserted, narrowly oblong-sagittate, 0.48-0.5 cm. long, glabrous or very minutely puberulent-papillate; ovary ovoid, about 0.125 em. long, gla- brous; stigma 0.1 em. long; nectaries concrescent, entire or slightly undulate, somewhat surpassing the ovary; follicles unknown. British HONDURAS: low banks of Rio Grande, March 25, 1907, Peck 953 (G, TYPE, MBG, photograph and analytical drawings). 39. Prestonia versicolor Woodson, spec. nov. Suffruticosa volubilis omnino glabra; ramulis crassiusculis maturitate conspicue lenticellatis; foliis ovalibus oblongo- ellipticisve apice abrupte breviterque acuminatis basi obtusis vel rotundatis 10-17 em. longis 5-9 em. latis rigide membrana- ceis opacis ; petiolis 1.3-2.1 em. longis ; appendicibus stipulaceis interpetiolaribus multis minutissime denticularibus; inflores- centiis subeorymbosis dichotome divisis; pedunculo foliis multo breviore; pedicellis 1.4-1.6 em. longis post maturi- tatem paulo accrescentibus glabris vel minutissime papillatis ; bracteis minutissimis; calycis laciniis oblongo-ellipticis acutis acuminatisve 0.8-1.0 em. longis membranaceis viridibus incon- (504) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 333 spicue purpurissatis squamellis denticulato-trigonalibus integ- ris; corollae salverformis tubo 1.2-1.6 cm. longo basi ca. 0.3 em. diametro metiente prope fauces paulo angustato appendicibus epistaminalibus omnino inclusis 0.15 em. longis annulo fau- cium conspicue incrassato lobis oblique obovatis breviter acu- minatis 0.7-0.9 em. longis patentibus; antheris paulo exsertis oblongo-sagittatis paululo panduratis 0.6 em. longis glabris; ovario ovoideo ea. 0.15 em. longo glabro; stigmate 0.15 cm. longo; nectariis conerescentibus margine leviter erosis ten- uibus haud incrassatis ovarium conspicue superantibus; fol- liculis ignotis. Plants completely glabrous ; stems relatively stout, conspicu- ously lenticellate when fully mature; leaves oval or oblong- elliptie, apex abruptly and shortly acuminate, base obtuse or rounded, 10-17 em. long, 5-9 em. broad, firmly membranaceous, opaque, petioles 1.3-2.1 em. long; stipular appendages inter- petiolar, numerous, very minutely dentiform; inflorescence subeorymbose, dichotomous; peduncle much shorter than the subtending leaves; pedicels 1.4-1.6 em. long, somewhat ac- crescent after maturity, glabrous or very minutely papillate ; bracts very minute, scarcely visible ; calyx-lobes oblong-elliptic, acute to acuminate, 0.8-1.0 em. long, membranaceous, green in- conspicuously flushed with purple, the internal squamellae dentiform-trigonal, entire; corolla salverform, glabrous with- out, the tube 1.2-1.6 em. long, about 0.3 em. in diameter at the base, slightly constricting toward the orifice, epistaminal ap- pendages wholly included, 0.15 em. long, faucal annulus con- spicuously thickened, the lobes obliquely obovate, shortly acu- minate, 0.7—0.9 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers slightly exserted, oblong-sagittate, slightly pandurate, 0.6 cm. long, glabrous; ovary ovoid, about 0.15 cm. long, glabrous; stigma 0.15 em. long ; nectaries completely concrescent, thin and more or less diaphanous, not fleshy, slightly erose, conspicu- ously surpassing the ovary; follicles unknown. PANAMA: COLON: Cana and vicinity, alt. 2000-6500 ft., April 17—June 8, 1908, R. S. Williams 940 (NY, TYPE, MBG, photograph and analytical drawings); Perme, April 23, 1933, Cooper 237 (NY). (505) [Vor. 23 334 ANNALS OF THE MISSOURI BOTANICAL GARDEN The color of the flowers is reported by Williams as ‘‘Tube pink, divisions greenish outside, purple lines within.’’ 40. Prestonia peregrina Woodson, spec. nov. Suffrutieosa volubilis; ramulis crassiusculis juventate mi- nute appresse-puberulis maturitate minute irregulariterque scabridulis conspicue lenticellatis ; foliis oblongo-elliptieis apice abrupte breviterque acuminatis basi obtusis 8-17 cm. longis 3-7 em. latis rigide membranaceis opacis omnino glabris ; peti- olis 0.9-2.0 em. longis minute sparseque puberulo-papillatis ; appendicibus stipulaceis interpetiolaribus multis minute denti- eulo-flagelliformibus ; inflorescentiis subeorymbosis dichotome subdivisis flores 30-60 albo-luteos gerentibus ; pedunculo folia superante glabro vel indistinete irregulariterque puberulo- papillato; pedicellis 0.7-1.5 cm. longis post maturitatemve paulo acerescentibus minute papillatis; bracteis ovato-lanceo- latis 0.1-0.15 em. longis seariaceis; calycis laciniis oblongo- ellipticis acuminatis 1.1-1.5 em. longis subcoriaceis saturate purpurissatis extus glabris squamellis deltoideis minute ero- sis; corollae salverformis extus glabrae vel minutissime papil- latae tubo 1.6-1.8 em. longo basi ca. 0.3 cm. diametro metiente appendicibus epistaminalibus omnino inclusis ca. 0.1 em. longis annulo faucium conspicue incrassato lobis oblique obovatis breviter acuminatis 0.7-0.9 em. longis patentibus; antheris paulo exsertis oblongo-sagittatis paululo panduratis 0.5-0.6 em. longis glabris vel minutissime papillatis; ovario ovoideo ca. 0.15 em. longo glabro; stigmate 0.125-0.15 em. longo; nec- tariis concrescentibus tenuibus haud incrassatis margine mi- nute erosis ovarium conspicue superantibus; follieulis gracil- limis continuis leviter falcatis apice saepius connatis 36-38 em. longis glabris; seminibus ignotis. Stems relatively stout, minutely appressed-puberulent when young, minutely and irregularly scabridulous and conspicu- ously lenticellate when fully mature; leaves oblong-elliptic, apex abruptly and shortly acuminate, base obtuse, 8-17 cm. long, 3-7 em. broad, firmly membranaceous, opaque, glabrous; petioles 0.9-2.0 cm. long, minutely and sparsely puberulent- papillate; stipular appendages interpetiolar, numerous, mi- (506) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 335 nutely dentiform-flagelliform; inflorescence subcorymbose, dichotomously subdivided, bearing 30-60 yellowish flowers; peduncle somewhat surpassing the subtending leaves, glabrous or indistinctly and irregularly puberulent-papillate; pedicels 0.7-1.5 em. long, or somewhat accrescent after maturity, mi- nutely papillate; bracts ovate-lanceolate, 0.1-0.15 cm. long, scariaceous; calyx-lobes oblong-elliptic, acuminate, 1.1-1.5 cm. long, subcoriaceous, deeply suffused with purple, glabrous, the internal squamellae deltoid, minutely erose; corolla salver- form, glabrous or very minutely papillate without, the tube 1.6— 1.8 em. long, about 0.3 em. in diameter at the base, epistaminal appendages wholly included, about 0.1 em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.7—0.9 em. long, reflexed or widely spreading; sta- mens inserted at about the upper 1⁄4 of the corolla-tube, the anthers slightly exserted, oblong-sagittate, slightly pandurate, 0.5-0.6 em. long, glabrous or very minutely papillate; ovary ovoid, about 0.15 cm. long, glabrous; stigma 0.125-0.15 cm. long; nectaries wholly concrescent, thin and somewhat diaph- anous, not fleshy, minutely erose, conspicuously surpassing the ovary; follicles relatively slender, continuous, somewhat fal- cate, the tips frequently connate, 36-38 cm. long, glabrous; seeds unknown. ECUADOR: GUAYAS: Balao, March, 1892, Eggers 14527 (M); PICHINCHA: Tan- dapi, ad marg. silvulae, July, 1920, Heilborn 771 (DL, G, S, TYPE, MBG, photograph and analytical drawings). The affinities of this species accentuate still further the inti- mate relationship of certain elements of the flora of Ecuador for the flora of Central America. 41. Prestonia vallis Woodson, spec. nov. Fruticosa vel suffruticosa volubilis; ramulis gracilibus gla- berrimis maturitate conspicue lenticellatis; foliis ovalibus apice breviter acuminatis basi late obtusis 10-14 cm. longis 4.5-6.5 cm. latis membranaceis glaberrimis utrinque opacis; petiolis 0.6-0.8 cm. longis glabris; appendicibus stipulaceis intrapetiolaribus minute pectinatis numerosis; inflorescentiis corymbosis dichotomis flores 20-25 speciosos brunneos gerenti- (507) [Vor. 23 336 ANNALS OF THE MISSOURI BOTANICAL GARDEN bus ; pedunculo petiolos bis terve superante minutissime puber- ulo haud ferrugineo; pedicellis 1.0-1.2 em. longis minutissime puberulo-papillatis; bracteis ovato-lanceolatis 0.15-0.25 cm. longis scariaceis; calycis laciniis ovato-lanceolatis acute acu- minatis 1.3-1.4 em. longis firmiter membranaceis foliaceis vel paululo purpurissatis glaberrimis squamellis deltoideis mi- nute irregulariterque erosis; corollae salverformis tubo 1.4— 1.5 em. longo basi ca. 0.18—0.2 em. diametro metiente extus gla- bro intus prope insertionem staminum molliter puberulo ap- pendicibus epistaminalibus oblongis 0.1-0,125 em. longis pro- funde inclusis annulo faucium conspicue inerassato lobis oblique obovatis 1.5-1.7 em. longis patentibus ; antheris lanceo- lato-sagittatis 0.6 cm. longis dorso minute puberulis apice paulo exsertis; ovario ovoideo 0.1 em. longo apice inconspicuissime barbellato; stigmate subcapitato-maniculato ca. 0.15 em. longo; nectariis basi concrescentibus ubique paululo incrassatis cae- terumque delicate membranaceis plus minusve foliaceis minute pilosulis ovarium ca. dimidio superantibus ; folliculis ignotis. Stems relatively slender, glabrous, conspicuously lenticel- late when fully mature; leaves oval, apex shortly acuminate, base broadly obtuse, 10-14 em. long, 4.5-6.5 em. broad, mem- branaceous, glabrous, opaque throughout; petioles 0.6-0.8 em. long, glabrous; stipular appendages intrapetiolar, minutely peetinate, numerous; inflorescence corymbose, dichotomous, bearing 20-25 showy brownish flowers; peduncle about twice to thrice surpassing the subtending petioles, minutely, but not ferruginously, puberulent; pedicels 1.0-1.2 em. long, minutely puberulent-papillate; bracts ovate-lanceolate, 0.15-0.25 em. long, scarious; calyx-lobes ovate-lanceolate, acutely acumi- nate, 1.3-1.4 cm. long, firmly membranaceous, foliaceous or very slightly suffused with purple, glabrous, the squamellae deltoid, minutely and irregularly erose; corolla salverform, the tube 1.4-1.5 em. long, about 0.18-0.2 em. in diameter at the base, glabrous without, softly pubescent within near the inser- tion of the stamens, epistaminal appendages oblong, 0.1-0.125 em. long, deeply included, faucal annulus conspicuously thick- ened, the lobes obliquely obovate, 1.5-1.7 em. long, reflexed ; (508) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV aot anthers lanceolate-sagittate, 0.6 em. long, minutely puberulent dorsally, the tips slightly exserted; ovary ovoid, about 0.1 em. long, the tips very inconspicuously barbellate, otherwise gla- brous; stigma subcapitate-maniculate, about 0.15 cm. long; nectaries somewhat surpassing the ovary, concrescent and somewhat incrassate at the base, otherwise membranaceous and more or less foliaceous, entire or essentially so save for the five component divisions, minutely and sparsely pilosulose ; follicles unknown. COLOMBIA: VALLE DEL CAUCA: Urwald, Candelaria, alt. 2200 m., Jan. 1, 1931, Dryander 1042 (B, TYPE, MBG, photograph and analytical iR d The foliaceous, puberulent nectary is the striking feature of this species, the membranaceous texture of which allies it to the species of $ Annulares indigenous to Central America. 42. Prestonia Schippii Woodson, spec. nov. Suffruticosa volubilis; ramulis gracilibus juventate minute puberulo-papillatis mox glabratis maturitate conspicue sed minute lenticellatis; foliis oblongo-ellipticis apice abrupte breviterque aeuminatis basi obtusis rotundatisve 10-17 em. longis 4-7 cm. latis rigidiuscule membranaceis opacis glabris; petiolis 1.2-1.6 em. longis glabris rariusve minutissime papil- latis; appendicibus stipulaceis intrapetiolaribus multis mi- nutissime denticulato-flagelliformibus; inflorescentiis subco- rymbosis simplicibus flores 10-12 gilvos gerentibus ; pedunculo foliis multo breviore glabro; pedicellis 1.2-1.5 cm. longis post maturitatem paulo accrescentibus glabris vel minutissime pap- illatis; bracteis ovato-lanceolatis 0.1-0.2 em. longis scariaceis vel rarius majoribus subfoliaceis; calycis laciniis obovato- ellipticis acutis 1.0-1.2 em. longis glabris subcoriaceis paululo purpurissatis squamellis dentieulato-trigonalibus integris; corollae salverformis extus minute papillatae tubo 1.3-1.4 em. longo basi ca. 0.3 em. diametro metiente appendicibus epistam- inalibus omnino inclusis ca. 0.3 em. longis annulo faucium con- spieue inerassato lobis oblique obovatis breviter acuminatis 0.9-1.0 em. longis patentibus ; antheris paulo exsertis oblongo- sagittatis 0.5 em. longis dorso minute hirtellis ; ovario ovoideo (509) [Vor. 23 338 ANNALS OF THE MISSOURI BOTANICAL GARDEN ca. 0.15 em. longo glabro; stigmate 0.125 em. longo; nectariis concrescentibus tenuibus haud inerassatis margine erosis ovarium conspicue superantibus ; folliculis ignotis. Stems relatively slender, minutely puberulent-papillate when young, soon becoming glabrate and conspicuously but minutely lenticellate when fully mature; leaves oblong-elliptie, apex abruptly and shortly acuminate, base obtuse or rounded, 10-17 em. long, 4-7 em. broad, firmly membranaceous, opaque, glabrous; petioles 1.2-1.6 em. long, glabrous or minutely pap- illate; stipular appendages intrapetiolar, numerous, very mi- nutely dentiform-flagelliform; inflorescence subcorymbose, simple, bearing 10-12 cream-colored flowers; peduncle much shorter than the subtending leaves, glabrous; pedicels 1.2-1.5 em. long, somewhat accrescent after maturity, glabrous or very minutely papillate; bracts ovate-lanceolate, 0.1-0.2 em. long, Sseariaceous, or rarely the lowermost somewhat larger and sub- foliaceous; calyx-lobes narrowly obovate-elliptic, acute, 1.0- 1.2 cm. long, glabrous, subcoriaceous, slightly purple-tinted, the internal squamellae denticulate-trigonal, entire ; corolla salver- form, minutely papillate without, the tube 1.3-1.4 em. long, about 0.3 cm. in diameter at the base, epistaminal appendages wholly included, about 0.3 em. long, faucal annulus conspic- uously thickened, the lobes obliquely obovate, shortly acumi- nate, 0.9-1.0 cm. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers oblong-sagittate, slightly pandurate, 0.5 cm. long, dorsally mi- nutely hirtellous, slightly exserted : ovary ovoid, about 0.15 em. long, glabrous; stigma 0.125 em. long; nectaries wholly con- erescent, thin and somewhat diaphanous, not fleshy, slightly erose, conspicuously surpassing the ovary; follicles unknown. BriTIsH HONDURAS: rare, climbing over low trees in forest, Eldorado, alt. 200 ft., Sept. 22, year lacking, Schipp S-388 (FM, TYPE, MBG, photograph and analytical drawings). The affinities of this species are outlined in the key to spe- cies. It is the second species of § Annulares to be reported from British Honduras, apparently the northernmost range of the section. (510) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 339 43, Prestonia guatemalensis Woodson, spec. nov. Fruticosa volubilis; ramulis sat crassiusculis teretibus gla- bris maturitate conspicue lenticellato-verrucosis; foliis brev- iter petiolatis late obovato-ellipticis apice breviter acuteque acuminatis basi obtuse cuneatis 18-25 cm. longis 8-10 em. latis firme membranaceis glaberrimis; petiolis 1.0-1.5 cm. longis; appendicibus stipulaceis interpetiolaribus anguste dentiformi- bus multis; inflorescentiis lateralibus bis terve dichotome di- visis plurifloris ; pedunculo post maturitatem 5.5—6.0 em. longo glabro lenticellato-verrucoso; pedicellis 0.9-1.2 em. longis mi- nute papillatis (?) ; calycis laciniis latiuscule ellipticis obtu- siusculis vel late acutis 0.7-0.9 em. longis subcoriaceis dilute viridibus ut ereditur haud purpurissatis extus intusque minute papillatis squamellis sat parvis acuminatis; corollae salver- formis ut creditur luteae tubo 1.8-2.0 em. longo basi ea. 0.2- 0.225 em. diametro metiente extus glabro intus prope inser- tionem staminum minute puberulo caeterumque glabriusculo appendicibus linearibus integris valde exsertis faucibus con- spicue calloso-incrassatis lobis oblique obovatis minute acumi- natis 1.2-1.3 em. longis extus minutissime papillatis intus gla- briusculis ; antheris valde exsertis 0.6 cm. longis glabris ; ovario ovoideo ca. 0.1 em. longo glabriusculo; stigmate 0.15 em. longo; nectariis concrescentibus membranaceis margine anguste lo- batis ca. 0.15 em. longis ovarium omnino superantibus ; follicu- lis teretibus falcatis 30-50 em. longis ca. 0.4—0.5 em. diametro metientibus apice saepe concrescentibus glabris; seminibus 1.5-1.8 em. longis como dilute luteo ca. 2.5 em. longo. Fruticose lianas; stems rather stout, terete, glabrous, warty- lenticellate when fully mature; leaves shortly petiolate, broadly obovate-elliptic, apex shortly and acutely acuminate, base obtusely cuneate, 18-25 cm. long, 8-10 em. broad, firmly membranaceous, glabrous throughout; petioles 1.0-1.5 cm. long; stipular appendages interpetiolar, numerous, narrowly dentiform ; inflorescence lateral, twice- to thrice-dichotomous, bearing several yellowish flowers; peduncle (somewhat after maturity) 5.5—6.0 em. long, glabrous, warty-lenticellate; pedi- cels 0.9-1.2 em. long, minutely papillate (?) ; calyx-lobes rather (511) [VoL. 23 340 ANNALS OF THE MISSOURI BOTANICAL GARDEN broadly elliptic, obtusish to rather broadly acute, 0.7-0.9 cm. long, subcoriaceous, pale green, evidently not suffused with purple, minutely papillate without and within, the squamellae relatively small, acuminate; corolla salverform, the tube 1.8— 2.0 em. long, about 0.2-0.225 em. in diameter at the base, gla- brous without, minutely puberulent within near the attachment of the stamens, otherwise glabrous, the internal appendages linear, entire, somewhat exserted, the orifice callose-annulate, the lobes obliquely obovate, minutely acuminate, 1.2-1.3 em. long, very minutely papillate without, essentially glabrous within; anthers exserted, 0.6 cm. long, glabrous; ovary ovoid, about 0.1 em. long, essentially glabrous; stigma 0.15 cm. long; nectaries concrescent, membranaceous, the margin narrowly and minutely lobed, about 0.15 em. long, completely concealing the ovary; follicles terete, falcate, usually united at the tips, 30-50 cm. long, about 0.4-0.5 em. in diameter, glabrous; seeds 1.5-1.8 em. long, the pale yellow coma about 2.5 cm. long. GUATEMALA: Sepacuite, Oct., 1901, Owen 1 (US, TYPE, MBG, photograph); SANTA ROSA: Volcan Tecuamburro, alt. 2000 m., Febr., 1893, Heyde $ Lux 4539 (G). Closely related to P. portobellensis, differing chiefly in the calyx-lobes, squamellae, and more extensive inflorescence. Heyde £ Lux 4539 is unfortunately long past prime and rather fragmentary; its relegation to this species is therefore some- what uncertain. 44, Prestonia portobellensis (Beurl.) Woodson, Ann. Mo. Bot. Gard. 18: 553. 1931. Echites portobellensis Beurl. Vet. Akad. Handl. Stockh. 137. 1854 (1856). Haemadictyon schizadenium Muell.-Arg. Linnaea 30: 431. 1860; Miers, Apoc. So. Am. 261. 1878. Prestonia schizadenia (Muell.-Arg.) Hemsl. Biol. Centr.- Am. Bot. 2: 312. 1881. Prestonia (Haemadictyon) macrocarpa Hemsl. loc. cit. 311. 1881. Stems relatively stout, minutely scabridulous in the vicinity of the nodes, conspicuously lenticellate when fully mature; leaves oblong-elliptie, apex obtuse to abruptly and shortly acu- (512) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 341 minate, base obtuse or rounded, 9-29 cm. long, 3-18 em. broad, firmly membranaceous to subcoriaceous, glabrous, opaque, or slightly nitidulous above; petioles 0.6—3.0 em. long; stipular ap- pendages interpetiolar, numerous, minutely dentiform; inflo- rescence subcorymbose, di- or trichotomous, less frequently simple, bearing 8-30 purplish-yellow flowers; peduncle much shorter than the subtending leaves, glabrous; pedicels 0.6—1.8 em. long, somewhat accrescent after maturity, glabrous or very minutely papillate; bracts ovate-lanceolate, 0.05—0.4 em. long, scarious or slightly foliaceous ; calyx-lobes oblong-elliptic, acu- minate, 1.1-1.7 em. long, coriaceous or subcoriaceous, more or less flushed with purplish, glabrous or very minutely and indis- tinetly papillate, the squamellae dentiform-deltoid, very mi- nutely erose; corolla salverform, glabrous or very minutely papillate without, the tube 1.1—1.7 em. long, about 0.3-0.35 cm. in diameter at the base, epistaminal appendages slightly ex- serted or at least attaining the faucal annulus, 0.2—0.4 cm. long, faueal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 1.0-1.5 em. long, reflexed or widely spreading; stamens inserted at about the upper 1⁄4 of the co- rolla-tube, the anthers slightly exserted, narrowly oblong- sagittate, slightly pandurate, 0.5-0.6 em. long, puberulent-pap- illate to minutely papillate dorsally; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.15 em. long; nectaries wholly con- crescent, thin and somewhat diaphanous, not thick and fleshy, minutely erose or lacerate, conspicuously surpassing the ovary ; follicles relatively long and stout, 33-35 em. long, con- tinuous, somewhat falcate, often united at the tips, glabrous; seeds 1.5-1.6 em. long, the pale yellowish coma 2.5-3.7 em. long. MEXICO: OAXACA: vicinity of Cafetal Concordia, alt. 400-650 m., April 1-15, 1933, Morton $ Makrinus 2348 (US). HONDURAS: ATLANTIDA: beside the trail, above Lancetilla, alt. 200 ft., July 15, 1934, Yuncker 4596 (Herb. Univ. Mich.). SALVADOR: SONSONATE: vicinity of Sonsonate, alt. 220-300 m., March 18-27, 1922, Standley 22349 (G). Costa Rica: Limon: Moin Hill, prés Limon, June, 1898, Pittier 12401 (G, V); Pinta prés Limon, litt. Atlantique, Sept., 1899, Pittier 16016 (B, G) ; PUNTARENAS: forêts de Santo Domingo de Golfo Dulce, March, 1896, Pittier 9935 (BB, Bx); en- virons de Sto. Domingo de Osa, March, 1896, Tonduz 9889 (Bx, V). (513) VoL. 23 342 ANNALS OF THE MISSOURI BOTANICAL GARDEN AMA: COLON: Portobello, in silvis ad littora, April, 1826, Billberg s.n. (S, eu photograph and analytieal drawings); Santa Rita trail, Febr. 27, 1905, Hisce 105 (NY); Chagres, Jan.—March, 1850, Fendler 250 (MBG, V); Gatun Sta., Jan. 28, 1860, Hayes 450 (NY). Remarks on this species and its nomenclature will be found in Ann. Mo. Bot. Gard. 18: 553-554. 1931. The species is not a very uniform nor constant one as interpreted above, and may require segregation at a future date. 45. Prestonia lacerata Woodson, spec. nov. Fruticosa volubilis altitudine ignota ; ramulis gracilibus gla- berrimis maturitate conspicue lenticellatis; foliis oblongo- ellipticis apice abrupte acuminato-mucronatis basi late obtusis 12-20 em. longis 5.5-9.5 em. latis firmiter membranaceis sub- coriaceisve glaberrimis supra sublutescentibus subtus opacis; petiolis 1.0-1.5 em. longis; appendicibus stipulaceis inter- petiolaribus numerosis minute dentiformibus vix bene visis; inflorescentiis corymbosis dichotome divisis flores gilvos 30-40 gerentibus; pedunculo foliis ea. dimidio breviore mi- nutissime puberulo; pedicellis 1.2-1.3 em. longis post maturi- tatem paulo accrescentibus minute ferrugineo-puberulis ; brac- teis acuminato-deltoideis scariaceis vix 0.1 em. longis; calycis laciniis oblongo-lanceolatis acute acuminatis 1.0-1.1 em. longis membranaceis foliaceis post exsiccationem paulo purpuris- satis sparse minuteque pilosulis glabratisve squamellis sub- trigonalibus acutis integris ; corollae salverformis tubo 1.5 em. longo basi ca. 0.18-0.2 cm. diametro metiente extus sparse mi- nuteque puberulo-papillato intus prope insertionem staminum molliter puberulo appendicibus epistaminalibus linearibus ca. 0.35 em. longis paulo exsertis annulo faucium conspicue incras- sato lobis oblique obovato-dolabriformibus 0.9 cm. longis patentibus ; antheris oblongo-sagittatis 0.5 em. longis dorso mi- nute puberulis apice exsertis ; ovario ovoideo ca. 0.15 em. longo glabro; stigmate subcapitato-fusiforme 0.15 em. longo; nec- tariis basi concrescentibus ubique aliquid inerassatis caete- rumque delicate membranaceis purpurissatisque hyalinis mar- gine profunde irregulariterque laceratis ovarium ca. dimidio superantibus ; follieulis gracilibus continuis falcatis 30-45 cm. (514) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 343 longis glaberrimis; seminibus immaturis 1.8-2.0 em. longis como dilute aurantiaco ca. 4 em. aequante. Stems relatively slender, glabrous, conspicuously lenticel- late at maturity; leaves oblong-elliptic, apex abruptly acumi- nate-mucronate, base broadly obtuse, 12-20 em. long, 5.5-9.5 em. broad, firmly membranaceous to subcoriaceous, glabrous, somewhat lustrous above, opaque beneath ; petioles 1.0—1.5 em. long; stipular appendages interpetiolar, numerous, minutely dentiform; inflorescence corymbose, dichotomously divided, bearing 30-40 pale yellowish flowers; peduncle about half as long as the subtending leaves, very minutely ferruginous-pu- berulent ; pedicels 1.2-1.2 em. long, somewhat accrescent at ma- turity, minutely ferruginous-puberulent; bracts acuminate- deltoid, searious, less than 0.1 em. long; calyx-lobes oblong- lanceolate, acutely acuminate, 1.0-1.1 cm. long, membrana- ceous, foliaceous (somewhat suffused with purple in desicca- tion), sparsely and minutely pilosulose to glabrate without, the squamellae subtrigonal, acute, entire; corolla salverform, the tube 1.5 em. long, about 0.18-0.2 em. in diameter at the base, sparsely and minutely puberulent-papillate without, softly pu- berulent near the insertion of the stamens within, epistaminal appendages linear, about 0.35 cm. long, slightly exserted, fau- eal annulus conspicuously thickened, the lobes obliquely obo- vate-dolabriform, 0.9 em. long, reflexed; anthers oblong-sag- ittate, 0.5 cm. long, minutely puberulent dorsally, the tips slightly exserted; ovary ovoid, about 0.15 cm. long, glabrous; stigma subcapitate-fusiform, 0.15 em. long; nectaries about twice surpassing the ovary, concrescent and somewhat incras- sate at the base, membranaceous and purplish-hyaline above, deeply and irregularly lacerate; follicles slender, continuous, falcate, 30-45 cm. long, glabrous; seeds (immature) 1.8-2.0 em. long, the pale tawny coma about 4 em. long. PERU: LORETO: Regenwald, Iquitos, alt. 100 m., May 12, 1925, Tessmann 5106 (B, TYPE, MBG, photograph and analytical drawings). Unique among the South American Prestonias because of the subpetalaceous nectary of the flowers, by which it simulates the Central American species of $ Annulares. (515) [Vor. 23 344 ANNALS OF THE MISSOURI BOTANICAL GARDEN Sect. 4. roMENTOSAE Woodson. Relatively stout, usually densely ferruginous-pubescent lianas; leaves membranaceous to subcoriaceous; inflorescence simple or variously compound, typically bostrychoid-racemose to subumbellate; calyx-lobes relatively large and conspicuously foliaceous; corolla salver- form, rarely infundibuliform, more or less densely pubescent without, appendiculate or exappendiculate within, the epistam- inal appendages occasionally replaced by callous ridges, the faucal annulus conspicuously thickened or tuberculate ; anthers included or the tips more or less exserted. Spp. 46-60. This is the most difficult section of Prestonia. Since the de- limitation of the species appears rather vague and confusing in certain instances, in spite of the relatively large number of herbarium specimens available for study, an endeavor has been made to maintain conservatively the specific limits recognized by previous botanists. Subsequent study, particularly if sup- plemented by extensive field observation, may well differ in this regard. It is scarcely to be avoided that our present conceptions of taxonomic units in the Apocynaceae, as in other tropical American groups, should be other than purely elemental. In several instances in the delimitation of the species to fol- low, the color of the indument has appeared significant. Here the difficulty of ascribing intelligibly the color to desiccated specimens has been considerable, and doubtless will be found unsatisfactorily anticipated in many instances. It should be borne in mind that the color of the indument, described as such, refers in all instances to the collective color of the trichomes themselves, and not, for instance, to the gross shade of the indument viewed in conjunction with a variable background of leaf- or stem-surface, which would necessarily vary greatly according to the relative density of the trichomes and the con- sequently varying proportion of the vegetative ground tissue visible. KEY TO THE SPECIES A. Corolla-tube appendiculate within. B. Epistaminal appendages exserted, or at least attaining the faucal annulus. (516) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 345 C. Inflorescence dichotomous, occasionally rather obscurely so; foliar indument very finely velutinous, umber-brown, opaque.......... CC. Inflorescence simple, or rarely very obscurely compound. x-lobes as long as the corolla-tube, or nearly so; inflorescence many-flow E. Anthers ae? F. Foliar indument dull yellowish- or buffy-brown, opaque, or searcely lustrous; calyx tomentulose, subappressed; plants of northern Colombia to south-central Brazil an aguay AE OUO SE dO o STEEN GE 7. P. tomentosa FF. Foliar indument dark orange-brown, lustrous; d hispid- hirsute to -hirsutulose. G. Foliar indument velutinous; calyx hispid-hirsutulose; plants theast-central Brazil................. 48. P. bahiensis GG. Foliar Gre hispid to hispidulose, much sparser than in the preceding; calyx hispid-hirsute ; He. "i Panama to Erench3Quistig.. EE e . P. ipomaeif olia EE. t gii Eo at least puberulent-papillate de is . Corolla-tube 1.6-2.0 em. long, oer -villosulose without, the a em. long; nectaries surpassing the ovary; plants of southern Brazil and goth Paraguay.50. P. calycina FF. Corolla-tube 2.0-2.2 em. long, minutely puberulent without, the lobes 2.0-2.2 cm. a nectaries about half ny the ; plants of Perl... -tuur UEM Ee . P. cordifolia DD. Calyx- joa osi half as c as the corolla-tube; "vhi dh nce Tiw-HOWENEOd Lr ioo E e A 52. P. brachypoda BB. T al appendages deeply included. C. Inflorescence simple, or essentially so, subumbellate; corolla 2.0-2.3 ded A og em. long; anthers barely included.................. mucronata CG: pl repeatedly dichotomous, corymbose or thyrsiform; .0—1.3 em. long; anther-tips exserted......... o. F. persino AA. Gorilla: Sa princi within, or with callous ridges or protuber- ances in the position of epistaminal appenda B. Follicles Abe "ni and stout, napiform, Ped divaricated; species of Mexico and Centrai America. C. Corolla poate uai the tube not dilating above the insertion of the stamens, or scarcely so. D. Nectary annular, broadly 5-lobed, or ia entire, about equal- ng or barely surpassing the ov E. Corolla 3.5—4.0 em. long; cite protuberances mee 0.2- 0.25 em. long; plants of southern Mexico and Guatemala.. EE. Corolla 2.8-3.0 em. long; epistaminal protuberances nearly quad- rate, 0.1—0.15 em. long; plants of British Ho we "nor A UE e EE . P. amanuensis DD. Nectary tubular or gé eng E 5-lobed, Geen sur- passing the ovary; corolla long; plan f Costa Ee be. M crc Ania Nr AO r^s P. isthmica [Vor. 23 346 ANNALS OF THE MISSOURI BOTANICAL GARDEN BB. Follieles relatively long and slender, not divaricate; species of South America. CC. Inflorescence congested, subumbellate; anther-tips exserted, glabrous 0. P. Schumanniana $8 hoo s o9 9 9 9 9 9 d 9 o9 9 o3 à 9 9 à 9 5 9 $0 39 5 2 8 8 8] ££ £ | | ]] £2: n À | c 9 n9 n 9 46. Prestonia surinamensis Muell-Arg. Linnaea 30: 433. 1860; Miers, Apoc. So. Am. 147. 1878. Stems relatively stout, densely and minutely ferruginous- tomentose, eventually becoming glabrate; leaves broadly ovate to ovate-elliptie, apex shortly and abruptly acuminate, base obtuse to rounded, 10-24 em. long, 6-17 em. broad, firmly mem- branaceous, uniformly and very densely velutinous when young, densely and very finely velutinous beneath, above mi- nutely scabridulous to glabrate generally, pilosulose along the midrib and veins when fully mature, the indument dull umber- brown, opaque, not lustrous; petioles 0.7-3.5 em. long, densely ferruginous-pubescent; stipular appendages interpetiolar, nu- merous, pectinate-flagelliform, more or less ferruginous-pilosu- lose; inflorescence corymbose, dichotomous, occasionally rather obscurely so, bearing 10-40 whitish-yellow flowers; peduncle minutely ferruginous-velutinous, much shorter than the sub- tending leaves; pedicels 0.4—0.7 cm. long, somewhat accrescent after maturity, minutely ferruginous-velutinous; bracts ovate- lanceolate, somewhat foliaceous, 0.6-0.8 cm. long; calyx-lobes broadly ovate- to oblong-elliptic, shortly acuminate to acute, 0.8-1.2 em. long, foliaceous, minutely appressed ferruginous- velutinous, the internal squamellae broadly dentiform-ligular, emarginate, somewhat pilosulose; corolla salverform, densely ferruginous-villosulose without, the tube 1.5-1.8 em. long, about 0.3 em. in diameter at the base, epistaminal appendages slightly exserted or at least attaining the orifice, 0.2-0.4 em. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.6-0.9 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers barely exserted, oblong-sagit- (518) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 347 tate, 0.6-0.7 cm. long, glabrous; ovary ovoid, about 0.15 cm. long, glabrous or essentially so; stigma 0.1-0.15 cm. long; nectaries concrescent, rather irregularly 5-cleft or essentially entire, fleshy, conspicuously surpassing the ovary; follicles very stout and rigid, narrowly napiform, sharply divaricate, 9-11 em. long, minutely ferruginous-velutinous to glabrate; seeds 1.4-1.7 em. long, the pale tawny coma 2.0-3.2 em. long. BRITISH GUIANA: upper Rupununi River, near Dadanawa, May 30, 1922, Cruz 1419 (MBG, NY); same locality, June 10, 1922, Cruz 1489 (MBG, NY). DuTCH GuIANA: Plantatio Beekhuizen, date lacking, Wullschlágel 1029 (Bx, V); exact locality and date lacking, Hostmann 981 (B, U, V, TYPE, MBG, photograph and analytical drawings); fluv. Tapanahoni, Aug., 1904, Versteeg 753 (U); fluv. Gonimi, Aug., 1903, Versteeg 72 (U). BRAZIL: PARA: insula Mexiana, in fauce Amazonum fluvii, Sept. 20, 1901, Guedes 21640 (B); Rio Branco de Obidos, ad ripas fluminis, Jan. 28, 1918, Ducke 21628 (B) ; Ourem, Rio Guaura, Dec. 4, 1903, Siqueira 21633 (B). 47. Prestonia tomentosa R. Br. Mem. Wern. Soc. 1: 70. 1811; A. DC. in DC. Prodr. 8: 429. 1844; Muell.-Arg. in Mart. Fl. Bras. 6!: 163. 1860; Miers, Apoc. So. Am. 144. pl. 20B. 1818. Prestonia latifolia Benth. in Hook. Jour. Bot. 3: 250. 1841; A. DC. loc. cit. 429. 1844; Miers, loc. cit. 145. 1878. Prestonia lutescens Muell.-Arg. loc. cit. 164. 1860; Miers, loc. cit. 147. 1878. Prestonia lanata Muell.-Arg. loc. cit. 1860; Miers, loc. cit. 1878. Prestonia Cearensis Miers, loc. cit. 148. 1878. Prestonia sericocalyx Malme, Bihang till K. Sv. Vet. Akad. Handl. Afd. IIT. 24°: 29. 1899. Stems relatively stout, densely and rather pale ferruginous- tomentose, eventually becoming glabrate; leaves broadly ovate, apex shortly acuminate, base broadly obtuse to rounded, infre- quently very obscurely cordate, 8-19 em. long, 5-12 em. broad, firmly membranaceous, above rather sparsely and minutely ferruginous-hirtellous, more densely so along the midrib and veins, beneath densely and rather closely velutinous-tomentose, the indument dull yellowish- or buffy-brown, opaque, not lus- trous; petioles 0.3-1.2 cm. long, minutely ferruginous-tomen- tose as upon the stem; stipular appendages interpetiolar, (519) [VoL. 23 348 ANNALS OF THE MISSOURI BOTANICAL GARDEN pectinate-flagelliform, somewhat ferruginous-pilosulose at the base; inflorescence subumbellate, simple, bearing 10-30 waxy, yellow flowers; peduncle much shorter than the subtending leaves, pale ferruginous-tomentulose; pedicels 0.4—1.0 em. long, somewhat accrescent after maturity, minutely ferruginous- tomentulose as upon the peduncle; bracts ovate-lanceolate, 0.8-1.8 cm. long, slightly foliaceous; calyx-lobes oblong- to ovate-elliptic, acute to acuminate, 1.0-1.8 em. long, foliaceous, subappressed-tomentulose, the internal squamellae broadly dentiform, entire or very minutely erose; corolla salverform, densely and appressed ferruginous-villous without, the tube 1.5-1.8 em. long, about 0.3-0.35 em. in diameter at the base, epistaminal appendages barely exserted, or about attaining the orifice, 0.15-0.4 em. long, faucal annulus conspicuously thickened or tuberculate, the lobes obliquely obovate, shortly acuminate, 0.7-1.2 em. long, reflexed or widely spreading; sta- mens inserted at about midway or the upper 14 of the corolla- tube, the anthers barely exserted, sagittate, 0.5-0.6 em. long, glabrous; ovary ovoid, about 0.15 em. long, minutely puber- ulent-papillate; stigma 0.05-0.1 em. long; nectaries conerescent, broadly and rather shallowly 5-lobed, fleshy, conspicuously sur- passing the ovary; follicles stout and rigid, narrowly napiform, sharply divaricate, 6-9 em. long, densely ferruginous-hispid; seeds 0.8-0.9 em. long, the pale tawny coma 3.5-4.2 em. long. COLOMBIA: MAGDALENA: growing over a shrub on open, low grassland e a forest- lined stream, near Jordan, alt. 800 ft., Nov. 5, 1898, H. H. Smith 165 (N VENEZUELA: MER a few e s. of Guigue, alt. 2000 ft., ie 5 1855, Fendler 1052 (G); EUM alt. 450 m., Sept., 1899, Preuss 1617 (B). BRITISH GUIANA: savannah, Pirara, ni. Zeng Schomburgk 755 (B, Camb., K, NY, V); Pirara, 1841-2, Schomburgk 374 (DL, V BRAZIL: AMAZONAS: bei der Serra do Stel., Rio Branco; Surumu, Nov., 1909, Ule 8266 (B, DL) ; am Igarape Imelu bei Pracana, Rio Branco, Sur ODE 1909, Ule 7939 (B, DL); Boa Vista, Rio Branco super., Ss Geier July 1913 Kuhlmann 3646 (B) ; MATTO GROSSO: Coxipo, prope Cuyaba, in margine silvula non in **cerrado,”” Dec. 27, 1893, Malme 1276B (DL, S) ; MINAS GERAES: exact locality and date lacking, Claussen s.n. (DL); RIO DE JANEIRO: Palmita, date lacking, Pohl 2214 (Bx, V); exact Mom ES date lacking, Glaziou 8800 (Bx); Rio de Janeiro, 1910, EE s.n. (B, M); SAO PAULO: Rincão, Jan. 25, 1928, Toledo 23550 (B); DATA INCOMPLETE: Pohl s.n. (Bx) ; Claussen 339 (B, BB); Glaziou 19625 (B). PERU: LORETO: Tarapoto, Dec., 1902, Ule 6650 (B, DL). (530) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 349 BOLIVIA: LA PAZ: Ixiamas, alt. 800 ft., Dec. 17, 1921, Cardefias 1928 (NY); SANTA CRUZ: im Wald zwischen Rio Pirai und Rio Cuchi bei Santa Cruz, Jan., 1911, Herzog 1514 (B, S, V). PARAGUAY: collines herbeuses entre le Cerro-Hu et le Cerro San Tomas, prés Paraguari, Jan. 31, 1877, Balansa 1376 (Bx, DL); in capueras Caaguazu, date lack- ing, Jórgensen 4708 (MBG); Villa Concepcion, Febr., 1896, Anisits s.n. (S) ;Cor- dillera de Altos, Cerro Chochi, Dee. 1902, Fiebrig 640 (B, DL); Wald, Cordillera de Altos, Dec., 1902, Fiebrig 647 (B) ; zwischen Rio Apa und Rio Aquidaban, Centurion, Jan., 1909, Fiebrig 4462 (B); in regione lacus Ypacaray, Febr., 1913, Hassler 11547 (B, DL, MBG); in regione cursus superioris fluminis Apa, Nov., 1901, Hassler 7820 (B, BB, V). 48. Prestonia bahiensis Muell.-Arg. in Mart. Fl. Bras. 6': 164. 1860; Miers, Apoc. So. Am. 147. 1878. Stems relatively stout, densely ferruginous-tomentose, even- tually becoming glabrate; leaves broadly oval to oblong- elliptic, apex acuminate to obtuse, base very broadly obtuse or rounded, occasionally very obscurely cordate, 5-16 cm. long, 3.0-7.5 em. broad, firmly membranaceous, above ferruginously hispid-hirtellous, beneath densely velutinous, the indument dark orange-brown, very lustrous; petioles 0.3-1.0 em. long, indument as upon the stem; stipular appendages intrapetiolar, several, pectinate-flagelliform, ferruginous-pilosulose, rela- tively ineonspieuous; infloreseenee densely subumbellate, simple, bearing 10-30 yellowish flowers; peduncle much shorter than the subtending leaves, indument as on the stem; pedicels 0.4-1.0 em. long, somewhat accrescent after maturity, fer- ruginous-tomentulose; bracts linear-lanceolate, 0.7-1.2 em. long, slightly foliaceous; calyx-lobes ovate-lanceolate, acu- minate, densely hispid-hirsutulose, foliaceous, the internal squamellae dentiform-ligular, entire or very minutely erose, minutely pilosulose; corolla salverform, densely ferruginous- villosulose without, the tube 1.5-1.8 em. long, about 0.3 em. in diameter at the base, epistaminal appendages slightly exserted, 0.25—0.3 em. long, faueal annulus eonspieuously thickened or tubereulate, the lobes obliquely obovate, shortly acuminate, 0.7-0.9 em. long; stamens inserted somewhat above midway within the corolla-tube, the anthers barely exserted, oblong- sagittate, 0.55-0.6 em. long, glabrous; ovary ovoid, about 0.15 em. long, papillate or essentially glabrous; stigma 0.1-0.15 em. (521) [Vor. 23 350 ANNALS OF THE MISSOURI BOTANICAL GARDEN long; nectaries concrescent, more or less deeply lobed or cleft, fleshy, eonspieuously surpassing the ovary; follieles unknown. BRAZIL: BAHIA: S. Thome, June, 1844, Blanchet 3776 (BB, Bx, TYPE, DC, MBG, photograph and analytical drawings) ; MINAS GERAES: Lagóa Santa, 1870, Warm- ing s.n. (DC); inter Ponte do Sapueahy et Retero, Jan. 23, 1868, Regnell III 1600 (B, 8); Campo Corrego do Leitáo, Bello Horizonte, Aug., year lacking, Gehrt 3183 (B); Sabara, Jan., 1916, Hoehne 6874 (B) ; exact locality and date lacking, Claus- sen s.n. (V); SAO PAULO: Magy-Mirim, March 20, 1874, Mosen 1461 (S); DATA IN- COMPLETE: Sellow 1406 (B). 49. Prestonia ipomaeifolia A. DC. in DC. Prodr. 8: 429. 1844; Miers, Apoc. So. Am. 145. 1878. Prestonia Seemanni Miers, loc. cit. 146. 1878. Stems relatively slender, somewhat laxly ferruginous- pilose; leaves broadly elliptie to oval, apex very abruptly and shortly acuminate, base broadly obtuse or rounded, 9-14 em. long, 9-8 em. broad, firmly membranaceous, above rather sparsely and uniformly hispid-hirtellous, beneath somewhat more densely hispid to hispidulose, the indument dark orange- brown, lustrous; petioles 0.4-0.6 em. long, indument as on the stem; stipular appendages intrapetiolar, numerous, narrowly pectinate-flagelliform, ferruginous -pilose at the base, relatively conspicuous; inflorescence densely subumbellate, simple, bear- ing 10-30 yellowish flowers; peduncle much shorter than the subtending leaves, indument as on the stem; pedicels 0.4-0.7 em. long, somewhat accrescent after maturity, ferruginous- hirtellous; bracts ovate-lanceolate to linear-lanceolate, 0.7—0.9 em. long, somewhat foliaceous; calyx-lobes ovate-lanceolate, acuminate, 0.9-1.1 em. long, somewhat foliaceous, 0.9-1.1 cm. long, hispid-hirsute, the internal squamellae broadly dentiform, minutely erose; corolla salverform, densely ferruginous-villous without, the tube 1.7-1.8 em. long, about 0.3 em. in diameter at the base, epistaminal appendages barely exserted, or about at- taining the orifice, 0.15 em. long, faucal annulus conspicuously thickened or tuberculate, the lobes obliquely obovate, shortly acuminate, 1.2-1.4 cm. long, reflexed or widely spreading; sta- mens inserted at about the upper 14 of the corolla-tube, the anthers slightly exserted, oblong-sagittate, 0.6 em. long, gla- brous; ovary ovoid, about 0.15 em. long, essentially glabrous; (522) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 351 stigma 0.1 cm. long; nectaries concrescent, tubular, barely 5- lobed, fleshy, conspicuously surpassing the ovary; follicles unknown. PANAMA: PANAMA: Camino del Boticario, near Chepo, alt. 30-50 m., Oct., 1911, Pittier 4700 (G, US). COLOMBIA: MAGDALENA: Sta. Isabel de Hungaria bei Tucurinca, alt. ea. 150 m., Oet. 16, 1926, Schultze 605 (B). FRENCH GUIANA: Cayenne, date lacking, le Blond s.n. (DC, TYPE). 50. Prestonia calycina Muell.-Arg. in Mart. Fl. Bras. 67: 162. 1860; Miers, Apoc. So. Am. 146. 1878, not Lindl. Prestonia hirsuta Muell.-Arg. loc. cit. pl. 48. 1860; Miers, loc. cit. 147. 1878, not Spreng. Stems relatively stout, coarsely ferruginous-hirsute to gla- brate; leaves broadly oval to oblong-elliptic, apex very abruptly and shortly aeuminate to obtuse or rounded, base broadly obtuse or rounded to obscurely cordate, 9-17 em. long, 6-13 em. broad, firmly membranaceous, above appressed fer- ruginous-hispidulous to seabridulous, occasionally essentially elabrate, beneath densely ferruginous-tomentulose, rarely glabrate; petioles 0.5-1.1 cm. long, ferruginous-tomentulose to elabrate; stipular appendages interpetiolar, numerous, rather broadly pectinate; inflorescence subumbellate, simple, bearing 6-18 yellowish flowers; peduncle much shorter than the sub- tending leaves, indument as upon the stem; pedicels 0.3-1.0 em. long, ferruginous-tomentulose to glabrate; bracts ovate to ovate-lanceolate, 0.7—1.4 em. long, foliaceous; calyx-lobes ovate- to oblong-lanceolate, acuminate, 1.2-1.8 em. long, foliaceous, ferruginous-hirtellous, the internal squamellae irregularly erose or lacerate, sparsely pilosulose; corolla salverform, densely and appressed ferruginous-villosulose, the tube 1.6-2.0 em. long, about 0.3-0.35 cm. in diameter at the base, epistaminal appendages barely exserted, or about attaining the orifice, 0.25—0.4 em. long, faucal annulus conspicuously thickened or tubereulate, the lobes obliquely obovate, shortly acuminate, 1.2-1.5 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers barely exserted, oblong-sagittate, 0.5-0.7 cm. long, more or less pu- berulent dorsally, rarely nearly glabrate; ovary ovoid, about (523) VoL. 23 352 ANNALS OF THE MISSOURI BOTANICAL GARDEN 0.15 em. long, glabrous; stigma 0.1—0.15 em. long; nectaries con- crescent, broadly tubular, more or less deeply 5-lobed, fleshy, conspicuously surpassing the ovary; follicles stout and rigid, narrowly napiform, sharply divaricate, 6.5-7.0 em. long, re- trorsely hirtellous to obsoletely scabridulous; seeds 1.1-1.2 em. long, the pale tawny coma 3.0-3.5 em. long. BRAZIL: RIO DE JANEIRO: Cantogallo, 1861, Peckholt 559 (Bx); exact locality lacking, Beyrich s.n. (B); Glaziou 12944 (B); Glaziou 8173 (B); DATA INCOM- PLETE: Pohl 5167 (V, MBG, photograph and analytical drawings) ; Sellow 550 (B); Glaziow 18363 (B). PARAGUAY: in regione cursus superioris fl. Y-aca, Dec., 1900, Hassler 6776 (B, BB, V). Hassler 6776 is more nearly glabrate than the majority of the Brazilian representatives, but is surpassed in this respect by Glaziou 18363, from an unspecified locality in Brazil. The specimens cited appear conspecific in all essential respects. 51. Prestonia cordifolia Woodson, spec. nov. Ut creditur suffruticosa vel fruticosa volubilis; ramulis sat gracilibus dense ferrugineque puberulo-tomentulosis; foliis op- positis manifeste petiolatis ovatis vel ovato-ellipticis apice abrupte acuminatis basi conspicue cordatis superne late obtusis rotundatisve 6-12 em. longis 4-9 cm. latis firmiter mem- branaceis supra minute scabridulo-puberulis subtus dense ferrugineo-tomentulosis; petiolis 1.3-1.5 em. longis ut in ramulo vestitis; appendicibus stipulaeeis interpetiolaribus inconspicue dentiformibus utroque latere 3-4; inflorescentiis simplicibus corymbosis foliis multo brevioribus flores (ut cred- itur aut gilvos aut flavos) 8-10 gerentibus; pedunculo dense ferrugineo-tomentuloso petiolis subaequante; pedicellis 1.0— 1.2 em. longis ut in pedunculo vestitis; bracteis oblongo- lanceolatis 1.0-1.5 cm. longis foliaceis minute ferrugineo- puberulis; ealycis laciniis ovato-oblongis abrupte acuminatis basi usque 14 altitudinem connatis 2.0-2.2 em. longis extus leviter intusque minute ferrugineo-puberulis squamellis in- conspieuis late triangulo-deltiformibus integerrimis apice ob- tusis ca. 0.05 em. longis minutissime puberulis; corollae salver- formis tubo 2.0-2.2 em. longo basi ca. 0.18-0.2 em. diametro (524) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 353 metiente faucibus 0.25-0.3 cm. diametro metientibus extus omnino minute ferrugineo-puberulo intus supra insertionem staminum retrorse pilosulo appendieibus epistaminalibus ob- longis 0.45-0.5 em. longis conspicue exsertis annulo faucium manifeste inerassato, lobis oblique obovato-oblongis 2.0-2.2 em. longis patentibus extus minute ferrugineo-puberulis intus glabris; staminibus prope fauces insertis antheris conspicue exsertis lanceolato-sagittatis 0.7—0.75 em. longis dorso minute hispidulis ; stigmate oblongo-subeapitato 0.17 em. longo; ovario ovoido ca. 0.2 em. longo glabro; nectariis concrescentibus haud conspicue lobatis incrassatis ovarium ca. dimidio aequantibus ; folliculis ignotis. Stems relatively slender, densely ferruginous-tomentulose ; leaves opposite, petiolate, ovate to ovate-elliptic, apex abruptly acuminate, base conspicuously cordate to broadly obtuse or rounded above, 6-12 cm. long, 4—9 em. broad, m mem- branaceous, above minutely scabridulous-puberulous, beneath densely rin ENN petioles 1.3-1.5 em. long, mi- nutely ferruginous-tomentulose; stipular appendages inter- petiolar, inconspicuously dentiform, 3-4 upon either side of the node; inflorescence simple, corymbose, much shorter than the leaves, bearing 8-10 (probably yellowish or cream-colored) flowers; peduncle densely ferruginous-tomentulose, about equalling the subtending petioles; pedicels 1.0-1.2 cm. long, the indument as upon the peduncle; bracts conspicuously foli- aceous, oblong-lanceolate, 1.0-1.5 em. long, very minutely fer- ruginous-puberulous; calyx-lobes ovate-oblong, abruptly acu- minate, connate at the base for about 14 their length, 2.0-2.2 em. long, conspicuously foliaceous, without softly and densely, within minutely ferruginous-puberulous, the squamellae broadly triangular-deltiform, entire, the apex broadly obtuse, about 0.05 em. long, very minutely puberulent; corolla salver- form, the tube 2.0-2.2 em. long, about 0.18-0.2 cm in diameter at the base and 0.25-0.3 em. in diameter at the orifice, minutely ferruginous-hirtellous or -puberulous without, retrorsely pilosulose within above the insertion of the stamens, epi- staminal appendages oblong, 0.45-0.5 cm. long, conspicuously (525) [Vor. 23 354 ANNALS OF THE MISSOURI BOTANICAL GARDEN exserted, faucal annulus conspicuously thickened; lobes obliquely obovate-oblong, 2.0-2.2 em. long, reflexed, without minutely ferruginous-puberulent, within glabrous; stamens inserted near the orifice of the corolla-tube, the anthers con- spieuously exserted, lanceolate-sagittate, 0.7—0.75 em. long, mi- nutely hispidulous dorsally; ovary ovoid, about 0.2 em. long, glabrous; stigma oblong-subcapitate, 0.17 em. long; nectaries conerescent, scarcely lobed, greatly thickened, about half equalling the ovary ; follieles unknown. PERU: CAJAMARCA: Catache, Prov. Contumaza, alt. 5000 ft., May 27, 1875, Raimondi 8228 (B, TYPE, MBG, photograph and analytieal drawings). Closely related to P. calycina Muell.-Arg., but differing, in addition to the key characters, in such features as the vege- tative induments, stipular appendages, length and degree of connation of calyx-lobes, and character of the squamellae. 92. Prestonia brachypoda Blake, Contr. U. S. Nat. Herb. 20: 530. 1924. Stems relatively stout, ferruginous-velutinous when young, soon becoming scabridulous to glabrate; leaves broadly ob- ovate-oblong, apex obtuse to rounded, base obscurely cordate, 14-25 em. long, 9-12 em. broad, firmly membranaceous, above sparsely ferruginous-pilosulose, beneath densely and minutely ferruginous-velutinous; petioles 0.3-0.5 em. long, very minutely ferruginous-vellutinous; stipular appendages interpetiolar, numerous, pectinate-flagelliform, sparsely and minutely pilosu- lose; inflorescence densely subumbellate-capitate, simple, bear- ing 3-6 yellowish flowers; peduncle extremely short, scarcely surpassing the subtending petioles, ferruginous-pubescent; pedicels 0.7-0.8 cm. long, minutely appressed-puberulent; bracts linear-lanceolate, long-acuminate, 0.4-0.6 em. long, slightly foliaceous; calyx-lobes ovate-lanceolate, acuminate, 0.8-0.9 em. long, somewhat foliaceous, minutely appressed- puberulent, the internal squamellae broadly dentiform-ligular, entire or very minutely erose or emarginate; corolla salver- form, appressed ferruginous-villosulose without, the tube 1.5— 1.7 em. long, about 0.25 em. in diameter at the base, epistaminal (526) 1936] : WOODSON—STUDIES IN THE APOCYNACEAE. IV 309 appendages slightly exserted, 0.3-0.35 em. long, faucal an- nulus conspicuously thickened, the lobes obliquely obovate, shortly acuminate, 0.6-0.7 cm. long, reflexed or widely spread- ing; stamens inserted at about the upper 14 of the corolla-tube, the anthers barely exserted, oblong-sagittate, 0.5 cm. long, gla- brous; ovary ovoid, about 0.15 em. long, glabrous; stigma 0.1- 0.15 em. long; nectaries concrescent, somewhat fleshy, tubular or subtubular, broadly 5-lobed, conspicuously surpassing the ovary; follicles unknown. VENEZUELA: CARABOBO: Guaremales, road from Puerto Cabello to San Felipe, in forest, alt. 10-100 m., May 15-29, 1920, Pittier 8832 (G, US, TYPE, MBG, photo- graph and analytical drawings). 53. Prestonia mucronata Rusby, Descr. So. Am. Pl. 90. 1920. Stems relatively slender, densely and minutely ferruginous- tomentose, eventually becoming glabrate when fully mature; leaves broadly oval, apex very abruptly and shortly acuminate, base broadly obtuse or rounded, 4-8 em. long, 3-5 em. broad, firmly membranaceous, above rather sparsely and minutely ferruginous-hispidulose, beneath minutely and rather sparsely ferruginous-hirtellous; petioles 0.7-1.5 em. long, minutely ferruginous-tomentulose; stipular appendages interpetiolar, numerous, minutely pectinate-flagelliform, very minutely fer- ruginous-pilosulose at the base; inflorescence subumbellate, simple, bearing 4-12 yellowish flowers, peduncle minutely ferruginous-tomentulose, somewhat shorter than the subtend- ing leaves; pedicels 1.0-1.2 em. long, somewhat accrescent after maturity, minutely ferruginous-tomentulose; bracts nar- rowly lanceolate, 0.2-0.5 em. long; calyx-lobes oblong-elliptic, acute to acuminate, 0.9-1.2 em. long, somewhat foliaceous, minutely appressed ferruginous-puberulent, the internal squa- mellae broadly deltoid-dentiform, very minutely erose, mi- nutely puberulent-papillate; corolla salverform, minutely fer- ruginous-puberulent without, the tube 0.9-1.2 em. long, about 0.2 em. in diameter at the base, dilated to about 0.4 em. at the orifice, epistaminal appendages wholly included, 0.1-0.15 em. long, faucal annulus conspicuously thickened or tuberculate, the lobes obliquely obovate, shortly acuminate, 0.9-1.0 cm. (527) [Vor. 23 356 ANNALS OF THE MISSOURI BOTANICAL GARDEN long, reflexed or widely spreading; stamens inserted slightly above midway within the corolla-tube, the anthers barely in- cluded, very narrowly sagittate, 1.0-1.1 em. long, glabrous; ovary ovoid, about 0.1 cm. long, glabrous or essentially so; stigma 0.2 cm. long; nectaries more or less concrescent at the base, compressed-oblongoid, truneate or slightly erose at the tips, more or less fleshy, conspicuously surpassing the ovary; follieles unknown. COLOMBIA: MAGDALENA: occasional in damp, somewhat open places in the forest and clearings near streams, alt. 4500—6000 ft., Las Nubes, Dee. 18, 1898, H. H. Smith 1656 (DL, MBG, NY, TYPE, S). 94. Prestonia parviflora Benth. in Benth. € Hook. Gen. Pl. 2: 709. 1876. Haemadictyon parviflorum Benth. Pl. Hartw. 355. 1857. Temnadema parviflora (Benth.) Miers, Apoc. So. Am. 215. 1878. Stems relatively slender, densely ferruginous-tomentulose, eventually becoming glabrate; leaves rather broadly elliptic, apex shortly acuminate, base broadly obtuse, 7-13 cm. long, 2.5-6.0 em. broad, firmly membranaceous, above rather sparsely and minutely hispid-hirtellous, more densely along the midrib and veins, beneath densely and ferruginously sericeous- velutinous; petioles 0.8-1.5 em. long, densely ferruginous- tomentulose; stipular appendages interpetiolar, numerous, pectinate-flagelliform, minutely pilosulose below; inflorescence corymbose to thyrsiform, repeatedly dichotomous, bearing many small, yellowish flowers; peduncle minutely ferruginous- tomentulose, somewhat shorter than the subtending leaves; pedicels 0.4—0.6 em. long, somewhat accrescent after maturity, minutely ferruginous-tomentulose; bracts minutely linear- lanceolate, 0.1-0.3 cm. long; calyx-lobes ovate-lanceolate, acu- minate, 0.3-0.4 em. long, densely and minutely ferruginous- puberulent, the internal squamellae dentiform, minutely pecti- nate; corolla salverform, minutely and appressed ferruginous- villosulose without, the tube 0.5-0.6 cm. long, about 0.1 em. in diameter at the base, somewhat dilated above the insertion of the stamens, epistaminal appendages wholly included, about (528) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV Sot 0.1 cm. long, faucal annulus conspicuously thickened, the lobes obliquely obovate, very shortly acuminate, 0.6-0.7 cm. long, re- flexed or widely spreading; stamens inserted about midway within the corolla-tube, the anthers slightly exserted, sagittate, 0.35 em. long, glabrous; ovary ovoid, about 0.1 cm. long, mi- nutely papillate; stigma 0.05 em. long; nectaries imperfectly concrescent, rather thin, somewhat shorter than the ovary; follicles unknown. COLOMBIA: CUNDINAMARCA: Pandi, date lacking, Hartweg 1053 (Camb., K, TYPE); Cena, Prov. de Bogota, alt. 1300 m., 1851-57, Triana s.n. (Bx, MBG, photo- graph and analytical drawings); DATA INCOMPLETE: Triana 1979 (B). 55. Prestonia mexicana A. DC. in DC. Prodr. 8: 429. 1844; K. Sch. in Engl. & Prantl, Nat Pflanzenfam. 4?: 188. 1895. Haemadictyon Meaicanum A. DC. loc. cit. 428. 1844. Prestonia sericea Mart. € Gal. Bull. Acad. Roy. Brux. 11!: 360. 1844. Haemadictyon contortum Mart. & Gal. loc. cit. 1844. Mitozus Mexicanus (A. DC.) Miers, Apoc. So. Am. 225. 1878. Exothostemon sericeum (Mart. E Gal.) Miers, loc. cit. 241. 1878. Exothostemon contortum (Mart. € Gal.) Miers, loc. cit. 1878. Stems relatively stout, densely ferruginous-tomentose, even- tually becoming glabrate; leaves broadly ovate to oval or ob- ovate-elliptie, apex acute to very abruptly and shortly acumi- nate, base rounded to obscurely cordate, 7-23 em. long, 4-15 em. broad, firmly membranaceous, above ferrugi ly hispid- hirtellous to scabridulous upon older specimens, beneath densely ferruginous-tomentose; petioles 0.3—0.9 em. long, fer- ruginous-tomentulose; stipular appendages intrapetiolar, nu- merous, pectinate-flagelliform, ferruginous-tomentulose be- low; inflorescence closely subumbellate, simple, bearing 8-20 yellowish flowers; peduncle much shorter than the subtending leaves, indument as upon the stem; pedicels 0.5-1.0 em. long, ferruginous-tomentulose; bracts ovate-lanceolate, acuminate, 0.6-1.7 cm. long, foliaceous; calyx-lobes ovate- to oblong- (529) [Vor. 23 358 ANNALS OF THE MISSOURI BOTANICAL GARDEN lanceolate, acute to acuminate, 1.2-2.3 em. long, foliaceous, ap- pressed-tomentulose, the internal squamellae broadly denti- form, minutely erose to essentially entire; corolla salverform, densely ferruginous-villosulose without, the tube 2.2-3.0 em. long, about 0.3-0.35 em. in diameter at the base, epistaminal appendages replaced by linear callous ridges 0.2-0.25 em. long, faueal annulus conspicuously thickened or tuberculate, the lobes obliquely obovate, shortly acuminate, 1.3-1.5 em. long, reflexed or widely spreading; stamens inserted at about the upper 14 of the corolla-tube, the anthers slightly exserted, oblong-sagittate, 0.6-0.8 em. long, glabrous; ovary ovoid, about 0.15 em. long, minutely papillate or essentially glabrous; nec- taries conerescent, annular, broadly 5-lobed or essentially en- tire, about equalling or slightly surpassing the ovary; follicles stout and rigid, sharply divaricate, narrowly napiform, 6-12 em. long, densely and rigidly ferruginous-hispid; seeds 1.0-1.2 em. long, the pale tawny coma 2.5-3.0 cm. long. MEXICO: COLIMA: Manzanillo, Dee. 1-31, 1890, E. Palmer 1028 (B, G); MORE- LOS: barranca near Cuernavaca, alt. 4700 ft., June 17, 1896, Pringle 6341 (B, Bx, BB, DL, G, M, MBG, V); same locality, alt. on 00 ft., Nov. 13, 1895, Pringle 6224 (B, Bx, BB, DL, G, M, MBG, V); VERA CRUZ: Valle de Don Oct. 15, 1865, Bourgeau 1500 (B, S); Wartenberg, near Tantoyuea, 1858, Ervendberg 127 (DC); Zaeuapan, June, 1916, Purpus 7665 (MBG); open forests, Zacuapan, 1928, Purpus 11134 (MBG); oaxaca: Ojitlan, alt. 700 ft., Aug. 31, 1895, L. C. Smith 661 (G); San Bartolo, Aug., year lacking, Andrieux 251 (M); DATA INCOMPLETE: Schiede 488 Sd Pavon s.n. (BB); Andrieux 398 (DC). GUATEMALA: SANTA ROSA: Naranjo, alt. 1100 m., May, 1893, Heyde $ Lux 4496 (B, m M, MBG); DATA INCOMPLETE: Warszewicz s.n. (B Haemadictyon Mexicanum A. DC., based upon a tracing of a plant of Mociño & Hesse, diverges from typical P. mexicana only in the absence of the heavy indument characteristic of the latter species. This absence is considered to be an omission rather than a positive indieation of glabrism, since the thick napiform-fusiform follicles illustrated for Mocifio & Sesse's plant are only found in a relatively small group of species always bearing a characteristic ferruginous indument. Upon the basis of a wide representation of the Mexican flora, only one species of Prestonia with thick follicles is known, P. mexi- cana; consequently H. Mexicanum has been placed in synonymy (530) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 359 under it. The peculiar conditions under which the tracings of Mociño & Sesse's plants were prepared would also support the supposition of an accidental omission of the indument. 56. Prestonia amanuensis Woodson, spec. nov. Suffruticosa volubilis; ramulis crassiusculis ferrugineo-to- mentosis; foliis late ovatis apice abrupte brevissimeque acu- minatis basi late obscureque cordatis 6-9 em. longis 4-7 cm. latis rigide membranaceis supra dense ferrugineo-hispidulis subtus densius ferrugineo-tomentulosis; petiolis 0.2-0.3 em. longis ut in ramulo vestitis; appendicibus stipulaceis intra- petiolaribus multis minute pectinatis; inflorescentiis dense sub- umbellatis simplicibus flores gilvos 6-12 gerentibus; pedunculo foliis multo breviore ut in ramulo vestito; pedicellis 0.3-0.5 em. longis post maturitatem paulo acerescentibus ferrugineo- tomentulosis; bracteis ovato-lanceolatis acuminatis 0.3-0.7 em. longis foliaceis; calycis laciniis ovato-lanceolatis acuminatis 1.1-1.3 em. longis foliaceis minute appresseque hirtellis squa- mellis dentiformibus minute emarginatis erosisve; corollae salverformis extus dense appresseque ferrugineo-villosulae tubo 2.0-2.2 em. longo basi ea. 0.3 em. diametro metiente intus plieas subquadratas callosas ea. 0.1-0.15 em. longas pro ap- pendieibus epistaminalibus gerente faucibus callosis lobis oblique obovatis breviter acuminatis 0.8-0.9 cm. longis pa- tentibus; staminibus prope fauces insertis antheris paululo exsertis sagittatis 0.6 em. longis glabris ; ovario ovoideo ca. 0.14 em. longo glabriusculo; stigmate 0.2 em. longo; nectariis con- crescentibus annularibus subintegris ovario subaequantibus ; folliculis ignotis. Stems relatively stout, ferruginous-tomentose; leaves broadly ovate, apex abruptly and shortly acuminate, base broadly and obscurely cordate, 6-9 cm. long, 4—7 cm. broad, firmly membranaceous, above densely ferruginous-hispidulous, beneath densely ferruginous-tomentulose; petioles 0.2-0.3 cm. long, indument as upon the stem; stipular appendages intra- petiolar, numerous, minutely pectinate; inflorescence densely subumbellate, simple, bearing 6-12 cream-colored flowers; peduncle much shorter than the subtending leaves, ferrugi- (531) [Vor. 23 360 ANNALS OF THE MISSOURI BOTANICAL GARDEN nous-tomentose; pedicels 0.3-0.5 em. long, somewhat accrescent after maturity, ferruginous-tomentulose; braets ovate-lance- olate, acuminate, 0.3-0.7 em. long, foliaceous; calyx-lobes ovate- lanceolate, acuminate, 1.1-1.3 em. long, foliaceous, minutely appressed-hirtellous, the internal squamellae dentiform, mi- nutely emarginate or erose; corolla salverform, densely and appressed ferruginous-villosulose without, the tube 2.0-2.2 em. long, about 0.3 em. in diameter at the base, epistaminal ap- pendages replaced by callous subquadrate protuberances 0.1— 0.15 em. long, faucal annulus conspicuously thickened or tuber- culate, the lobes obliquely obovate, shortly acuminate, 0.8—0.9 em. long, reflexed or widely spreading; stamens inserted at about the upper 1% of the corolla-tube, the anthers slightly ex- serted, sagittate, 0.6 em. long, glabrous; ovary ovoid, about 0.14 em. long, essentially glabrous; stigma 0.2 em. long; neetaries concrescent, annular, essentially entire, about equalling the ovary; follieles unknown. BRITISH HONDURAS: growing over low bushes in open places, rare, Stann Creek Railway, alt. 50 ft., Aug. 29, 1929, Schipp S? (B, TYPE, G, MBG, photograph and analytical drawings). Similar to P. mexicana, to which it is doubtless closely re- lated, but differing in its smaller, more shortly petiolate leaves, smaller flowers, and partieularly in the shape and size of the epistaminal protuberances of the corolla as indicated in the key to species. 57. Prestonia isthmica Woodson, Ann. Mo. Bot. Gard. 18: 595. 1931 Prestoma longituba K. Sch. in Engl. & Prantl, Nat. Pflan- zenfam. 47: 188. 1895, nom. subnud. Stems relatively stout, rather finely ferruginous-tomentulose to glabrate, conspicuously lenticellate when fully mature; leaves broadly obovate to oblong-elliptic, apex abruptly and shortly acuminate, base obtuse or rounded, rather obscurely cordate, 9-22 em. long, 5-13 em. broad, firmly membranaceous, above rather minutely and sparsely hirtellous-strigillose, be- neath minutely ferruginous-tomentulose; petioles 0.3-1.2 em. (532) 1936] WOODSON—STUDIES IN THE APOCYNACEAE, IV 361 long, indument as upon the stem; stipular appendages intra- petiolar, rather broadly pectinate; inflorescence densely sub- umbellate, simple, bearing 4-12 yellowish flowers; peduncle much shorter than the subtending leaves, indument as upon the stem; pedicels 0.5-1.0 em. long, Somewhat — after oum minutely ferruginous-t ; braets ovate-lanceolate, 0.7—1.2 em. long, foliaceous; calyx: lobes ovate to ovate-lanceolate, acuminate, 2.0-2.5 em. long, foliaceous, minutely and rather irregularly hirtellous-papillate without, the internal squamellae broadly dentiform-ligular, minutely erose or lacerate, pilosulose; corolla salverform, ferrugi- nous-villosulose without, the tube 2.5-3.5 em. long, about 0.3—0.4 em. in diameter at the base, epistaminal appendages replaced by callous linear ridges 0.2-0.25 cm. long, faucal annulus conspicuously thickened or tuberculate, the lobes obliquely obovate, shortly acuminate, 1.3-1.7 cm. long, re- flexed or widely spreading; stamens inserted at about the upper 1, of the corolla-tube, the anthers oblong-sagittate, 0.6— 0.8 em. long, glabrous; ovary ovoid, about 0.2 cm. long, gla- brous; stigma 0.15-0.2 em. long; nectaries concrescent, tubular, deeply 5-lobed, fleshy, eonspieuously surpassing the ovary; follicles stout and rigid, narrowly napiform, sharply divaricate, 7-12 em. long, densely and rigidly ferruginous-hispid; seeds 1.0-1.6 em. long, the bright tawny coma 3.5-4.5 em. long. Costa RICA: GUANACASTE: collines de Nicoya, May, 1900, Tonduz 13945 (B); ALAJUELA: buissons, haies, aux eollines de Santiago prés S. Ramon, May 31, 1901, Brenes 14276 (B, G); buissons au bord du Tiliri à la Verbena, prés Alajuelita, Aug., 1894, Tonduz 8904 (Bx); SAN JOSE: wet thicket, between Aserri and Tar- baca, alt. 1200-1700 m., Dee. 6, 1925, Standley 41332 (US, TYPE, MBG, photograph and analytical drawings); Rio Virilla, alt. 1160 m., Oct., 1898, Tonduz 7441 (B, BB); Rio Virilla prés de San Juan, Oct., 1898, Tonduz 12711 (BB, M, V); S. Jose, date laeking, Hoffmann 522 (B); PUNTARENAS: foréts du R. Ceibo, prés Buenos Aires, alt. 200 m., Febr., 1892, Pittier 6652 (Bx). 58. Prestonia speciosa Donn. Sm. Bot. Gaz. 27: 435. 1899. Stems relatively stout, ferruginous-tomentose, eventually be- coming glabrate; leaves ovate to ovate-elliptie, apex abruptly and shortly acuminate, base broadly obtuse or rounded, 10-17 em. long, 7-11 em. broad, firmly membranaceous, above rather (533) [Vor. 23 362 ANNALS OF THE MISSOURI BOTANICAL GARDEN sparsely and minutely hispidulous-strigillose generally, densely ferruginous-tomentulose upon the midrib and veins, beneath finely and minutely ferruginous-tomentulose gen- erally; petioles 0.5-1.2 cm. long, densely ferruginous-tomen- tose; stipular appendages interpetiolar, numerous, pectinate- flagelliform; inflorescence subumbellate, simple, bearing 4-8 bright yellow flowers; peduncle scarcely surpassing the sub- tending petioles, or slightly shorter, ferruginous-tomentulose; pedicels 1.0—1.5 em. long, somewhat acerescent after maturity, minutely ferruginous-tomentulose; bracts narrowly lanceolate, 0.2-0.5 em. long, slightly foliaceous; calyx-lobes ovate to ovate- oblong, acute, 1.2-1.5 em. long, minutely and rather irregularly hirtellous, the internal squamellae broadly dentiform-deltoid, minutely emarginate or erose; corolla infundibuliform, mi- nutely ferruginous-villosulose without, the proper-tube 1.5-1.7 em. long, about 0.35 em. in diameter at the base, the throat rather narrowly conical-campanulate, 1.5-1.6 em. long, about 0.8-0.9 em. in diameter at the orifice, epistaminal append- ages replaced by rather inconspicuous, callous, obtriangular, foveolate protuberances 0.1—0.12 cm. long, faucal annulus con- spicuously thickened or tuberculate, the lobes obliquely ob- ovate, shortly acuminate, 2.0-2.5 em. long, widely spreading; stamens inserted at the base of the corolla-throat, the anthers included, oblong-sagittate, 0.8 em. long, glabrous; ovary ovoid, about 0.15 em. long, glabrous or minutely papillate; stigma 0.15 em. long; nectaries concrescent, annular, rather obscurely and irregularly lobed, fleshy, somewhat surpassing the ovary; fol- licles unknown. GUATEMALA: SANTA ROSA: Buena Vista, alt. 1700 m., April, 1893, Heyde § Lua 4497 (B, BB, G, US, TYPE, MBG, photograph and "ES drawings). SALVADOR: LA LIBERTAD: eultivated in the garden of the finea, Puerta de la La- guna, April 27, 1922, Standley 23673 (G, US). 59. Prestonia Riedelii (Muell.-Arg.) Mgf. in Fedde, Rep. Spec. Nov. 20: 26. 1924 Haemadictyon Riedelii Muell.-Arg. in Mart. Fl. Bras. 6': 170. 1860. (534) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 363 Temnadenia Riedelu (Muell.-Arg.) Miers, Apoc. So. Am. 216. 1878. Prestonia Muelleri Rusby, Mem. Torrey Bot. Club 4: 217. 1895. Echites (?) Riedel (Muell.-Arg.) Malme, Bull. Herb. Boiss. IT. 4: 196. 1904. Stems relatively stout, rather finely ferruginous-tomentose to glabrate; leaves broadly ovate to ovate-elliptie, apex acute to acuminate, base broadly obtuse to rounded, 5-18 cm. long, 3-11 em. broad, membranaceous, above densely hispidulous- strigillose to essentially glabrate, beneath minutely tomentu- lose to sparsely and irregularly puberulent or pilosulose; peti- oles 0.7—4.0 em. long, indument as upon the stem; stipular appendages intrapetiolar, extremely inconspicuous, pectinate; inflorescence bostrychoid-racemose, simple, relatively lax and elongate, bearing 20-45 brownish-yellow flowers; peduncle somewhat shorter than the subtending leaves, indument as upon the stem; pedicels 1.5-2.0 em. long, somewhat accrescent after maturity, minutely tomentulose; bracts lanceolate to nar- rowly ovate-lanceolate, 0.5-3.3 em. long, foliaceous; calyx-lobes elliptic-lanceolate, acuminate, 0.8-2.0 em. long, foliaceous, mi- nutely puberulent, the internal squamellae broadly deltoid- dentiform, minutely and irregularly erose to essentially en- tire; corolla salverform, rather indistinetly and irregularly puberulent without, the tube 1.3-1.5 em. long, about 0.25 em. in diameter at the base, epistaminal appendages replaced by callous, linear ridges 0.15-0.3 em. long, faucal annulus con- spieuously thickened or tuberculate, the lobes obliquely oblong- elliptie to obovate, 1.1-1.7 em. long, reflexed or widely spread- ing; stamens inserted at slightly below midway within the corolla-tube, the anthers wholly included, narrowly oblong- sagittate, shortly auriculate, 0.55-0.65 em. long, sparsely pi- losulose; ovary ovoid, 0.1-0.125 cm. long, minutely papillate; stigma 0.125-0.15 em. long; nectaries separate or essentially so, compressed-obovoid or oblong-obovoid, about as long as the ovary; follicles relatively elongate and slender, somewhat ar- ticulated, usually more or less faleate, occasionally persistently (535) [Vor. 23 364 ANNALS OF THE MISSOURI BOTANICAL GARDEN united at the tips until maturity, 10-25 em. long, minutely and densely hispidulous; seeds 1.1-1.3 em. long, obsoletely rostrate, the pale tawny coma about 2.5 cm. long. PERU: AYACUCHO: Aina, between Huanta and Rio Apurimac, open woods, alt. 750-1000 m., May 7-17, 1929, Killip $ Smith 22708 (US). BOLIVIA: LA PAZ: ur alt. 1300 m., Dee., 1917, Buchtien 277 (MBG); Polo-Polo bei Coroieo, alt. 1100 m., Oet.-Nov., 1912, Buchtien 279 (B, DL, MBG); ** Yungas,'' 1890, Bang 403 ae BR, DL, G, M, MBG); TARIJA: Soledad bei S. Bs: b and alt. 1500 m., Jan. 30, 1904, Fiebrig 2675 (B, DL, M, U, V); SANTA : Cue WEE e alt. 1200 m., Dec. 23, 1921, Steinbach 6065 (B); Samaipata, FOE alt. 1200 m., March 10, 1920, Steinbach 3732 (B); im Wald der Que- brada de Charagua, Dec., 1910, Herzog 11230 (B, DL). PARAGUAY: in regione fl. Alto Parana, 1909-10, Fiebrig 5947 (B); in reg. col- lium ** Cerros de Tobaty,’’ Sept., 1900, Hassler 6424 (B, V); in reg. lacus Ypaearay, April, 1913, Hassler 12166 (C, MBG); ad ripam Piribibuey, Aug., year lacking, Hassler 3211 (B, V); in altaplanitie et declivibus ‘‘Sierra de Amambay,’’ Dee., 1907, Rojas 9768a (B); Cordillera de Altos, Aug., 1902, Fiebrig 64 (B, DL, M); zwischen Rio Apa und Rio Aquidaban, San Luis, Dee., 1908, Fiebrig 4447 (B, G, M); in reg. collium, Cordillera de Villa-Rica, Jan., 1905, Hassler 8678 (B). BRAZIL: SAO PAULO: campinas, date lacking, Novaes 11207 (B); Canna Velha, April, 1848, Regnell III 584 (S); data incomplete, De Jonghe s.n. (Camb.); PARANA: ad marginem silvulae, Jaguariahyva, alt. 740 m., April 15, 1911, Dusén 11624 (MBG, S); ad marginem silvulae, Itarare opp., alt. 730 m., Jan. 21, 1915, Dusén 16454 (MBG, S, US); Villa Velha, in campo, Febr. 4, 1905, Dusén 7699 (S). ARGENTINA: MISIONES: Posadas, La Granja, Dee. 3, 1907, Ekman 1592 (8). This is a relatively uniform species. It departs from other representatives of $ Tomentosae in the relatively scant in- dument of the corolla. The color of the corolla is reported by Herzog as ‘‘rot-braun mit gelbem schlund.’’ 60. Prestonia Schumanniana Woodson, spec. nov. Suffrutieosa volubilis ; ramulis gracilibus minute ferrugineo- tomentulosis; foliis late ellipticis apice abrupte breviterque acuminatis basi obtusis 12-15 em. longis 5-8 em. latis rigide membranaceis supra sparse minutissimeque strigilloso-papil- latis nervo medio nervisque minute ferrugineo-pilosulis subtus minutissime ferrugineo-tomentulosis; petiolis 1.1-1.5 «em. longis minutissime tomentulosis; appendicibus stipulaceis interpetiolaribus multis minute pectinatis; inflorescentiis sub- umbellatis simplicibus vel obscure dichotomis flores 6-12 luteos (536) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 365 gerentibus ; pedunculo petiolis subaequanto minute ferrugineo- tomentuloso; pedicellis 1.0-1.1 cm. longis post maturitatem paulo acerescentibus minutissime tomentulosis ; bracteis ovatis 0.1-0.3 em. longis ; calycis laciniis oblongo-ellipticis apice acutis 0.9-1.1 em. longis foliaceis minute hirtello-papillatis squa- mellis late dentiformibus minute emarginatis puberulis; corol- lae salverformis extus minute ferrugineo-villosulae tubo 1.5 cm. longo basi ea. 0.2 em. diametro metiente intus plieas eallosas minutas late oblongas pro appendicibus epistaminalibus gerente annulo faucium conspicue incrassato tumido lobis oblique obovatis brevissime acuminatis 1.0-1.2 cm. longis pat- entibus; antheris paululo exsertis elliptico-sagittatis 0.7 cm. longis glabris; ovario ovoideo ca. 0.2 em. longo glabro; stig- mate 0.2 em. longo; nectariis basi concrescentibus carnosis ovarium paulo superantibus ; follieulis ignotis. Stems relatively slender, minutely ferruginous-tomentulose ; leaves broadly elliptic, apex abruptly and shortly acuminate, base obtuse, 12-15 cm. long, 5-8 em. broad, firmly membrana- ceous, above sparsely and very minutely strigillose-papillate generally, the midrib and nerves minutely ferruginous-pilosu- lose, beneath very minutely ferruginous-tomentulose ; petioles 1.1-1.5 em. long, very minutely tomentulose; stipular append- ages interpetiolar, numerous, minutely pectinate ; inflorescence subumbellate, congested, simple, or obscurely dichotomous, bearing 6-12 yellowish flowers; peduncle about as long as the subtending petioles, minutely ferruginous-tomentulose; ped- icels 1.0-1.1 em. long, slightly accrescent after maturity, very minutely tomentulose; bracts ovate, 0.1—0.3 em. long; calyx- lobes oblong-elliptic, acute, 0.9-1.1 em. long, foliaceous, mi- nutely hirtellous-papillate, the internal squamellae broadly dentiform, minutely emarginate, puberulous; corolla salver- form, minutely ferruginous-villosulose without, the tube 1.5 cm. long, about 0.2 em. in diameter at the base, epistaminal ap- pendages replaced by broadly oblong, callous ridges, faucal annulus conspicuously thickened or tuberculate, the lobes obliquely obovate, very shortly acuminate, 1.0-1.2 em. long, re- (537) [Vor. 23 366 ANNALS OF THE MISSOURI BOTANICAL GARDEN flexed or widely spreading; stamens inserted slightly above midway within the corolla-tube, the anthers barely exserted, elliptie-sagittate, 0.7 cm. long, glabrous; ovary ovoid, about 0.2 em. long, glabrous; stigma 0.2 em. long; neetaries concrescent at the base, fleshy, somewhat surpassing the ovary ; follicles unknown. ECUADOR: GUAYAS: Balao, May, 1892, Eggers 14722 (M, TYPE, MBG, photograph and analytieal drawings). Important features of morphology distinguish this species from P. Riedelii, the only other representative of $ Tomentosae in South America with the epistaminal appendages replaced by callous ridges. These may be summarized as follows: P. Riedelu P. Schumanniana Inflorescence: simple, lax and simple or obscurely dichoto- elongate. mous, congested, subum- bellate. Anthers: included, pilosu- barely exserted, glabrous, lose, auricles auricles long and slender. very short. Epistaminal ridges: linear. broadly oblong. EXCLUDED OR UNCERTAIN SPECIES Prestonia Goudotiana Baill. Bull. Mens. Soc. Linn. Paris 1: 192. 1889, nom. subnud. Possibly refers to a species of $ T'o- mentosae. The nectaries are briefly described, without citation of specimens or other characters by means of which interpre- tation can be based. Prestonia hirsuta (R. & P.) Spreng. Syst. 1: 637. 1825, not Muell.-Arg. (Echites hirsuta R. & P. Fl. Peruv. 2: 19. pl. 136. 1799, not A. Rich.) - Mandevilla Pavonii (A. DC.) Woodson, Ann. Mo. Bot. Gard. 19: 73. 1932 (Echites Pavonii A. DC. in DC. Prodr. 8: 463. 1844). Prestoma Langlassei Standl. Contr. U. S. Nat. Herb. 23: 1159. 1924 = Laubertia Pringlei (Greenm.) Woodson, Ann. Mo. (538) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 367 Bot. Gard. 18: 555. 1931 (Streptotrachelus Pringles Greenm. Proc. Am. Acad. 32: 298. 1897). Prestonia peruviana Spreng. loc. cit. 1825 = Mandevilla glandulosa (R. € P.) Woodson, loc. cit. 66. 1932 (Echites glandulosa R. & P. loc. cit. pl. 135. 1799). Haemadictyon bracteosum Muell.-Arg. in Mart. Fl. Bras. 6: 168. 1860. Ihave not been able to examine the type specimen of this species, collected by Riedel near Rio de Janeiro. The description accords with that of Prestonia perplexa Woodson (vide ante) in all essential characters save the elaborate bracts, which the specific adjective commemorates. In view of that fact, it appears wise to maintain the entities separate. XXVII. Ruopocatyx Muell.-Arg. Rhodocalyx Muell.-Arg. in Mart. Fl. Bras. 6': 172. 1860; Miers, Apoc. So. Am. 138. 1878 (as to Rh. rotundifolius). Lactescent, suffrutescent herbs. Stems terete, erect, simple, or branching at the very base. Leaves opposite, petiolate, en- tire, penninerved, eglandular, the petioles exappendiculate or essentially so. Inflorescence terminal, simply racemose, con- spicuously bracteate. Calyx 5-parted, the lobes somewhat unequal, imbricated, cleft nearly to the receptacle, conspicu- ously foliaceous or subpetaloid, bearing within at the base deeply lacerate, opposite squamellae. Corolla salverform, ex- appendiculate within, the orifice of the tube conspicuously an- nulate, the limb actinomorphic, 5-parted, dextrorsely convolute. Stamens 5, included, the anthers connivent and agglutinated to the stigma, consisting of 2 parallel, basally protuberant sporangia borne ventrally near the apex of an enlarged, nar- rowly sagittate connective; pollen granular. Carpels 2, united at the apex by a common stylar shaft surmounted by the fusiform-capitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, separate. Fol- licles 2, apocarpous, terete, dehiscing along the ventral suture, containing numerous dry, rostrate, apically comose seeds (ac- cording to Muell.-Arg.). (539) (VoL. 23 368 ANNALS OF THE MISSOURI BOTANICAL GARDEN Type species: Rhodocalyx rotundifolius Muell.-Arg. in Mart. Fl. Bras. 61: 173. pl. 51. 1860. 1. Rhodocalyx rotundifolius Muell.-Arg. in Mart. Fl. Bras. 6': 173. pl. 51. 1860; Miers, Apoc. So. Am. 138. 1878. Echites erecta A. DC. in DC. Prodr. 8: 469. 1844. Stems erect, simple, or branched at the base, 2.0-4.5 dm. tall, ferruginous-hirtellous ; leaves opposite, shortly petiolate, ovate to suborbicular, apex rounded to very shortly and abruptly acuminate, base broadly obtuse to rounded, the uppermost and the lowermost usually much reduced, 5-9 em. long, 3.0-9.5 em. broad, firmly membranaceous, above ferruginous-hispidulous, beneath densely ferruginous-tomentulose; petioles 0.3-0.5 em. long ; inflorescence somewhat surpassing the subtending leaves, bearing 3-18 rather showy flowers; peduncle densely ferrugi- nous-hirtellous; pedicels 1-2 em. long, ferruginous-hirtellous ; bracts very conspicuous and laminate, foliaceous to sub- petaloid, oblong to ovate-oblong, about equalling to somewhat surpassing the subtended pedicels ; calyx-lobes oblong to ovate- oblong, acute to acuminate, 2.0-2.5 em. long, the margins ciliolate, otherwise essentially glabrous, rather delicately membranaceous, purplish (according to Muell.-Arg.), the squamellae very deeply lacerate; corolla salverform, glabrous without, the tube 2.0 em. long, about 0.1 em. in diameter at the base, the orifice conspicuously annulate, the lobes obliquely obovate, 1.2-1.3 em. long, somewhat reflexed ; stamens inserted somewhat above midway within the corolla-tube, the anthers 0.6 cm. long, essentially glabrous dorsally ; ovary ovoid, about 0.1 em. long, essentially glabrous; stigma 0.15 em. long; nec- taries separate, somewhat shorter than the ovary; follicles unseen. BRAZIL: MINAS GERAES: data incomplete, Regnell III 885 (S, US); Glaziou 12951 (US); MATTO GROSSO: campo, Procedencia Caceres, Sept., 1911, Hoehne 4697 (US); DATA INCOMPLETE: Riedel s.n. (G, NY). Also reported from the Brazilian states of S&o Paulo, Bahia, and Espiritu Santo by Mueller, who describes the color of the corolla as dark erimson to rose-purple. (540) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 369 EXCLUDED SPECIES Rhodocalyx calycosus (A. Rich.) Miers, Apoc. So. Am. 140. 1878 (Echites calycosa A. Rich. in Sagra, Hist. Cuba 11: 94. 1850) = Asketanthera calycosa (A. Rich.) Woodson, Ann. Mo. Bot. Gard. 19: 47. 1932. Rhodocalyx cinereus (A. Rich.) Miers, Apoc. So. Am. 141. 1878 (Echites cinerea A. Rich. in Sagra, Hist. Cuba 11: 93. 1850) = Haplophyton cinereum (A. Rich.) Woodson, comb. nov. Notes on the relegation of this species will be found on p. 231. Rhodocalyx coccineus (Hook. & Arn.) Miers, Apoc. So. Am. 141. 1878 (Echites coccinea Hook. & Arn. in Hook. Jour. Bot. 1: 286. 1834) = Mandevilla coccinea (Hook. € Arn.) Woodson, Ann. Mo. Bot. Gard. 20: 734. 1933. Rhodocalyzx crassifolius (Muell.-Arg.) Miers, Apoc. So. Am. 139. 1878 (Amblyanthera crassifolia Muell.-Arg. in Mart. Fl. Bras. Bi: 143. 1860) = Galactophora crassifolia ( Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 19: 50. 1932. Rhodocalyz crassipes (A. Rich.) Miers, Apoc. So. Am. 140. 1878 (Echites crassipes A. Rich. in Sagra, Hist. Cuba 11: 91. 1850) = Echites umbellata Jacq. var. crassipes (A. Rich.) Gomez, Anal. Soc. Espan. Hist. Nat. 23: 274. 1894. Rhodocalyx cuneifolius Miers, Apoc. So. Am. 142. 1878. Perhaps equivalent to Mandevilla velutina (Mart.) Woodson. Rhodocalyx hypoleucus (Benth.) Miers, Apoc. So. Am. 140. 1878 (Echites hypoleuca Benth. Pl. Hartw. 23. 1839) = Macro- siphonia hypoleuca (Benth.) Muell.-Arg. Linnaea 30: 452. 1860. Rhodocalyx lanuginosus (Mart. & Gal.) Miers, Apoc. So. Am. 139. 1878 (Echites lanuginosa Mart. & Gal. Bull. Acad. Roy. Brux. 11': 357. 1844) = Macrosiphonia lanuginosa (Mart. A Gal.) Hemsl. Biol. Centr.-Am. Bot. 2: 316. 1882. Rhodocalyx ovatus Miers, Apoc. So. Am. 141. 1878 = Mandevilla coccinea (Hook. & Arn.) Woodson, Ann. Mo. Bot. Gard. 20: 734. 1933 (Echites coccinea Hook. & Arn. in Hook. Jour. Bot. 1: 286. 1834). Rhodocalyx suaveolens (Mart. & Gal.) Miers, Apoc. So. Am. 139. 1878 (Echites suaveolens Mart. & Gal. Bull. Acad. Roy. (541) [Vor. 23 310 ANNALS OF THE MISSOURI BOTANICAL GARDEN Brux. 11*: 356. 1844, not A. DC.) = Macrosiphonia hypoleuca (Benth.) Muell.-Arg. Linnaea 30: 452. 1860 (Echites hypo- leuca Benth. Pl. Hartw. 23. 1839). Rhodocalyx Tweedianus Miers, Apoc. So. Am. 142. 1878. Probably a species of Mandevilla. XXVIII. Lavsertia A. DC. Laubertia A. DC. in DC. Prodr. 8: 486. 1844; Benth. & Hook. Gen. Pl. 2: 724. 1876; Miers, Apoc. So. Am. 124. 1878; K. Sch. in Engl. & Prantl, Nat. Pflanzenfam. 42: 170. 1895. Streptotrachelus Greenm. Proc. Am. Acad. 32: 298. 1897. Lactescent, fruticose or suffruticose lianas. Stems volubile, terete; branches alternate or opposite below. Leaves opposite to occasionally ternate or quaternate, petiolate, entire, penni- nerved, eglandular; petioles somewhat girdling at the nodes, subtended by numerous minute, pectinate, adaxial, stipular appendages. Inflorescence lateral, alternate, occasionally sub- terminal or terminal, di- or trichotomously scorpioid, bracteate, bearing few to numerous rather mediocre flowers. Calyx 5- parted, the lobes equal to subequal, slightly foliaceous, cleft nearly to the receptacle, imbricated, eglandular within. Co- rolla salverform, minutely ferruginous-pubescent without, the tube straight or spirally contorted, exappendiculate within, the orifice conspicuously annulate, the limb actinomorphic, 5- parted, dextrorsely convolute. Stamens 5, the anthers some- what exserted, connivent and agglutinated to the stigma, con- sisting of 2 parallel, basally protuberant sporangia borne ventrally near the apex of an enlarged, acutely sagittate con- nective; pollen granular. Carpels 2, connected at the apex by the common stylar shaft surmounted by the fusiform-capitate stigma; ovules many, several-seriate, borne upon an axile, binate placenta. Nectaries 5, separate or somewhat concres- cent at the base. Follicles 2, apocarpous, terete, dehiscing along the ventral suture, containing many dry, truncate, apically comose seeds. Type species: Laubertia Boissierii A. DC. in DC. Prodr. 8: 487. 1844. (542) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 371 KEY TO THE SPECIES a. Corolla-tube straight, not spirally contorted; species of South America. b. Inflorescence much surpassing the subtending leaves, conspicuously and usually repeatedly compound; calyx-lobes ovate-trigonal....1. L. Boissierii bb. Inflorescence about equalling, or somewhat shorter than the subtending leaves, rather obseurely compound to essentially simple; ealyx-lobes E, A es RUD MA VE NET MEVS 2. L. Sanctae-Martae aa. Corolla-tube spirally contorted; species of Mexico and Central America. b. Calyx-lobes ovate-lanceolate, 0.3-0.4.5 em. long; eorolla-tube 2.0-2.3 em. long, anther tips experted.......:oooooooocosrossrocrioaso 3. L. Pringlei bb. Calyx-lobes narrowly oblong-elliptic, 0.9—1.1 em. long; iir tube 1.3- 1.4 em. long; anthers barely ineluded.................. . L. peninsularis 1. Laubertia Boissierii A. DC. in DC. Prodr. 8: 487. 1844; Miers, Apoc. So. Am. 124. 1878. Echites Eggersti Mgf. Notizblatt 9: 78. 1924. Stems relatively stout, finely ferruginous-hirtellous when young, becoming glabrate and conspicuously lenticellate when fully mature; leaves opposite, occasionally ternate or qua- ternate, petiolate, ovate to oblong-lanceolate, apex acute to acuminate, base broadly obtuse to rounded or obscurely cor- date, 4-18 cm. long, 1-9 em. broad, firmly membranaceous, above minutely hirtellous to glabrate, beneath finely and rather sparsely puberulent when young, eventually glabrate; petioles 1.0-1.5 em. long, minutely and rather sparsely puberulent to glabrate; inflorescence much surpassing the subtending leaves, seorpioid, the peduncle conspicuously and usually repeatedly compound, bearing relatively numerous, greenish-purple or reddish flowers; pedicels 0.8-1.2 em. long, finely puberulent- papillate; bracts minutely ovate, scarious to only slightly foli- aceous ; calyx-lobes ovate-trigonal, sharply acute to acuminate, 2.0-3.5 em. long, minutely and ferruginously puberulent-papil- late without, slightly foliaceous to nearly scarious; corolla salverform, rather indistinctly puberulent-papillate without, the tube 1.5-2.0 em. long, about 0.25 cm. in diameter at the base, straight, not spirally contorted, the lobes obliquely obovate- dolabriform, 1.0-1.5 em. long, reflexed; stamens inserted near the orifice of the corolla-tube, the anthers somewhat exserted, 0.6 em. long, glabrous to minutely puberulent-papillate dor- sally ; ovary about 0.125 em. long, glabrous to indefinitely papil- (543) [Vor. 23 372 ANNALS OF THE MISSOURI BOTANICAL GARDEN late; nectaries nearly equalling the ovary; follicles relatively slender, rather obscurely moniliform, 25-40 em. long, glabrous ; seeds 1.0-1.5 cm. long, the pale tawny coma 2.0-2.5 em. long. ECUADOR: MANABI: im Wäldern bei El Recreo, April 30, 1897, Eggers 15684 (B, IBG, NY); DATA INCOMPLETE: Pavon s.n. (BB, TYPE, MBG, photograph and analytical drawings). PERU: SAN MARTIN: on riverside brush, La Merced, alt. 2000 ft., Aug. 10-24, 1923, Macbride 5473 (FM, US); edge of montaña along trail, Muña, alt. 7000 ECH May 23-June 4, 1923, Macbride 3902 (FM, US). 2. Laubertia Sanctae-Martae (Rusby) Woodson, Ann. Mo. Bot. Gard. 18: 555. 1931. Echites Sanctae-Martae Rusby, Deser. So. Am. Pl. 85. 1920 Stems relatively slender, minutely ferruginous-puberulent when young, becoming glabrate at maturity; leaves opposite, petiolate, elliptic, apex acuminate, base acutely cuneate, 10- 15 em. long, 1.5-4.5 em. broad, rather delicately membrana- ceous, minutely and rather sparsely pilosulose to glabrate above, glabrous beneath; petioles 1.0-1.3 em. long; inflores- cence about equalling, or somewhat shorter than the subtending leaves, rather obscurely, and usually only once dichotomous to essentially simple, bearing several greenish-purple or reddish flowers ; pedicels 0.8-1.0 cm. long, minutely and rather sparsely puberulent, greatly accrescent in fruit; bracts minutely lance- olate to linear, 0.1-0.3 em. long, somewhat foliaceous; calyx- lobes oblong-linear, acute to acuminate, 0.4—0.6 em. long, rather conspicuously foliaceous, minutely and ferruginously ap- pressed-puberulent without; corolla salverform, minutely and ferruginously appressed-puberulent without, the tube straight, not spirally contorted, 2.4-2.7 em. long, about 0.3 em. in di- ameter at the base, the lobes obliquely obovate-dolabriform, 0.9-1.1 em. long, reflexed ; stamens inserted at about the upper 1% of the corolla-tube, the anthers slightly exserted, 0.55-0.6 em. long, minutely papillate dorsally; ovary ovoid, about 0.15 cm. long, essentially glabrous; nectaries about equalling the ovary ; follicles relatively slender, rather distantly moniliform, 40-45 em. long, glabrous ; seeds 1.0-1.3 cm. long, the pale tawny coma 3.0-3.5 em. long. (544) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 313 COLOMBIA: MAGDALENA: rare in ravines and on wooded hillsides near Valparaiso, alt. 4000—5000 ft., Jan. 20, 1899, Smith 1643 (MBG, NY, TYPE) ; data incomplete, Smith 2525 (MBG, NY). 3. Laubertia TEES (Greenm.) Woodson, Ann. Mo. Bot. Gard. 18: 555. pO hata Pringlei Greenm. Proc. Am. Acad. 32: 208. 1897. Prestoma Langlassei Standl. Contr. U. S. Nat. Herb. 23: 1159. 1924. Stems relatively stout, minutely hirtellous when young, becoming glabrate at maturity; leaves opposite, petiolate, broadly ovate- to oblong-elliptic, apex shortly acuminate, base rounded to very obscurely cordate, 5-10 em. long, 2.5-5.0 em. broad, membranaceous, very minutely and rather sparsely puberulent to essentially glabrate above and beneath ; petioles 1.5-3.0 em. long, minutely and rather sparsely puberulent ; in- florescence simple or essentially so, about equalling, or some- what shorter than the subtending leaves, bearing several greenish-purple flowers; pedicels 1.0-1.5 cm. long, minutely and rather sparsely appressed-puberulent; bracts narrowly lanceolate, 0.1-0.3 cm. long, somewhat foliaceous; calyx-lobes ovate-lanceolate, acuminate, 0.3-0.45 cm. long, minutely hirtel- lous without, somewhat foliaceous; corolla salverform, mi- nutely ferruginous-puberulent without, the tube spirally con- torted at about midway below the insertion of the stamens, 2.0-2.3 em. long, about 0.3 em. in diameter at the base, the lobes rather broadly dolabriform, 0.7—0.8 cm. long, reflexed ; stamens inserted somewhat above midway within the corolla-tube, the anthers 0.7—0.75 em. long, minutely hirtellous dorsally, the tips slightly exserted; ovary ovoid, about 0.15 cm. long, minutely hirtellous ; neetaries somewhat shorter than the ovary ; mature follicles unknown, the immature relatively slender, rather distantly moniliform, finely and ferruginously appressed- hirtellous. MEXICO: MORELOS: lava beds near Cuernavaca, alt. 5200 ft., Sept. 23, 1896, Pringle 6554 (G, TYPE, MBG); SINALOA: near Colomar, July, 1897, Rose 1716 (US). (545) [Vor. 23 314 ANNALS OF THE MISSOURI BOTANICAL GARDEN 4. Laubertia peninsularis Woodson, spec. nov. Frutieosa volubilis altitudine ignota e fragmento vix facile deseripta; ramulis erassiuseulis dense ferrugineo-tomentu- losis; foliis oppositis petiolatis ovato-elliptieis apice acuminatis basi rotundatis vel obscurissime cordatis 6-13 cm. longis 4-7 em. latis (fide cl. Schipp) firme membranaceis supra juventate minute puberulis maturitate glabratis subtus minute ferrugi- neo-puberulis; petiolis 2.0-2.3 em. longis minute ferrugineo- puberulis; inflorescentiis foliis brevioribus flores mediocres gilvos (fide el. Schipp) 10-20 gerentibus ; pedunculo di- vel tri- chotomo dense ferrugineo-hirtello; pedicellis congestis 0.8-1.0 em. longis dense ferrugineo-hirtellis; bracteis anguste lance- olatis 0.2-0.4 em. longis caducis; calycis laciniis oblongo- ellipticis acutis acuminatisve 0.9-1.1 em. longis conspicue sub- foliaceis extus intusque ferrugineo-hirtellis; corollae salver- formis extus minute ferrugineo-hirtellae tubo 1.3-1.4 cm. longo basi ca. 0.25 cm. diametro metiente prope medium paulo con- torto lobis oblique obovatis obtusis 0.9-1.1 em. longis patenti- bus; antheris 0.5 cm. longis dorso minute puberulo-papillatis apice paulo inclusis; ovario ovoideo ca. 0.15 cm. longo glabro; nectariis ovoideis ovario subaequantibus; folliculis ignotis. Stems relatively stout, densely ferruginous-tomentulose; leaves opposite, petiolate, ovate-elliptic, apex acuminate, base rounded to very obscurely cordate, 6-13 cm. long, 4-7 cm. broad, firmly membranaceous, above minutely puberulent when young to essentially glabrate at maturity, beneath minutely ferrugi- nous-puberulent; petioles 2.0-2.3 cm. long, minutely ferrugi- nous-puberulent; inflorescence somewhat shorter than the subtending leaves, bearing 10-20 mediocre, cream-colored flowers; peduncle di- or trichotomous, densely ferruginous- hirtellous; pedicels congested toward the upper half of the peduncle, 0.8-1.0 cm. long, ferruginous-hirtellous; bracts nar- rowly lanceolate, 0.2-0.4 cm. long, caducous; calyx-lobes ob- long-elliptic, acute to acuminate, 0.9-1.1 cm. long, conspicu- ously subfoliaceous, minutely ferruginous-hirtellous within and without; corolla salverform, minutely ferruginous-hirtel- lous without, the tube 1.3-1.4 em. long, about 0.25 em. in di- (546) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 375 ameter at the base, spirally contorted at above midway immediately below the insertion of the stamens, the lobes obliquely obovate, obtuse, 0.9—1.1 cm. long, reflexed; stamens inserted at about midway within the corolla-tube, the anthers barely included, 0.5 em. long, minutely puberulent-papillate dorsally; ovary ovoid, about 0.15 em. long, essentially glabrous; nectaries ovoid, somewhat shorter than the ovary; follicles unknown. BRrrisH HONDURAS: B. H.-Guatemala Boundary Survey, spring or summer, 1934, Schipp s.n. (MBG, TYPE). In the spring of 1934, when Schipp accompanied an expedi- tion to determine the boundary between Guatemala and British Honduras, he carried vials containing alcoholic preservative for apocynaceous flowers intended for me. While on this trip Mr. Schipp collected a number of interesting novelties, includ- ing the type of Odontadenia Schippu Woodson, the only known representative of the genus north of Panama. Unfortunately, however, herbarium specimens of several Apocynaceae were completely spoiled in transit, with the survival only of the alcoholic specimens intended primarily for anatomical studies. Such circumstances befell the collections of Laubertia penin- sularis. Nevertheless, the preserved specimen was ample to show the plant to be a new species of this poorly understood genus, and a fragment of it is now incorporated in the herbar- ium of the Missouri Botanical Garden. Mr. Schipp describes his plant as “a tall vine growing in dense forest shade, the leaves of which are about five inches long and two and one-half wide. Flowers cream and slightly perfumed.” EXCLUDED SPECIES Laubertia (?) laxiflora Rusby, Bull. N. Y. Bot. Gard. 4: 408. 1907 = Odontadenia laxiflora (Rusby) Woodson, Ann. Mo. Bot. Gard. 19: 386. 1932. CoRRIGENDA Page references are to the parenthetical numbers at the bot- tom of the pages of ‘‘Studies in the Apocynaceae. IV.’’ Pages (547) 376 [VoL. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN 1-186 are contained in Ann. Mo. Bot. Gard. 20; 187-340 in 22; and 341-563 in 23. p. 10 p. 12 p. 18 p. 266. PA p. 105. 8. 3. O) Remove Echites obovata Nees from synonymy. Mandevilla pulchra Woodson, nom. nov. Dipladenia glabra Rusby, Descr. So. Am. Pl. 88. 1920. Mandevilla glabra (Rusby) Woodson, Ann. Mo. Bot. Gard. 20: 709. 1933, not N. E. Br. Mandevilla velutina (Mart.) Woodson, var. angusti- folia (Stadelm.) Woodson, comb. nov. Echites Pohliana Stadelm. var. a angustifolia Stadelm. Flora 24': Beibl. 73. 1841. Dipladenia gentianoides Muell.-Arg. B glabra Muell.- Arg. in Mart. Fl. Bras. 6': 124. pl. 37. fig. 2. 1860. Mandevilla velutina (Mart.) Woodson, var. glabra (Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 20: 732. 1933. Macrosiphonia petraea (St. Hil.) K. Sch. var. minor (Hook.) Woodson, comb. nov. Echites grandiflora Desf. var. minor Hook. Jour. Bot. 1: 286. 1834. Macrosiphonia verticillata Muell.-Arg. 9 pinifolia Muell.-Arg. in Mart. Fl. Bras. 6': 141. 1860. Macrosiphonia petraea (St. Hil.) K. Sch. var. pinifolia (Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 20: 787. 1933. In place of Angadenia Lindeniana (Muell.-Arg.) Miers read: Angadenia Berterii (A. DC.) Miers, Apoc. So. Am. 180. 1878 (Echites Berterii A. DC. in DC. Prodr. 8: 447. 1844). ADDENDA Page references are to the parenthetical numbers at the bot- tom of the pages of “Studies in the Apocynaceae. IV.” Pages 1-186 are contained in Ann. Mo. Bot. Gard. 20; 187-340 in 22; and 341-563 in 23. (548) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 377 II. Mesecuires Muell.-Arg. p. 26. To synonymy of M. trifida (Jacq.) Muell.-Arg., add: Echites chlorantha Schlecht. Linnaea 26: 663. 1853; Miers, Apoc. So. Am. 196. 1878. p. 36. To synonymy of M. angustifolia (Poir.) Miers, add: Echites concolor Ham. Prodr. 31. 1825. III. Maxpzvirra Lindl. p. 99. To synonymy of M. Martiana (Stadelm.) Woodson, var. glabra (Muell.-Arg.) Woodson, add: Micradenia acuminata (Hook.) Miers, Apoc. So. Am. 162. 1878. To synonymy of M. crassinoda (Gardn.) Woodson, add: Micradenia nodulosa Miers, Apoc. So. Am. 159. 1878. p. 120. To synonymy of M. atroviolacea (Stadelm.) Woodson, add: Micradenia atroviolacea (Stadelm.) Miers, var. ovata Miers, Apoc. So. Am. 159. 1878. p.124. To synonymy of M. illustris (Vell. Woodson, var. typica, add: Echites Gardneriana A. DC. in DC. Prodr. 8: 483. 1844. Echites Gardneriana A. DC. B grandiflora A. DC. loc. cit. 1844. P E b2 00 . To synonymy of M. velutina (Mart.) Woodson, var. angustifolia (Stadelm.) Woodson, comb. nov., add: Dipladenia longiloba A. DC. in DC. Prodr. 8: 485. 1844; Miers, Apoc. So. Am. 157. 1878. p. 130. To synonymy of M. coccinea (Hook. & Arn.) Woodson, add. Dipladenia ? coccinea (Hook. & Arn.) Muell.-Arg. in Mart. Fl. Bras. 61: 132. 1860. p. 135. To synonymy of M. subspicata (Vahl) Mgf., add: Laseguea latiuscula Miers, Apoc. So. Am. 251. 1878. p. 148. Tosynonymy of M. scabra (R. & S.) K. Sch., add. Echites canescens Willd. ex R. & S. Syst. 4: 795. 1819. (549) [Vor. 23 318 ANNALS OF THE MISSOURI BOTANICAL GARDEN Echites bicolor Miq. Stirp. Surinam. Select. 154. 1851. Laseguea bicolor (Miq.) Miers, Apoc. So. Am. 2951. 1878. p. 152. To synonymy of M. leptophylla (A. DC.) K. Sch., add: Amblyanthera leptophylla (A. DC.) Muell.-Arg. in Mart. Fl. Bras. 6': 142. 1860. p. 154. To synonymy of M. Fendleri (Muell.-Arg.) Woodson, add: Mitozus discolor Miers, Apoc. So. Am. 224, 1878. Echites discolor Moritz, ex Miers, loc. cit. nom. nud. in synon. p. 173. To the excluded species of Mandevilla, add: Mandevilla crassifolia (Spruce) K. Sch. ex Mgf. in Fedde, Rep. Sp. Nov. 20: 24. 1924, sphalm. = Galacto- phora crassifolia (Muell.-Arg.) Woodson, Ann. Mo. Bot. Gard. 19: 50. 1923 (Amblyanthera crassifolia Muell.-Arg. in Mart. Fl. Bras. 6': 143. 1860). Mandevilla Wrightiana (Muell.-Arg.) Benth. & Hook. Gen. Pl. 2: 727. 1876 (Rhabdadenia Wrightiana Muell.-Arg. Linnaea 30: 438. 1860) = Neobracea Valenzuelana (A. Rich.) Urb. Symb. Ant. 9: 241. 1924 (Echites Valenzuelana A. Rich. in Sagra, Hist. Cub. 11: 93. 1850). To the recognized species of Mandevilla add the following: Mandevilla Pittieri Woodson, spec. nov. Suffrutescens e rhizomate subtuberoso volubilis; ramis gra- cilibus ca. 4.5—6.0 dm. altis glaberrimis ; foliis oppositis petiola- tis oblongo-ellipticis apice breviter acuminatis basi late obtusis rotundatisve 7-8 em. longis 3.0-3.5 em. latis firmiter mem- branaceis glaberrimis supra sublutescentibus nervo medio basi pauciglanduligero subtus opacis; petiolis 1.2-1.5 em. longis glabris; appendicibus stipulaceis minutissimis vix bene visis; racemis simplicibus terminalibus subterminalibusve flores 3-5 gilvos (?) speciosos gerentibus; pedunculo petiolos ca. bis superante glabro; pedicellis 2 em. longis glaberrimis ; bracteis (550) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 379 haud visis; calycis laciniis ovatis acuminatis 0.45-0.5 cm. longis extus minutissime papillatis squamellis multis ; corollae infundibuliformis extus omnino glabrae tubo proprio 1.3 cm. longo ca. 0.18 em. basi diametro metiente faucibus anguste conicis 1.4 cm. longis ostio ca. 0.7 cm. diametro metiente lobis oblique obovatis 1.6-1.7 em. longis patulis; antheris 0.9 cm. longis glabris ; ovario oblongoideo in stylo gradatim angustato ca. 0.3 em. longo glabro; stigmate umbraculiforme breviter obtuseque apieulato 0.35 em. longo; nectariis 2 manifeste in- aequalibus ovario multo brevioribus; folliculis ignotis. Suffrutescent; rhizome subtuberous; stems relatively slen- der, glabrous, 4.5-6.0 em. tall, somewhat twining at the tips; leaves opposite, petiolate, oblong-elliptic, apex shortly (some- what subeaudate-) acuminate, base broadly obtuse to rounded, 7-8 em. long, 3.0-3.5 em. broad, firmly membranaceous, gla- brous, somewhat lustrous above, glandular at the base of the midrib, opaque beneath; petioles 1.2-1.5 em. long, glabrous; stipular appendages extremely minute, scarcely visible; ra- cemes simple, terminal or subterminal, bearing 3-5 showy, creamy white flowers; peduncle about twice surpassing the sub- tending petioles, glabrous; pedicels 2 em. long, glabrous; bracts not seen; calyx-lobes ovate, acuminate, 0.45-0.5 cm. long, very minutely papillate without, the squamellae numerous, indefi- nitely distributed; corolla infundibuliform, glabrous without, the proper tube 1.3 em. long, about 0.18 em. in diameter at the base, the throat narrowly conical, 1.4 em. long, about 0.7 em. in diameter at the orifice, the lobes obliquely obovate, 1.6-1.7 em. long, spreading; anthers 0.9 em. long, glabrous; ovary oblong- oid, about 0.3 em. long, glabrous; stigma umbraculiform, shortly and obtusely apiculate, 0.35 em. long, nectaries 2, mani- festly unequal, much shorter than the ovary; follicles unknown. Costa Rica: foréts de Luis, alt. 650 m., Nov., 1897, Pittier 11551 (B, TYPE, MBG, photograph and analytical drawings). This is the only species of the Dipladenia plexus of Mande- villa known from north of Colombia. Its most pronounced af- finities appear to be with M. cereola Woods., of Ecuador and Bolivia, from which it may be distinguished by its foliar glands (551) [Vor. 23 380 . ANNALS OF THE MISSOURI BOTANICAL GARDEN at the base of the midrib above, slight tendency to twine, and narrower corolla throat. Mandevilla collium Woodson, spec. nov. Fruticosa volubilis altitudine ignota; ramulis gracilibus glaberrimis maturitate bene lenticellatis; foliis oppositis peti- olatis obovatis apice brevissime acuminatis basi late obscuris- simeque cordatis 5-8 em. longis 3.0-4.5 em. latis membranaceis supra nervo medio basi pauciglanduligero ibique molliter puberulo eaeterumque glabris glabratisve subtus nervo medio basi puberulis caeterumque glabris; petiolis 1-2 em. longis molliter puberulis; appendicibus stipulaceis interpetiolaribus conspieuis maturitate coriaceis unguiformibus; racemis sim- plieibus lateralibus; peduneulo ca. 7 em. longo glaberrimo prope apicem flores gilvos (?) speciosos 5-7 gerente ; pedicellis 0.5-0.7 em. longis glabris; bracteis caducis haud visis; calycis laciniis ovatis longe acuminatis 0.6 em. longis squamellis multis; corollae infundibuliformis extus glabrae tubo proprio 1.0-1.1 em. longo ca. 0.2 em. basi diametro metiente faucibus anguste conicis 2.0-2.2 em. longis ostio ca. 0.65 em. diametro metiente lobis oblique obovatis breviter acutis 2.0-2.2 em. longis patulis; antheris oblongo-sagittatis 0.6 em. longis gla- bris; ovariis oblongoideis ca. 0.1 em. longis glabris; stigmate umbraculiforme breviter obtuseque apiculato ca. 0.3 em. longo; nectariis 2 valde inaequalibus ovarium ea. dimidio aequantibus; follieulis ignotis. Frutieose lianas; stems relatively slender, glabrous, con- spicuously lenticellate when fully mature; leaves opposite, petiolate, obovate, apex very shortly acuminate, base broadly and very obscurely cordate, 5-8 em. long, 3.0-4.5 em. broad, membranaceous, above inconspicuously glandular at the base of the midrib, where softly puberulent as well, otherwise gla- brous, beneath softly puberulent toward the base of the midrib and otherwise glabrous; petioles 1-2 em. long, softly puberu- lent; stipular appendages interpetiolar, conspicuous, cori- aceous and unguiform at maturity; racemes simple, lateral; peduncle about 7 cm. long, glabrous, bearing 5-7 creamy-white flowers toward the tip; pedicels 0.5-0.7 em. long, glabrous; (552) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 381 bracts caducous, not seen; calyx-lobes ovate, narrowly acumi- nate, 0.6 em. long, glabrous, squamellae numerous, indefinitely distributed; corolla infundibuliform, glabrous without, the proper tube 1.0-1.1 em. long, about 0.2 em. in diameter at the base, the throat narrowly conical, 2.0-2.2 em. long, about 0.65 em. in diameter at the orifice, the lobes obliquely obovate, shortly acute, 2.0-2.2 em. long, spreading; anthers oblong- sagittate, 0.6 em. long, glabrous; ovary oblongoid, about 0.1 em. long, glabrous; stigma umbraculiform, shortly and bluntly apiculate, about 0.3 em. long; nectaries 2, manifestly unequal, about half equalling the ovary ; follicles unknown. BOLIVIA: LA PAZ: Trockenbusch, Conzata, alt. 1300 m., Sept. 20, 1926, Troll 2687 (B, TYPE, MBG, photograph and analytical drawing). Closely related to M. oblongifolia Woods. and M. pulchra Woods., also indigenous to Bolivia, from which it may be dis- tinguished as follows: Leaves oid puberulent throughout; corolla throat broadly conical, about 1.5 n diameter at the orifice; squamellae in alternate groups of 4—8 PE EE EE CADET EN M. oblongifolia Leaves glabrous throughout, or essentially so. Leaves inconspicuously puberulent toward the base of the midrib; corolla throat narrowly conical, about 0.65 em. in diameter at the orifice; squa- mellae numerous, indefinitely distributed.............oo.oo.o.... M. collium Leaves glabrous throughout; corolla throat rather narrowly conical, 0.8—1.0 em. in diameter at the orifice; squamellae in alternate groups of 2—4 ODER REC CIO OO UTD TP D o d uUo IEEE M. pulchra Mandevilla Krukovii Woodson, spec. nov. Frutieosa volubilis; ramulis graciliusculis teretibus sparse pilosulis glabratisve; foliis oppositis breviter petiolatis ob- longo-oblaneeolatis apice breviter acuminatis basi obscure auriculatis 8-12 em. longis 3-4 cm. latis firme membranaceis supra nervo medio sparse pilosulo ibique sparse glanduligero eaeterumque glabris subtus sparse ferrugineo-pilosulis; peti- olis 0.7-1.0 em. longis pilosulis; inflorescentiis simplice ra- cemosis alterno-lateralibus; pedunculo foliis subaequante mi- nutissime pilosulo; pedicellis 0.2-0.3 em. longis similiter vesti- tis; bracteis magnis late ovatis conspicue caudatis 2.0-2.8 em. longis membranaceis planis petaloideis minutissime puberulis [Vor. 23 382 ANNALS OF THE MISSOURI BOTANICAL GARDEN caducis; calycis laciniis ovato-lanceolatis aeuminatis 0.7-0.8 em. longis minutissime puberulis squamellis oppositis ir- regulariter erosis; corollae infundibuliformis (colore luteo- aurantiacae ?) extus dense minutissime appresse-puberulae tubo proprio eylindrico 3.2-3.5 em. longo basi ca. 0.3 em. diametro metiente faucibus tubulo-conicis 0.9-1.0 em. longis ostio ca. 0.6 em. diametro metiente lobis oblique obovatis breviter acuminatis 1.7-1.8 cm. longis patulis; antheris latiuseule oblongo-oblanceolatis apiee obtusiusculis basi brev- issime auriculatis ca. 0.5 em. longis glabris; ovario oblongoideo ea. 0.125 em. longo glabro; stigmate umbraculiforme breviter apiculato ca. 0.15 em. longo; nectariis 5, compresse obovoideis ovario paulo brevioribus; follieulis crassiuculis conspicue moniliformibus 25-28 em. longis minute puberulis; seminibus 1.2 em. longis como aurantiaco ca. 2 em. longo. Fruticose lianas; branches relatively slender, terete, sparsely pilosulose to glabrate; leaves opposite, shortly petiolate, ob- long-oblanceolate, apex shortly acuminate, base obscurely auriculate, 8-12 em. long, 3-4 em. broad, firmly membranaceous above, sparsely pilosulose and glandular along the midrib, otherwise essentially glabrous, beneath sparsely and generally ferruginous-pilosulose; petioles 0.7-1.0 em. long, pilosulose; inflorescence simply racemose, alternate-lateral, bearing 20-25 showy, yellowish-orange (?) flowers; pedunele about equalling the subtending leaves, very minutely pilosulose; pedicels 0.2— 0.3 em. long, minutely pilosulose; braets showy, petalaceous, membranaceous, broadly ovate, conspicuously caudate, 2.0-2.8 em. long, very minutely puberulent, caducous; calyx-lobes ovate-lanceolate, acuminate, 0.7-0.8 em. long, very minutely puberulent, the opposite squamellae irregularly erose; corolla infundibuliform, slightly gibbous, the proper-tube cylindrical, 3.2-3.5 em. long, about 0.3 em. in diameter at the base, the throat tubular-conical, 0.9-1.0 em. long, about 0.6 em. in diameter at the orifice, the lobes obliquely obovate, shortly acuminate, 1.7-1.8 cm. long, patulous; anthers rather broadly oblong- oblanceolate, apex obtusish, base very shortly auriculate, about 0.5 em. long, glabrous; ovaries oblongoid, gradually narrowed (554) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 383 to the style, about 0.125 em. long, glabrous; stigma umbraculi- form, shortly apiculate, about 0.15 em. long; nectaries 5, com- pressed-obovoid, somewhat shorter than the ovary; follicles rather stout, conspicuously moniliform, 25-28 em. long, mi- nutely puberulent; seeds 1.2 em. long, the bright orange coma about 2 em. long. BRAZIL: AMAZONAS: Municipality Humayta, near Tres Casas, on low terra firma, Sept. 14-Oet. 11, 1934, Krukoff 6335 (NY, TYPE, MBG, photograph and analytical drawings). Perhaps most closely related to M. hirsuta ( A. Rich.) K. Sch., because of its conspicuously petalaceous bracts, but differing in its narrow corolla-throat and relatively scant indument gen- erally. In shape of foliage and general structure of the corolla M. Krukovii displays an affinity with M. lasiocarpa (A. DC.) Malme, which, however, has much smaller, lanceolate bracts. Were it not for the fact that M. hirsuta has been found to be surprisingly eonstant throughout a suite of hundreds of speci- mens representing a majority of extant herbarium specimens, the three species might be suspected as phases of a single, com- plex entity. V. FonsrERONIA G. F. W. Meyer p. 224. To synonymy of F. leptocarpa (Hook. & Arn.) A. DC., add: Forsteronia rotundiuscula Miers, Apoc. So. Am. 248. 1878. p. 235. To synonymy of F. thyrsoidea (Vell.) Muell.-Arg. var. glabriuscula ( A. DC.) Woodson, add: Forsteronia divaricata Miers, Apoc. So. Am. 247. 1878. p. 252. To synonymy of F. corymbosa (Jacq.) G. F. W. Meyer, add: Thyrsanthus pyriformis Miers, Apoc. So. Am. 100. 1878. p.257. To the list of excluded or doubtful species of For- steronia, add: Forsteronia ovalif olia (Poir.) Miers, Apoc. So. Am. 248. 1878 (Echites ovalifolia Poir. Encycl. Suppl. 2: 535. (555) [VoL. 23 384 ANNALS OF THE MISSOURI BOTANICAL GARDEN 1811). Perhaps equivalent to F. spicata (Jacq.) G. F. W. Meyer, although the latter is apparently unknown from Hispaniola. Forsteroma ? linearis (Vell.) Muell.-Arg. in Mart. Fl. Bras. Di: 107. 1860 (Echites linearis Vell. Fl. Flum. 111. 1830; Icon. 3: pl. 36. 1827). Probably a For- steronia but incapable of exact determination. IX. Opontapenta Benth. p. 313. To synonymy of O. verrucosa (R. & S.) K. Sch., add: Anisolobus rubidulus Miers, Apoc. So. Am. 173. 1878. p. 317. To synonymy of O. lutea (Vell.) Mef., add: Echites densevenulosa Stadelm. Flora 24!: Beibl. 47. 1841. Anisolobus Salzmanni A. DC. in DC. Prodr. 8: 395. 1844; Muell.-Arg. in Mart. Fl. Bras. 6*: 113. 1860; Miers, Apoc. So. Am. 169. 1878. Anisolobus Stadelmeyeri Muell.-Arg. loc. cit. 113. 1860; Miers loc. cit. 170. 1878. p. 320. To synonymy of O. puncticulosa (A. Rich.) Pulle, add: Amsolobus distinctus Miers, Apoc. So. Am. 169. 1878. Anisolobus oblongus Miers, loc. cit. 1878. To the recognized species of Odontadenia add the following: Odontadenia caudigera Woodson, spec. nov. Fruticosa volubilis; ramulis sat crassiusculis glabriusculis vel minutissime scabridiusculis haud conspicue lenticellatis; foliis oppositis longiuscule petiolatis late ellipticis apice acute breviterque caudato-acuminatis basi late obtusis 14-23 em. longis 7-11 em. latis membranaceis opacis glaberrimis ; petiolis 2.0-2.5 em. longis ; inflorescentiis lateralibus obscure compositis longe (10-12 em.) pedunculatis glaberrimis flores 3-6 gilvos speciosos gerentibus; bracteis ovatis 0.2-0.3 em. longis scari- aceis haud caducis; pedicellis ca. 2 em. longis; calycis laciniis (556) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 385 late ovatis obtusis 0.6-0.7 cm. longis glaberrimis squamellis alternatis 2-3; corollae subinfundibuliformis extus omnino glabrae tubo proprio urceolato 0.8-0.9 cm. longo basi ca. 0.5 em. diametro metiente faucibus cylindricis 1.6-1.7 cm. longis ostio ca. 0.6 cm. diametro metiente lobis obovato-dolabriformibus rotundatis 2.0-2.1 em. longis patulis; antheris anguste sagit- tatis acuminatis 1.2-1.3 em. longis dorso dense hirtellis ; ovario ovoideo ea. 0.2 em. longo glabro; stigmate acute 2-lobato 0.5 em. longo; nectariis carnosis profunde multidentatis ovarium paulo superantibus; follieulis ignotis. Stems relatively stout, glabrous or very minutely scabri- dulous, not conspicuously lenticellate; leaves opposite, peti- olate, broadly elliptic, apex acutely and shortly caudate-acu- minate, base broadly obtuse, 14-23 em. long, 7-11 em. broad, membranaceous, opaque, glabrous; petioles 2.0-2.5 em. long; inflorescence lateral, obscurely compound, glabrous, bearing 3-6 showy cream-colored flowers; peduncle 10-12 cm. long; bracts ovate, 0.2-0.3 em. long, scarious, persistent ; calyx-lobes broadly ovate, broadly obtuse, 0.6-0.7 em. long, glabrous, the squamellae in alternate groups of 2-3; corolla subinfundibuli- form, glabrous without, the proper-tube urceolate, 0.8-0.9 cm. long, about 0.5 em. in diameter at the base, abruptly constricted at the insertion of the stamens, the throat cylindrical, 1.6-1.7 em. long, about 0.6 cm. in diameter at the orifice, the lobes lous; anthers narrowly sagittate, acuminate, 1.2-1.3 em. long, densely hirtellous dorsally; ovary ovoid, about 0.2 em. long, glabrous; stigma acutely 2-lobed, 0.5 cm. long; nectaries fleshy, deeply multifid, slightly surpassing the ovary ; follicles unknown. BRITISH HONDURAS: data lacking, Schipp s.n. (MBG, TYPE). It is exasperating that no data accompanied the material of this species, supposedly from the interior of the colony. It is readily distinguished from both O. Hoffmannseggiana (Steud.) Woods., known only from Costa Rica and Panama in Central America and northern South Ameriea, and the Amazonian (557) (Vou. 23 386 ANNALS OF THE MISSOURI BOTANICAL GARDEN O. stemmadeniaefolia Woods. by the narrow corolla-throat which gives it the superficial aspect of a member of § Nitidae. XXVI. Prestoni R. Br. The following species, although included in the Key to Species, has been received too late to be incorporated within the revision proper of the genus: Prestonia discolor Woodson, spec. nov. Fruticosa volubilis; ramulis erassiuseulis glabris; foliis late ellipticis vel late oblongo-elliptieis apice brevissime acumi- natis acutisve basi late obtusis 11-21 em. longis 6-11 em. latis membranaceis post exsiecationem livide discoloratis opacis omnino glabris; petiolis 1.5-1.8 em. longis; appendicibus inter- petiolaribus sat numerosis late dentiformibus; inflorescentiis corymbosis dichotome divisis flores 30—40 luteo-roseos gerenti- bus; peduneulo foliis subaequante; pedicellis ca. 1.5 em. longis glabris; bracteis linearibus ca. 0.08 em. longis; calycis laciniis elliptico-oblongis acuminatis 1.2-1.3 em. longis delicate foli- aceis glabris squamellis dentiformibus minute erosis vel sub- integris; corollae salverformis extus omnino glaberrimae tubo 1.2-1.3 em. longis basi ca. 0.3 em. diametro metiente appendici- bus epistaminalibus omnino inclusis ca. 0.1 em. longis faucibus conspicue incrassatis lobis oblique obovatis haud acuminatis 1.4-1.5 cm. longis patentibus; antheris elliptico-sagittatis 0.6 em. longis dorso distincte pilosulis manifeste exsertis; ovario oblongoideo ca. 0.15 em. longo glabro; stigmate ca. 0.15 em. longo; nectariis compresse ovoideis basi irregulariter concres- centibus ovarium aequantibus; folliculis ignotis. Stems relatively stout, glabrous; leaves broadly elliptie to broadly oblong-elliptic, apex very shortly acuminate to acute, base broadly obtuse, 11-21 em. long, 6-11 em. broad, mem- branaceous, lividly discolored in desiccation, opaque, glabrous; petioles 1.5-1.8 em. long; interpetiolar appendages rather nu- merous, broadly dentiform; inflorescence corymbose, dichoto- mously divided, bearing 30-40 yellowish-pink flowers; pe- duncle about equalling the subtending leaves; pedicels about (558) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 387 1.5 em. long, glabrous; bracts linear, about 0.08 em. long; calyx- lobes elliptic-oblong, acuminate, 1.2-1.3 em. long, delicately foliaceous, glabrous, the squamellae dentiform, minutely erose to subentire; corolla salverform, glabrous without, the tube 1.2-1.3 em. long, about 0.3 cm. in diameter at the base, the epi- staminal appendages wholly included, about 0.1 em. long, the orifice conspicuously callose-incrassate, the lobes obliquely ob- ovate, not acuminate, 1.4-1.5 em. long, reflexed; anthers elliptic- sagittate, 0.6 em. long, distinctly pilosulose dorsally, the tips manifestly exserted; ovary oblongoid, about 0.15 em. long, gla- brous; stigma about 0.15 em. long; nectaries compressed-ovoid, irregularly concrescent at the base, equalling the ovary; fol- licles unknown. BRITISH GUIANA: Mora forest on edge of Karau Creek, Mazaruni River, May 25, 1933, Tutin 141 (BM, TYPE, MBG, photograph). Closely related to P. purpurissata, but evidently differing specifically in the smaller, greenish calyx-lobes, and shorter corolla-tube with proportionally longer lobes. Mr. Tutin re- ports that the corolla is yellow and pink, the calyx-lobes green, and the pedicels pale lilac in color. To the American genera of Echitoideae add the following: XXIX. TINTINNABULARIA Woodson Tintinnabularia Woodson, gen. nov. Apoeynacearum (Echi- toideae). Calyx majusculus profunde 5-partitus; laciniae foliaceae subaequales margine imbricatae intus basi in marginibus pauciglanduligerae. Corolla speciosissima magna infundibuli- formis; tubus inferne latiuscule cylindricus dein late dilatatus ibique staminiger; limbi laciniae 5 aequales oblique obovatae aestivatione dextrorsum convolutae. Stamina 5 omnino in- clusa; antherae inter se adglutinatae et stigmati adplicatae oblongo-sagittatae apice longe caudatae ibique plus minusve eonvolutae basi obtuse 2-auriculatae dimidia parte superiore ventro pollinigerae, sporangia uniforme fertilia pollinibus (559) [Vor. 23 388 ANNALS OF THE MISSOURI BOTANICAL GARDEN granulosis; filamenta filiformia antheris conspieue longiora haud adglutinata. Ovarii carpella gemina basi distineta apice in stylo gracili producta ovulis multis in quoque loculo pluri- seriatim positis; stigma capitato-fusiforme apice obscure ob- tuseque 2-partitum basi 5-manieulatum. Nectarii glandulae 9 saepissime separatae vel inter se plus minusve adglutinatae. Fructus ignotus ut creditur follieulus apocarpus. Frutices volubiles; folia opposita petiolata membranacea supra nervo medio inconspieue glanduligero subtus in axillis nervi medii inconspicue foveata. Inflorescentia lateralis alternata corym- boso-trichasialis pluriflora bracteis foliaceis oppositis. Lactescent (?), fruticose lianas. Stems volubile, terete; branches alternate above. Leaves opposite, membranaceous, the ventral surface bearing rather few, inconspicuous glandu- lar emergences indefinitely clustered at the base of the midrib, the dorsal surface bearing rather inconstantly inconspicuous elliptic foveae in the axils of the midrib; petioles somewhat girdling at the node into a slightly dilated, minutely appendic- ulate, stipular ring. Inflorescence lateral, alternate, corym- bose-trichasial, pluriflorous, the pedicels subtended by solitary, foliaceous bracts. Calyx 5-parted, the lobes subequal, foli- aceous, cleft nearly to the receptacle, imbricated, bearing within small groups of alternate, glandular squamellae. Co- rolla infundibuliform, the tube straight, rather broadly cy- lindrieal below, dilated into a broad throat at the insertion of the stamens, the limb 5-parted, actinomorphie, dextrorsely convolute. Stamens 5, inserted at the base of the corolla-throat, included; anthers connivent and agglutinated to the stigma, consisting of 2 parallel, uniformly fertile sporangia borne ven- trally near the apex of an enlarged, caudate, obtusely 2- aurieulate connective; pollen granular; filaments very con- spicuously longer than the anthers, filiform. Carpels 2, united at the apex by an elongate stylar shaft surmounted by the fusi- form-capitate, obtusely 2-lobed, basally 5-maniculate stigma; ovules many, several-seriate, borne upon an axile, binate pla- centa. Nectaries 5, separate or somewhat irregularly connate. Fruit unknown, supposedly follicular, apocarpous. (560) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 389 Tintinnabularia Mortonii Woodson, spec. nov. Plate 7. Fruticosa volubilis altitudine ignota; ramulis teretibus vel leviter compressis glabris maturitate inconspicue lenticellatis; foliis oppositis petiolatis oblongo-ellipticis apice obtuse cau- dato-acuminatis basi obtusis 9-10 em. longis 3.0-3.5 em. latis firme membranaceis omnino glabris supra nervo medio basi inconspicue glanduligeris subtus in axillis nervi medii incon- spicue irregulariterque foveolatis; petiolis 0.7-1.0 em. longis glabris; inflorescentiis corymboso-trichasialibus alterno-lat- eralibus folia ca. bis superantibus flores speciosos albidos ca. 9 gerentibus; pedunculo 4.0-5.5 em. longo glabro; bracteis foli- aceis ovato-lanceolatis acuminatis 1-2 em. longis; pedicellis 2 em. longis glabris; calycis laciniis oblongo-elliptieis acuminatis foliaceis 1.2-1.3 em. longis glabriuseulis intus basi squamellas 2-4 minutas alternatas gerentibus; corollae infundibuliformis tubo proprio latiuscule eylindrico 0.7—0.9 em. longo basi ca. 0.3 cm. diametro metiente extus puberulo-papillato intus prope in- sertionem staminum dense villosulo faucibus tubularibus 3.0-3.5 em. longis ostio ca. 0.8 em. diametro metiente extus intusque dense puberulo-papillatis lobis oblique obovatis ob- tusiusculis 0.9-1.0 em. longis extus puberulo-papillatis intus minute denseque puberulis paululo patulis; staminum antheris oblongo-sagittatis caudiculatis basi obtuse 2-aurieulatis 1.3 em. longis caudiculis minute pilosulis inter se convolutis fila- mentis filiformibus 3.5 em. longis basi minute pilosulis; ovariis oblongoideis ca. 0.35 cm. longis glabris; stigmate subcapitato basi 5-maniculato ca. 0.3 em. longo; nectariis compresse ovoideis ovario subaequantibus; follieulis ignotis. Frutescent lianas of unknown height; branches terete or slightly compressed, glabrous, inconspicuously lenticellate at maturity; leaves opposite, petiolate, oblong-elliptie, apex ob- tusely caudate-acuminate, base obtuse, 9-10 em. long, 3.0-3.5 em. broad, firmly membranaceous, glabrous throughout, upper surface bearing relatively few, inconspicuous glands at the base of the midrib, lower surface bearing with relative in- (561) [Vor. 23 390 ANNALS OF THE MISSOURI BOTANICAL GARDEN constancy solitary elliptic foveae in the axils of the midrib; petioles 0.7-1.0 em. long, glabrous; inflorescence alternate- lateral, corymbose-trichasial, about twice surpassing the length of the subtending leaves, GE about 9 showy, cream-colored flowers; peduncles 4.0-5.5 cm. long, glabrous; pedicels 2 cm. long, glabrous; bracts foliacecus ovate-lanceolate, acuminate, 1-2 em. long; calyx-lobes oblong-elliptic, acuminate, 1.2-1.3 cm. long, foliaceous, essentially glabrous, the squamellae in alter- nate groups of 2-4; corolla infundibuliform, the proper-tube rather broadly cylindrical, 0.7-0.9 em. long, about 0.3 em. in diameter at the base, puberulent-papillate without, densely vil- losulose toward the insertion of the stamens within, the throat tubular, 3.0-3.5 em. long, about 0.8 cm. in diameter at the ori- fice, densely puberulent-papillate within and without, the lobes obliquely obovate, obtusish, 0.9-1.0 em. long, somewhat patu- lous, puberulent-papillate without, minutely and densely pu- berulent within; anthers oblong-sagittate, caudate, obtusely 2—auriculate at the base, 1.3 em. long, the convolute apical ap- pendages minutely pilosulose, otherwise glabrous, the filaments filiform, 3.5 em long, minutely pilosulose toward the base; ovaries oblongoid, rather gradually produced into the style, about 0.35 em. long, glabrous; stigma subcapitate, basally 5- manieulate, about 0.3 em. long; nectaries compressed-ovoid, somewhat shorter than the ovary; follicles unknown. GUATEMALA: ALTA VERAPAZ: quebradas secas, at 2000 ft. alt., in jungle, June 8, 1920, Johnson 200 (US, TYPE, MBG, photograph and analytical drawings). The Key to Genera (p. 20) may be amplified to include Tintinnabularia as follows (under A.): B. Inflorescence bostrychoid, di- or trichotomously compound. C. Corolla infundibuliform. D. Calyx-lobes conspicuously foliaceous; staminal filaments filiform, eonspieuously longer than the anthers; leaves foveate in the axils he midrib beneath...... get XXIX. TINTINNABULARIA DD. Calyx-lobes scarious or only slightly foliaceous; staminal pee ol eylindrical, shorter than the anthers; leaves not fove Deeg tio I. ALLOMARKGRAFIA Tintinnabularia is of great interest because of its obvious affinity with the Asiatic genus Beaumontia, a widely cultivated (562) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 391 liana bearing large, infundibuliform corollas with conspicu- ously foliaceous calyx-lobes, filiform staminal filaments much longer than the anthers, an anomaly in the Apocynaceae, and leaves which are foveate in the axils of the midrib. Tintinna- bularia differs from Beaumontia, however, in the somewhat smaller corollas with tubular throat and shorter lobes, anthers with convolute apical appendages, carpels which are not united save when immersed within the receptacle, maniculate stig- mata, and leaves which are glandular at the base of the midrib. Although the fruit of Tintinnabularia is not yet known, it is doubtful that it agrees with that of Beawmontia which is es- sentially synearpous until the dehiscence of the valves. The specific adjective commemorates Mr. C. V. Morton, who called to my attention the type specimen which had been laid amongst the undetermined exsiccatae of the United States National Herbarium. INDEX ro ExsiccATAE Italicized numerals refer to collectors’ numbers, s.n. (sine numero) to unnumbered collections; parenthetical numerals refer to the numerals of taxonomic entities on in ca revision—the Roman to genera, and the Arabic to species. Vari- are undesignated. Genera Nos. I-IV are contained in Ann. Mo. Bot. Gard. 20: Dee (1)- (186). 1933; Nos. V-IX, in 22: 153-306. (187)-(340). 1935; Nos. X-XXIX, in 23: 169-391. (341)-(563). 1936. Abbott, E. K. s.n. (XIX 1). André, E. F. 1173 (III 91); 1942 (III Abbott, W. L. 2275 (XX 6). 96); 2488 (II 6); 2497 (III 85); Abrams, L. R. 1920 (XIX 1). 3590 (III 93). Andrieux, G. 248 (III 32); 249 (III 34); 251,398 (XXVI a Anisits, J. D. s.n. (XXVI 47). Appun, C. F. 1771 (II 1); 1838 (III m 1914 Pos 106); 2351 (III 86). Ackermann, —. s.n. (III 30; IV 6; Andrews, S. s.n. (III 9). VI 4) Acuiia, J. 3795 (II 10). Aguiel, Bro. 10408 (III 11). Alexander, R. C. s.n. (III 9; V 43). Allart, —. 203 (III 81). raujo, — ). Allemáo, F. s.n. (V 34). pé laez, E. ei Mo (XXVI 11). Alprato, —. 40E (XXVI 20). Archer, W. A. 573 E 85); 1714 (III Altamirano, F. 1639 (IIT 11). 97); 1749 (XXVI 3 Anderson, —. 60 (VIII 10). Arechavaleta, J. s.n. be 8); 18 (V 30). Andersson, N. E. 79 (V 27; XXVI 18). Ariste-Joseph, Bro. s.n. (XXVI 15). (563) 392 pi G. s.n. (III 11, 12) ; 2668 (III ud W. 2354 (VII 1). Austin, Mrs. R. M. s.n., 370, 376 (XIX 1 M Austin, Mrs. R. M. & Mrs. C. C. Bruce. 2422 (XIX 1). Bailey, —. 104 (III 86). Bain, S. M. 428 (VII 1). Baker, C. E. sn (XIX 1); 84 (IX 1); 165 (III 96); 2180A (XIX 1). er E 1353 (III 76); 1354 (III 74); 9 (V 30); 13704, 1370B (V 34); pis 13714 (V 23); 1872 (II 3); 1376 (XXVI 47); 4606 (V 30). Bang, M. s.n. (III 72); 249 (III 73); 274 (V 25); 402 (III 24); 403 (XXVI 59); 461 (III 23); 551 (III 79); 850 (V 20); 855 (V 34); 1065 (III 40) ; 1120 (III 39, 40) ; 1689 (V 20); 2053 (XXVI 8); 2056 (IX 26); 2057 (III 79); 2267 (XXVI 5); 2271 (III 20); 2404, EH (EMG): Bangham, —. 494 ie Bartlett, A. W. GE Gua Bartlett, H. H. 10873 que 1); 12366 Bassett, —. s.n. (XIX 1). Bere Na R sn (VIEL): Berlandier, J. L. 3197 (IV 5). M G. & R. Cario. 1821 (III a Bertero, C. s.n. (V 14, 43). Be 2 943 (V 34). Beyrich C. sn, (XXVI 50). Billberg, J. I. s.n. (XXVI 44); 254 (XVII 1). Bireher, F. W. n. (III 81) Blake, 3 F. a (HI 81). Blanchet, J. S. 677 (III 86); 1387 (III 96); 1578 (VIII 6); 1679 (III 49); 1738 (VIII 6) ; 1745 (V 2) ; 2452 (V 23); 2652 (V 23); 2807 (III 30); 3023, 31854 (III 86); 3229 (V 27); 3370 (VI 4); 3373 (IV 10); 3382 (III 72); 3406 (III 30); 3431 (V 27); 3467 (III 86); 3513 (XXVI 2); [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN 3635 (VI 4) ; 3636 (III 86, 96) ; 3637 (III 86); 3757 (V 28); 3776 (XXVI 2 3797 (III 86, 96); 3960 (III am —. 181 (V 45). leBlond, J. B. s.n. (III 78, XXVI 49); (?) 379 (V 9); (?) 381 (V 12). Bogusch, E. R. ME d 5). Bolland, G. s.n. 86). Boon, H. 1044 = 10); 1120 (III 78 ). ico J. 348 (III 76) ; 427, 693 (XVII 3). m —. 6302 (VIII 11). Bo oed 1500 (XXVI 55); (HT Brace, F^ K. 493 (X 3); 1492, 3588 (XX 6); 3601 (XVI 1); 4053 (X 3); 4373 (XVI 1); 4399 (XX 6); 4670 (X35 d (XX 6); 6177 (X 3). 2710 Brade, A. C. 5688 (XIII 4); 5696 (XXVI Ze 6700 (III 43); 9672 (V 47). Brandegee, T. S. s.n. (IV 2, 3; XIX 1). Brenes, A. 14272 (V 32); 14273 (III 757 8 (XIX 1). Bridges, T. s.n. (III 40). Brito, S. 43 (XVII 3). Britton, N. L. 423 (XVI 1); 640 (II 1); 641 (V 43); 961 (XVI 1); 1008 (III 9); 2080 (II 9); 2110 (XVI 1); 3127 (III 9); 3196 (V 43); 7159 (XVI 1). Ee N. L. & L. J. K. Braee. 526 (X mods E L. & E. G. Britton. 2900 Britton, N. L., E. G. Britton & J. F. Cowell. 12848 (XV 1); 13085 (II ES Britton, N. L., Britton & W. Freeman. 2195 (III 96). Britton, N. . G. Britton € C. $. Gages, Se e 9). Britton, N. L., E. G. Britton & J. A. Shafer. 91 ely ae 176 (XX 6). Britton, N Britton & A. M. Vail s.m. ur 5 (564) 1936] WOODSON—STUDIES IN Britton, N. L. & W. E. Broadway. 2626 (III 81). Britton, N. L., W. E. Broadway & T. E. AEN 310 (III 96). Britton, N. L. & J. F. Cowell. 365 (XVI 1); 12704 (II 9); 13290 (X 1). Britton, N. L. & F. S. Earle. 7652 (V 46). Britton, N. L., F. S. Earle & P. Wilson. 4 II 9) Britton, N. L. & W. C. Fishlock. 946 (XV. Britton, N. L. & T. E. Hazen. 717 (IX 21 » Britton, N. L. & C. F. Millspaugh. 2511, 2970, 5783 (X 3); 5887 (XX 0); 6093 (XVI 1 Britton, N. L. & J. A. Shafer. 911 (XVI ly. Britton, N. L. & P. Wilson. 10 (V 41); 59 (XX 6); 190 (XVI 1); 327 (XX way, W. E. s.n. (II 1; III 81, 96); 643 Kë 78); 826 ES 96); 2228 (III 88); 2582 (XXVI 7); 2592 (III 96); 2644 (III 88); 2707 (XXVI 7); 2730 (XXVI 14); 2799 DE Ede (III 88; I 14); 3122 (XXVI i 3373 (XXVI 14); 3375 Ge 96); 3848 (III 81); 4150, 4350 (XXVI 14); 4350 (XXVI 7); 5243 (IX 21); 6738 (II 1; III 8); 7369 (III 88); 10009 (XXVI Be W. 88 (II 7); 176, 339 (V 46); 1237 8); Buchtien, O. s.n. (XVIII 1); 250 (III 24); 277 (III 23); 277, 279 (XXVI 59) ; 590 (III 52); 609 (III 23) ; 1197 (ITE 90); 14195. CEDE 70); 17582 (TIE 57); 1743 (XXIV 1); 1968 (III 79) ; 3229 (III 53); 3677, 3876 (V 25); 8908 (III 73); 3953 (V 20); 4032 (III 73); 4088, 4140 (III 52); 4371 (VI 1); 4372 (XXVI 5); 4671 (III 24); 4672 (III 79); 4673 (III 96); 5099 (III 23); 5100 (III 90); 5918 THE APOCYNACEAE. IV 393 E 79); 7439 (III 23); 7441 (III 90). Burchell, W. J. 3255 (III 96); 3303, 3482 (III P 5946 (III 96); 8209 (III 30); 8510 (II 1); 9356, 9550 III 96); 10026 (III 86); 63961 (III 72 » Bush, B. F. 3742, 3751, 6238 (VII 1). Byars, J. s.n. (VII 1). Calderon, S. 815 (III 81); 938 (III 2); 2176 (V 41). Campbell, —. 5831 (V 43). Cardefias, M. 1144 (III 73); 1748 (XXVI 8); 1894 (XXIV 1); 1928 (XXVI 47); 2042 (II 4). Chapman, A. W. s.n. (VII 1). Chase, A. 7945 (III 30). Chaves, D. 313 (V 41). Chodat, R. & —. Vischer. 205 (IV 7). Christ, P. E. 1904 (II 7) ; 1937 (V 46); 48); 35 (XXI 2); 84 (V 3); 99, 100 (III 72); 103 (III 74); 106 (III 30); 108 (IV 9); 165 (III 30); 172 IV 10); 333 (IV 9); 334 511 ) IV 9); 1369 (III 96); 1673 (IV 9). Clemens, M. S. s.n Combs, R. 13 (XVI 1); 17 (XX 6); 78 (XVII 1); 138 (II T 524 (V 46). Commons, A. s.n. (VII Bo De pear SE 35); 8675 (V 24). Conzatti, C. 578 (III 32); 1837 (III 12); 3514 (III 81); 4277 (III 12); 4541 (III 4). Conzatti, C. & V. Gonzales. 1198 (III 12). be F. sm. (VII 1). Cook, O. & R. F. Griggs. 273 (III va s ae 81). (565) 394 Cooper, G. P. III. 23? (XXVI 39); X 19) Cooper, G. P. III & —. Slater. 113 (VIII 20). Cordoni, I. R. s.n. (XVIII 1). Costa, M. da 253 (VIII 10). III 17). 105 (XXVI 44); 134 Crawford, H s.n. (IV 2); 714 (V 14); $21 (II Crueger, E sn, (III 26, 96); 39 (III 88). Cruz, J. S. de la. s.n. (III 88); 1116, 1157 (VIII 3); 1299 (III 88); 1335 (III 78); 1347 (III 88); 1360 (VIII 8); 1417 (XVII 1); 1418 (IX 19); 1419 (XXVI 46); 1474 (III 88); 1476 (VIII 3); 1489 (XXVI 46); VEL > da M (VITI 3); 1990 (III 96); 2351 (IX 23); 2631 (XXVI 9); 3097 (XXVI 21); 3100 (III 96); 3135 (VIII 10); 3242 (VIII 3); 3274 (XVII 1); 3473 (VIII 10); 3703 (IX 23); 3542 (IX 13); 3924 (V 1); $935 (VIII 3); 4083 (VIII 10) ; 4154 (III 96) ; 4211 (IX 23); 4232 (VIII 3); 4584 (V ER Cumming, —. 51 (III 9). Curran, H. M. 158 (III 86). Curtiss A. H. 187 (X 8); 178: (XVI 1); 178, $95 (XV 2); 1102 (II 9); 2266 (XX 6); 2267 (XVI 1); 5448 (XX 6); 5893 (VII 1). Dahlgren, B. E. & E. Sella. 74 (IX 23) ; 96 (VIII 10). Dawe, M. T. 30 (III 49). Deam, C. C. s.n. (VIII 20); 3 (III 96); 6320 (XVI 2); 6368 (V 41); 6376 (XX 3); 37581 (VII 1). De Jonghe, —. s.n. (XXVI 59). Delesa, —. 1653 (III 11). Demaree, D. 9392 (VII 1). Dent, —. s.n. (IV 9). Diquet, L. s.n. (XXV 1). [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN edi BR. A-7491 (VII 1). Drummond, T. s.n. (II 9; VII 1). Dry mt —. 1042 SEN 41). Ducke, A. 8434 (XI 2); 11297 (III 30); 11393 (XI ya 11894 (VIII 1); 16075 (III 30); 17464 (VIII 8); 17465, 17466 (VIII 5); 17478 (V 10); 21618 (II 1); 21614 (V 13); 21622 (VI 1); 21624 (IX 1); 21625 (VIII 8); $1627 (V 19); #1628 (XXVI 46); 21634 (XXVI 36); 21639 (V 10); 21641 (VI 1); 21644 (IX 21) ; 21645 (IX 22); 21646 (III 86); 21647 (IX 12); 21650 (IX 23) ; 21758 (V 10) ; 21763 (VIII 10); 21766 (V 10); 21768 (III 79); 21772 (VIII 10) ; 21773, 21774 (VIII 11); 21775 (VIII 10); 21784 (VIII 4); 21809 (III 69) ; 21811 (IV 6) ; 21821 (IX 23); 21822 (VIII 8); 22429 (V 19; 22430 (VI 1); 22431 (V 18); 432 oe T 338952 (VIII 8); Hid (VIII 17). Duckett, F. s.n. (XV 2). Dudley, W. R. s.n. (XIX 1). Duges, A. s.n. (III 7, 11); 90 (III 7); 242 (IV 3). Dusén, P. s.n. (III 66, 69; IV 8; XXVI 2); 1317a (III 76); 1350a (III 74); 1090a (IV 8); 2939 (XXVI 2); 3564 (III 67); 5097 (V 27); 6988 (III 43); 7311 (III 76); 7409 (III 66); 7445 (V 47); 7648 (III 75); 7558 mun 1); 7699 (XXVI 59); 7781 3); 7884 (XIII 1); 8038 T 2); 8681 (III 67) ; 9086 (III 25); 9098 (III 75); 9487 (V 30); dou a 10); ); (III 74); 10851 (XIII 1); SE per (III 76); 11038 (XXI 4); 11449 (XXI 2); 11486 (XXVI 3); 11624 (XXVI 59); 11928 (III 43); 11988, 13430 (III 67); 13454 (III 74); 13652 (V (566) 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 27); 13696 (XXI 2); 13708 (III 43); 14272 (V 8); 14384 (III 67); 14632 (V 27); 14749 (III 74) ; 14750 (III 76); 15653 (IV 7); 15913 (III 72); 15946 (III 25) ; 16023 (III 76); 16031 (IV 10); 16454 (XXVI 59); 16608 (XXI 4); 16662 (V 8); 16684 (XXI 2); 16835 (XXI 4); 16998 (III 25) ; 17356 (III 76) ; 17456 (IV 10); 18012 (III 69); 18056 (IV 8). Duss, P. 2841, 3713 (XXVI 7). Dutra, —. 301 (XIII 1). Earle, F. S. & C. F. rl 2451 (II 9). Eastwood, A. s.n. Edwall, L. E. 11147 a me 11242 (VI 1). Eggers, H. s.n. (III 81); 438 (V 45); $36 (XVII 1); 1155 (III 96); 1638 46) ; 3389 (II 7) ; 4315 (XV 2) ; 4707 (XXIII 1); 5297 (XV 1); 5347 (V 41); 5482 (III 81); 5546 (III 96); 60) ; 14782, 14956 (XXVI 18); 15078 (XXVI 27); 15430, 15430bis (XXVI 16); 15618 (V 26); 15684 (XXVIII 1). Eggert, H. s.n. (IV 5; VII 1). Ehrenberg, C. A. 159 CH j ); 1869 (III 11 ). Ekman, E. L. s.n. (II 9); 659 (XXIII 2); 847 (V 46); 1083 (II 8); 1297 (V 41); 1355 (V 46); 1533 (XX 6); 1548 (V 46); 1580 (II 8); 1585, 1586 (V 30); 1591 (IV 8); 1592 (XXVI 59); 1594 (III 25); 1937 (V 46); 2426 (XXIII 4); 2431, 2664 598, 6762 (XXIII 4); 7191 (XXIII 2); 7549 (XX 6); 8020 (V 41); 8173 (II 7); 8289 (V 46); 8590 (567 395 (XXIII 2); 8698 (II 9); 8730 (V 46); 9051 (XXIII 2); 9314 (V 41); 9460 (XXIII 2); 9827 (V 46); 10224 (XVI 1); 10835 (II 7); 13037 (XXIII 1); 13334 (V 41); 15311 (XXIII 1); 15334 (II 10); 16390, 16409 (XXIII 1 Elias, Bro. 222 (V 41). Endlich, R. 1075 (III 11). Endres, M. 235 (III 96). ngle, —. s.n. (III 46, 69, 72, 74, 76, 77; IV 6,9; V 47; XXI 4 licvendberg, L. C. 127 (XXVI 55); 217 (III 81); 240 (III 11). Everts, M. M. s.n. (VII 1). Byerdam, W. J. 321 (XVII 1) ; 414 (II 7); 431 (XX 6). Fairchild, D. s.n. (XX 6 Faris, J. A. 119 (XVI 1); 568 (II 7). Faweett, W. s.n. (III 9). Fendler, A. hs (VIII 20) ; 250 (XXVI 44); 257 (XVII 1); 622, 624 (III 81); 624 (III 96); 1028 (III 37); 1030 (III 26); 1031 (II 1); 1032 (III 91); 1033 (III 81); 1052 (XXVI 47); 2111 (XXVI 14); 2381 (V 47); 2382 (III 98). Ferris, R. S. & C. D. Dunean. 2530, 2939 (IV 5) Fiebrig, K. 64 (XXVI 59); 64a (XXVI 6); 218 (V 30); 218a (V 23) ; 310 (III 74) ; 387, 474 (IV 10); 640, 647 (XXVI 47); 2153 (III 41); 2310 (V 30); 2336 (III 41); 2381 (V 47); 2455 (III 40); 2550 (III 24); 2675 (XXVI 59); 4157 (V 30); 4183 (IV 10); 4238 (III WE 4313 (II 3); 4447 (XXVI 59); 4462 (XXVI 47); 4541 (XXVI 8); 4545 (IV 8); 4601 (XXVI 2,8); 5162 (V 23); 5380 (V 47); 5677 (III 76); 5841 (XIII 2); 5947 (XXVI 59); 6267 (V 34); 6373 (III 25). Firmin, C. 241, 608 ims 17). Flint, M. B. s.n. Focke, W. O. 1056 6] 20). Forrest, —. s.n. (XXVI 35). 396 Frazáo, A. 7144 (XXVI 19); 7147 (V 27); 8674 (V 34). Fredholm, A. 3163 (XVI 1). Friedrichsthal, —. 138 (VIII 20). Fries, R. E. 12? (III 23). Fuertes, M. 34 (II 7); 220 (XVII 1); 340 (II 8) ; 410 (V 46); 453 (XXIII 3). Funck, N. € L. J. Schlim. 310 (V 29). Gabriel, —. s.». (III 96). Galander, C. s.n. (IV 8). Galeotti, H. 1565 (XXV 4); 1575 (III 81); 1579 (III 1); 1582 (III 32); 1593 (IV 3); 1602 (III 81); 1604 III 8 Garber, A. P. sn. (XVI 1; XVII 1); 11957 (XV 2). Gardner, G. 249 (III 69); 250 (III 47); 530 (V 8); 535 (III 96); 538 (III 43); 894 (III 86); 951 (VIII 10); 1059 (V 27); 1060 (XXI 2); 1761 (V 23); 1762 (VI 4); 1961 (III 96); 3311 (III 72); 3312 (IV 7 (VIII 7); 6058 (III 86) ; 6060 paca p Gaudiehaud, C. 533 (XXVI 35); 534 . ); 653 es 1); 815 (XXII 1); 817 (XX 4); (III 9, 81); 883 (III 9); 1158 SC 81); 1979 (XX 4); 2013 (III 9, 81); 2014, 2228 (III 81); 23423 (III 9); 23969 (III 1). Gaumer, G. F. & Sons. 23335 (XVII 1). Gay, C. 384 om 15 2 (III 26). 3 (XXVI 48); 3185 (III 86); es a 5); 4555 (V 34) ; 4652 a E Gentle, P. 439, 1022 (XXI 1); 1037 (XVI 2); 1281 (V 44). Ghiesbreght, A. 148 (III 1). Glatfelter, N. M. s.n. (VII 1 [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN Glaziou, A. s.n. (III 69, 72, 74, 86) ; 635 (III 69); 1989 (VIII 7); 2091 (III 4080 (V 48) ; 4086 (III 58); 4092 (V 34); 4879 (XIII 4); 4880 (V 2); 4881 (XXVI 35); 4934 (VIII 6); 5934 (VIH 7); 5938 (V 8); 5943 ( Wd 7 (XIII 4); D GE o. 7755 iud 58); 7756 (V 47); 7757 (V 35); 8170 (III 43); 8171 (III 47); 8173 (XXVI 50); 8363 (XXVI 50); 8796 (III 96); 8799 (V 8); 8800 (XXVI 47); 8803 (III 64); 9507 (IV 9); 9511 (VIII 7); 9932 (V 10); 11179 (VI 1); 11180 (III 86); 11187 (III 58); 11188 (XIII 4); 11190 (III 67); 11195 (III 96); 12939 (V 23); 12941 (VI 4); 12943 (XXVI 12); 12944 (XXVI 50); 12951 (XXVII 1); 12954 (XXVI 35); 12955 (III 63); 12957 (IX 9); 14061 (III 67); 14062 (III 70) ; 14063 (V 27); 14064 (V 24); 14065 (V 8); 14072 (XXVI 12) ; 14082 (VIII 11) ; 15214 (IV 9); 15215 (III 64; IV 9); 15216 (III 76); 15217 (III 72); 15218, 15219 16250 (III 30); 17134 (III 72); 17135 (III 70); 17136 (III 30); 17138 (V 34); 19623 (II 5); 19625 (XXVI 47); 19627 (III 67); 20412 (XXVI 8); 20416 (IX 17); 21718 (III 96); 21720 (VI 4); 21721 (III 31); 21724 (III 72; V 27); 21731 (IV 9); 21732 (IV 6); 21733 (IV 0) Gleason, H. A. 336 (XXVI 9); 774 (III 86 Goeldi, A. 21767 (III 86); 21777 (IX 10). Goldman, E. A. 646 (XX 6). Goldsmith, —. 126 (IV 3). Goll, G. P. 95 (III 81). Gomez, R. 777 (III 3). (568) 1936] WOODSON—STUDIES IN Gonggrijp, —. 2135 (VIII 10). Gonzalez, V. s.n. (III 32). Goodman, d J. & C. L. Hitchcock. 1720 XIX Goodwin, s. G. & B. W. Goodwin. 4, 16 (XVI Goudot, S 2 (II 6). III 86). Guedes, M. 21640 (XXVI 46); 21800 (IX 5). Guilding, —. s.n. (VIII 2). Guillemin, A. 168 (V 27). Gunckel, H. 65, 2361 (XVIII 1). Gwynne-Vaughan, D. T. 31 (III 107); 33 (XI 4 Hall, H. M. s.n., 1231 (XIX 1). Hamilton, W. s.n. (III 96). Hanson, H. C. 752 (IV 5). rper, R. M. 794 (VII 1). TE W. 5908, 7042 (V 43); 7237, 8285 (III 9); 8440, 8599 (V 43); 9007, 11732, 12085 (XVI 1); 12109 (XX 6). Hart, —. 128 (VIII 20). Hart, J. H. s.n. (III 9). Hartmann, C. V. 15? (IV 2). Hartweg, T. s.n. (XIX 1); is (Ey 39; 492 (III 32); 669 (III 81); 670 (XXVI 18); 1053 (XXVI 54); 1195 (II 6). Hasse, H E. s.n. (XIX 1). Gem E. 417 (V 30); 1088 (V 23); 0 (V 30); 1577 (V 34); 3032 (V Sie 3061 (XXVI 12); 3211 (XXVI 59); 3345 (V 30); 3494 (V 34); 3538 (IV 8); 3583 (XXVI 6); 4499 (III 74); 4528 (V 30); 4724 (IV 10); 4789 (III 76); 5136 (V 30); 5733 (III 75); 6001 (XXVI 8); 6362 (V 23); 6424 (XXVI 59); 6658 (III 74); 6776 (XXVI 50); 6816 (III 74); 7125 (III 76); 7204 (III 55); 7621 (II 3); 7820 (XXVI 42); 8065 (XXVI 12); 8143 (III 77); 8172 XXVI 8); 8678 (XXVI 59); 8765 397 (XXVI 6); 8768, 8825 (IV 10); 9081 (IV 8); 9609, 9801 (III 76); 11420 (VI 1); 11477 (IV 8); 11547 (XXVI 42); 11962 (V 30); 12166 (XXVI 59); 12229 (V 23); 12326 (V 30); 12527 GA 6). Hauman, —. 3591 (XVII 3). Havard, V. s.n. a ina 16 (II 9). Hayes, S. s.n. (III 96; V 32; XXII 2); 55 (VIII 20); 98 (III 96); 116 (III 82); 148 (II 1); 167 (III 96); 345 e 81); 450 (XXVI 44); 663 (IX THE APOCYNACEAE. IV nde T. E. & E. P. Killip. 11168 (III 91 ). Heilborn, O. 771 (XXVI 40). Heller, A. A. 18229 (XIX 1). endorff 36 (XXI 4). Herrera, F. L. 3123 (III 40) ; 3282 (III 73). Herter, W. 570, 3128 (IV 8); 3128 (IV 10). Herzog, T. 742 (III 54) ; 1512 (XXIV 1); 1514 (XXVI 47); 11230 (XXVI 59 ftejde, H. T. 152 (III 3) ; 172, 318 (III 81); 749 (III 3). Heyde, H. T. & E. Lux. 3992 (III 81); 3993 (III 3); 3994 (II 1); 4496 (XXVI 55); 4497 (XXVI 58) ; 4533, 4534 (V 32); 4539 (XXVI 43); 4540 (III 5); 6186 (III 81); 6345 (XVII 20069 (XXVI 18 20495 (III 37); 20593 (XXVI 18); 20595 (II 1 Hoehne, F. C. 385 (IV 10); 695 (III 77); 834 (III 70); 896 (III 74); 1200 Geck 12); 1270 (II 5); 1512 1 (V «d (V 35) ; 2857 ar d ; $185 (III 96) ; 4697 (XXVII 1); 5094 (IV 6); 6251 (569) 398 (VI 4); 6623 d 35); 6633 (III 96); 6634 (IV 10 715 (II 5); 6874 (XXVI 48); 11148 (V 27); (III 86) Hoehne, F. C. & A. Gehrt. 17319 (XXVI 35 11153 JE Hoffman, C. 522 (XXVI 57). Hohenkerk, L. S. 680 (III 96). Hollermayer, P. A. 326 e 2c Hollister, M. P. 46 (VII Holmgren, I. $46 (III b 920 (III 7). Holt, E. G. & E. R. Blake. 578 (III 86) ; 592 (IX 15); 593 (III 86) ; 594 (IX ; 789 (III 86); 819 (III Holt, E. G. & W. Gehriger. 202 (III 86); 269 (II 1). Holway, E. W. D. & M. M. Holway. 556 (III 40). Hopkins, M. H. s.n. (VII 1 Hostmann, F. W. 55 (II 1); 269 (VIII 946 (III 96) ; 981 (XXVI 46); 6118 (V 9). Hostmann, F. W. A. Kappler. s.n. (III 86; XXVI x 123 (III 78); 569 (V 9); 1203 (IX 10); 1236 (V 9). Houston, W. s.n. (III 21762 (11 1); 21770 (VIII 10) Hulk, J. F. 303 (XXVI 8). von Humboldt, A. & A. Bonpland. s.n. (II 6; III 16, 81, 85, 95, 107) ; 1296 (V 41). Hyams, M. E. s.n. (VII 1). Ihering, H. V. 8, 300 (V 30). Ingalls, T. J. s.n. (VIT 1). Jack, J. G. 6614 (V 46); 6730, 7397, 8013 (V 41). Jaeger, B. s.n. (II 7). [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN Jameson, W. 101 (III 17); 153 (III 18). de Jelski, C. s.n. (IX 21); B (III 29) ; 336 (III 17); 375 (III 15). Jenman, G. S. 151 (III 86); 447 (III 96); 739 (III 86); 799 (II 1); 1224 (III 106) ; 1313, 3867 (III 96) ; 4058 5550 (III 86); 6316 (V 40); 6965 1). Jennings, O. E. 463B (XVI 1 Johnson, H. 88 (II 1); $9 (IIT 96); 200 (XXIX 1); 410 (III 1). Johnson, W. s.n. T): Johnston, I. M. 1428, 1431, 1550 (XIX 1). Johnston, J. R. 53, 137 (III 81); 357 (V 45); 3807, 3888, 3984 (IV 1). le Jolis, —. s.n. (II Jones, M. E. s.n. (IV 2; XIX 1). Jones, V. s.n. (XIX 1). Jónsson, e 2a, 91a (XXI 2); 286a (XXVI Joor, J. SE am (VII 1). Jørgensen, P. 1605 (III 24); 1800 (III 41); 1956 (V 30); 2605 (III 56); 3450 (XVII 3); 3573 (V 30); 4038 (XVII 3); 4182 (XXVI 6); 4708 (XXVI 47); 4711 (XIII 2) Joseph, Bro. C. 2397, 3334, 3497, 4409 (XVII). Julio, Bro. 15, 128 (III 53). Jürgens, C. 29 (V 34). Kalbreyer, e 261 (II 1); 887 (96); 1955 (III 92). Kappler, A. 444a (VIII 10) ; 1449 (IX 21); 1605 (III 96) ; 1825, 1955 (VIII 10); 1989 (IX 19); 2087 (VIII 10); 2089 (VIII 11). rsten, H. an (III 17, 95). Karwinski, W. F. von. 269 (III 33). Kearney, T. H. Jr. 1874 (VII 1) Kegel, — 174 (XXIV 1). Kellermann, W. A. 7564 (III 81). Kellogg, J. H. s.n., 15263 (VII 1). (570) 1936] WOODSON—STUDIES IN Killip, E. P. 11409 (III 37); 11602 (III 91); 11775 (III 96). Killip, E. P. & T. E. Hazen. 9605 (III Killip, E. P. & A. C. Smith. 14370 (II 1); 15469 (III 91) ; 16181 (III 86); 16308, 16385 (XXVI 8); 17054 (III 91); 20568 (V 42); 22708 (XXVI te te e Get, Sage, Ze £o "ut Ki (VIII 13); 27493 (XXIV 1); 27579 (III 99); 27840 (XXVI 8); 28609 (IX 6); 29847 (IX p baste M III 15); 29916 (IX 8); 3 X 2); 30151 (V 16); 30257 Sos $2 30422 (III 86). Klug, G. 57 (III 83); 79 (VIII 15); 422 (IX 1); 541 (XXIV 1); 640 (V 13); 657 (IX 1); 774 (XXVI 30); 782 (IX 20); 915 (XXVI 30); 972 (VIII 13); 1026 (IX 19); 1053 (V 10); 1119 (XXVI 8); 1288 (III 83); 2761 (V 5) ; 2905 (V 21) ; 3066 (XXVI 31); 3104 (V 21). Koch ©. T. 70, 78 (LX 15). Koch-Griinberg, T. 30 (XXVI 23). Krug, L. 709 (V 45) Krukoff, B. A. 1247 (IX 23); 1427 (XXVI 34); 1457 (IX 19); 1545 (XXVI 30); 1854 (III 96); 1886 (XVII 1); 6335 (Addenda). Kuhlmann, J. G. 1234 (VI 1); 2252 (XI 2); 2832 (V a d (IX 23); X 19); 3254, : See (XXVI 47); 3647, 3648 (V 10); 3650 (XXVI 8); 3651 (III 96); 21849 (III 79); 21851 (XXVI 30) ; 21854, 21855 (III 86); 21856 (IX 19); 21858 (IX 10); 21863 (V 1); 21868 (VIII 5); 21873 (VI 1) ; 21874 (III 86) ; 21877 (VIII 7). Kuiper, —. 39 gens ul 40 (IX 21). Kunth, —. s.n. (XXVI 7). Kuntze, O. s.n. Ma Wé IV 10; V 22); 601 (III 96); 689, 1331 (III 81); 1891 (III 82). THE APOCYNACEAE. IV 399 Lagerheim, G. von. s.n. (III 17). Landlouw, D. 297 (III 78). Langlasse, E. 261 (V 41); 519 (III 81). Langlois, A. B. s.n. (VII 1). Lawrance, A. E. 241 (XXI 1); 453 (V 4); 534 (I 2) ; 710 (III 45). Lehmann, F. C. 439 (III 17); 704 (III 85); 3809 (II 6); 4801 (III 16); 5645 (II 1); 7580 (III 91); 7581 (III 85); 7885 (V 37); 8483 (III 37). Leithold, —. s.n. (XIX 1). Lemmon, J. G. an (IV 2; XIX 1); 119, 18? (XIX 1). Leon, Bro. 712 (XVI 1); 713 (XX 6); 3353 (X 1); 5108 (XVI 1); 5211 (XX 6); 13127 (X 1); 13138 (XX 6 . ); Leon, Bro. & F. R. Cazañas. 5974 (II 10). Leon, Bro. € G. M. Fortun. 8649 (II Leon, Bro. € Bro. Roca. 8237, 8904 (X 1 js Leonard, E. C. 5159, 5251, 7155, 9838 (II 7); 8048, 8332 (II 8 Leonard, E. C. & G. M. CHEN 11642 (XVI 1); 11735 (II 1); 11923 (XX 6); 14013 (IL 8); 14070 (XX 6); 15112 (II 8); 15543 (V 46); 15682 (II 7). Letterman, G. W. s.n. (VII 1). Levy, —. 1074 (XX 3). Liot J. s.n. (V 47); 29 (V 24); 55 (V 34); 74 (V 27); 224 (III 86). Liebmann, F. M. 11959, 11960 (III 1); p (III 33); 11976 (III 34); 11978 (III 1); 11981 (III 32); e (III 4). Lillo, M. 1314, 3200 (XXIV 1); 5988 (III 40). Lindberg, G. A. 191 (XIII 4) ; 193 (III 76); 194 E 72); Ec J. 267 (V 4); 359 5 (III E. 580 (II 1); 863 (III (571) 400 85); 967, 1503 (III 81); 1699 (XV 1); 1822 (V 46); 2167 (XV 2). Lindheimer, F. s.n. (VII 1); 128, 984 (IV 5). Lindmann, C. A. 957 (XXVI 2); 43161 (XXVI 13); Ce (XXVI 2); 43337 (IX 9); 43563 (IX 21). Lisbóa, A. 4786 (III 96). A. 282 (XXI 4 ; 339 (III 86); 622 (XXVI 2); 636 (III 86); 1141 (III 96) ; 11154 (III 86). Lorentz, P. G. 205 (III 41); 1207 (IV 8). Lorentz, P. CERAT); G. Hieronymus. s.n 216, 358 (V 23); 403 (III nd 102, —. 643 (III ee Ge l8 E (III 43; XXVI 47); 383 (III 30); 5000 (III 74); 6953 (III 70); 7155 (III 44) ; 12599 (III 30). Lund, P. W. s.n. (III 25, 74, 76; IV 6, 8,9; XXVI 17); 886 (IV 10). Lundell, c. L. s.n. (III 81); 515 (XX 6); 842, 1524 (XVI 2); 1350 (XX 4); 2540 (V 32); 3445, 3533, 4046, 4056, 4736 (XVI 2). Luschnath, B. s.n. (V 3, 34). Maebride, J. F. 2777 (II 1); (XXVI 14); 3730 (III 38); (XXVIII 1); 4720 (III 81); (X ps 8); 5473 (XXVIII 1); (III 93). Mae Nab, G. s.n. (VI 6). Malme, G. A. s.n. (11 5; III 84; IX 9; VI 12); 248, $64 (V 30); 502 (XXVI 2); 602 (V 30); 690 (XIII 1); 827 (XXVI 2); 1038 (V 23); 1058 (11 3) ; 1118 (VI 1); 1196 (III 84, 96); 1276B (XXVI 47); 1478 (XXVI 8); 1536 (IX 9); 1642 (IX 22); 1766 (IX 21); 2036 (IX 11); 2489 W 23); 3122 (XXVI 8); 3305 (IX 2 2785 5812 Mandon, rt 1472 (III 40). [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN Manso, e Ss S. s.n. (VI 1); 2 (TII 86); 398 74). Manso, p da S. & J. Lhotzky. 29 (I e 96); 33 oe 1255 44 (1100): 3 (XXVI Martin, J. s.n. (III 78; V 12; VIII 10; IX 19; XXIII 3). Martius, P. von. s.n. (III 30, 72 CH V 22) ; 292 (III 74); 293, 299 (III 72); 300 (III 46); 307 (III 71); 324 (VIII 7); 503 (III 74); 504 (III 76); 506 (III 69); 909 (III 46); 966 (VIII 7) ; 967 (VI 1) ; 1034 (IX 17); 2267 (III 30); 2663 (IX 5); 2960 (V 7) ; 3029 (XXVI 29). Matthews, A. s.n. (XXVI 18) ; 820 (III 23); 1327, 1977 (III 81); 1978 (III 22). Maxon, W. R. € A. D. Harvey. 6507 (II 1); 6527? (III 96). Maxon, W. R., A. D. Harvey € A. T. Val- entine. 7614 x 3). Max . P. Killip. 394 (XV 1; 1440 abs 1); 1616 (XX 6); 1675 (III 9). Maxon, "A R. & A. T. Valentine. 6972 (III 96). Mayerhoff, C. J. 28 (II 7). Me Clatehie, A. J. s.n. (XIX 1). Me Rae —. s.n. (VIII 7 Mearns, E. A. 1755 (IV 2); 1884 (IV 2). poe —. sn. (VIII 10); 460 (V 127. Mellichamp, J. is s.n. (VII 1). Mendonca, R. v 9). Mexia, Y. ne p 47); 538344 (VI 4); 4337 (XXVI 4); 5914 (V 47); 5994 5). Miers, J. 98 (III 96); 2418 (III 30); 3436 (III 43); 4020 (XXI 2); 4022 (III 96); 4027 (VIII 7); 4029 (III 43) ; 4031 (III 96) ; 4049 (V 8). Mikan, J. C. s.n. (III 49; VIII 7; XIII 4; XXVI 2) Millspaugh, C. F. 1102, 1113 (II 9). (572) 1936] WOODSON—STUDIES IN Rd sn. (VII Mo Moldenke, | H. N. 725 (XVII D: 399A (XV 2); 10064, 5784 (XVI 1). Montes, M. N. & A. E. Salazar. 486 (IV 3 d Moore, J. A. & J. A. Steyermark. 3243 (IV Moore, S. 819 (XXVI 30). Moritz, J. W. K. 31 (IIT 98) ; 1153 (III 81) ; 1899 (III 91). pires T. 380 (II 3) ; 420 (IV 10); A (IV 8) ; 810 (V 23) ; 895 (II 3). uen C. V. & —. Makrinus. 2348 (XXVI 44). Mosén, H. 622 (XIII 4) ; 947 (III 68); (XX 48) ; XXI 2); 3193 (III 43) ; 3433 (XXVI 3) ; 3434 (V 27); 4269 (XXVI 2); 4271 (V Moss, A.M. En. (11 06). Mee F. 94 (IV 2 ae (V 30). Munz, P. A. 6094 (X Mutis, J. C. s.n. T o 28); 97 (III 85). Nash, G. V. 185 (XVI 1); 136 (II 8); 138 (XX 6); 158, 286, y ER 7345 789 a 8) ; 940, 996 (II ash, G. V. & N. Taylor. MA (XVII); 1309, 1385 Gë 7); 1633, 3773 (XVI 1); 3890 (XX Nelson, E. W. p 866 (III 81); 3080 (XX 1); 3337 (III 2); 4658 (III 11); 4659 (IV 3); 4692 (IV 5). Wee do S. M. s.n. (III 69) ; 46 (VI "um G. E. p AES y^ Qs m A 3 (V 30); 92 965 Go e pe (V 34). o 9). 11202 zZ o = ME z^ C. (V 35); (XXVI 4); ra (XXVI 59). Nurse, H. A. 2158 (III 81). Oechioni, —. 4459 (VIII 7). d'Orbigny, A. C. V. 879 (IX 12). Orsted, A. S. 15510 (III 37) ; 15542 (II 1); 15544 (XX 1). THE APOCYNACEAE. IV 401 Ortega, J. G. 4712 (IV 3). Ortenburger, A. I. s.n. (VII 1). Osta, —. 1014 (III 48). Gegen C. wid hus T). —. 1 (XXVI 43). Pabst, C. 557 (V 27). Palmer, E. 28 (IV 3); 98 (III 7); 205 (IV 2); 238 (III 11); 259 (XXII 1); 296 (IV 3); 395 (III 11); 447 (IV 3); 571, 605, 697 (III 10) ; 702 1); 1028 Ge 55; 1127 (III 10). Palmer, E. J. X 6); 6063, 7958, IV 16542, 23132, 27116, 29164, 39314 Palmer, W. & J. H. Riley. 207, 837 (XV 2); 348 (II 9) ; 364, 833 (XVI 1); 842 (XX 6); 871 (II 9); 1010 (XVI 1); 1041 (XX 6). Parker, C. S. s.n. (III 96); 259 (III 86 Parry, C. C. & E. Palmer. 574 (IV 3). Passarge, S. 86 (III 100). Passarge, S. & —. Selwyn. 287 (IX 21). VI 16, 55; XXVIII 1); 389 (II 4). Pearce, y s.n. (III 23) ; 708 (III 60). Peek, M. E. 35 (III 81); 450 (V 11); 653 ES 38); 671 (VIII 20); 696 (III 96). Peckholt, T. 18 (IV 9); 559 (XXVI 2 Penn . 1330, 1890 (III 81); peu: 2 E 1633 (III 81) ; 2034 (III XVI 8); (XVII 3); 4155, 4181, 4617 (II 1); 5405 (III 91); 5604 (III 5719 (XXVI 22); 4 ( 95); 5719 10231, 10703, 10783, 10820 (III 91); 10884 (II 1, ed 12002 (II 1). Pennell, F. . Killip € T. E. Haz vie (IH 91). (573) 402 Perkins, J. R. 276 (III 9); 1100, 1378 (V 43) Perrottet, S. s.n. (II 1; III 78; VIII 10); 270 (IX 12) ; 272 (III 96). Persaud, —. ?4 (III 88). Piearda, L. 154 (V 46); 504 (XXIII 2); 813 (V 46). Piekel, D. B. 1222 (V 27 Pilger, R. 643 (IV 6); 670 (III 96); 799 ; Piper, C. V. 5850 (II 1); 5914 (III 96) Pittier, H. s.n. (III 96) ; 144 (III 3); 203 (III 96) ; 304 (III 81) ; 343 (III 96); 369 (III 37); 471 (II 1); 520 (III 93); 555 (III 96); 611 (III 95); 810 (III 81); 2498 (III 96); 2813 (VIII 20); 3147 (III 37) ; 3736 5490 (III 96); 6047 (XXVI 14); 6652 (XXVI 57); 6869 (II 1); 7300 (III 81); 7598 (III 82); 7638 (II 1); 8108 (III 61); 8540 (III 96); 8832 (XXVI 52); 8918 (V 4); 8920 (XIII 3); 9650 (III 81); 9730 (II 1); 9881 (IX 19) ; 9935 (XXVI 44); 10127 (III 98) ; 10760, 11212 (II 1); 11228, 11278 (III 81); 11551 (Ad- denda); 11597 (III 81); 12037 (II 1); 12401 (XXVI 44); 12556 (III 61); 12558 (III 81); 12759 (II 1); 12778 (III 81); 13121 (II 1); 13412 (XXVI 14); 13436 (III 96); 16016 (XXVI 44); 16137 (VIII 20); 16622 (V 32). Ke $ (II 4); 144 (III 81); 0-04 i 1233 (III 81) ; 1582 (VI ae 1840 (III 83) ; 2161 (XXVI 30); 2547 (VIII 10) ; 2657 (IX 10); 2946 Pohl, J. E. s.n. (III 21, 30, 58, 72, 74, 76; IV 9; V 27; VI 1; XXVI 47) 15 (III 43); 895 (IV 10); 970 (IV 6); 1383 (IX 9); 1592 (XIV 1); [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN 1845 (VI 1) ; 1846 (VI 4); 1898 (IX 22); 1899 (IX 9) ; 2189 (V 23) ; 2214 (XXVI 47); 2452 (XXVI 2); 5167 (XXVI 50); 5168, 5397 (XXVI 2). Poiteau, A. s.n. (II 1, 8; III 96; V 12). Pollard, C. L., E. Palmer, & W. Palmer. 2, 248 (II 9). Pollard, R. M. 40 (III 86). Porto, —. 7931 (IV 9); 8671 (VIII 7). Preuss, P. 7462 (III 96) ; 1617 (XXVI 47). Pride, —. s.n. (III 55). le Prieur, F. R. s.n. (III 78, 86, 96); 241 (III 78) ; 244 (III 96) Pringle, C. G. s.n. (III 8) ; 320 (IV 3); 690 (III 11) ; 694 (IV 5); 701 (III 11); 1108, 4393 (IV 3); 4662 (III 32); 4822 (XXV 3); 5422 (III 7); 6224 (XXVI 55) ; 6329 (III 7) ; 6341 (XXVI 55); 6554 (XXVIII 3); 6966, 7242 (XXV 2) ; 11014 (IV 3) ; 11015 (III 11); 11357 (III 13); 11838 (IV 4) ; 13106 (III 8); 13585 (XXV 4); 13760 (III 32) ; 13890 (III 8). Pulle, A. A. 78 (III 86); 291, 453 (V 12); 493 (III 78). Purdie, W. s.n. (II 1; III 9, 26, 81). Purpus, C. A. 851 (XIX 1); 1378 (IV 4); 1892 (III 10); 1985 (III 1); ~ el — mM en © w Q A e ~ $ Fi M ju w E o e Q ~ — ka ki 81) ; 6232 (XXII 1); 6929 (III 7274 (III 6); 72 e (III 81); 7665 X XVI 55); 10347 (V 41); 10739 (III 81); 10790 (III 1); 10885 (III 81); 11134 (XXVI 55). Quelch, J. J. & F onnell. 132 (III 89); 146 (IH E SE (III 89). Raben, (Graf) F. C. s.n. (III 74). Raimondi, A. $228 (XXVI 51). Ravenel, H. W. s.n. (VII 1). Record, S. J. s.n. (III 81). Regel, —. s.n. (V 24; VIII 6; XXVI (574) 1936] WOODSON—-STUDIES IN Regnell, A. F. 11 186 (III 25); 189 (III 96; IV 10); 280 (III 72); 287 77); 875 (III 68, 76); III 877 (V 35); 878 (IV 9); III 880 (IX 9); 881 (II 5) ; III 882 (V 30); III 883 (XIII 4); III 884 (XXVI 59); III 885 (XXVII 1); 1462 (III 86); III 1600 (XXVI 48). Reiehe, C. s.n. (III 11). ineck, E. M. 467 (V 30). Reineck, E. M. € J. Czermak. 467 (V 27) Reko, B. P. 3411 (III 81) ; 3511 (XXII 1); 3955 (III 32); 4823 (XXV 1); 5166 (III 11) ; 5212 (IV 4). Reverchon, J. 1378 (IV 5); 2558 (VII 1). Revirosa, J. N. 126, 225 (III 81). Bou E S. s.n. (VII 1). Richard, —. s.n. (VIII 11). Ricketts, —. s.n. (IV 2). Ridley, H. E T. S. Lea & G. Ramage. s.n. (III 96). Riedel, a an. (11 5; III 21, 30, 58, 72, 74, 76, 96; IV 8, 10; V 8, 22, 27, 34; VII SE 6:45 XIII 4; XXVI 2; XXVII 1); 60, 61 (III 43). Rimbach, A. 22 (II 1). itter, —. s.n. (XVII 1). Rodriguez, —. 605 (V 34); 3588 (V 30 ) von Rohr, J. P. Ban (III 78); 23 (XXVI 7). Roig, J. T. 109 (X 1); 3227 (X 2). Rojas, T. 2 (II 3); 2654 = 8); 9698 (V 23); 97684 (XXVI 59); 9838 (XIII 2); 10270 pats 8); 10652 V 47). Rose, J. N. 1716 (XXVIII 3); 3474 (IV 3); 16700 (IV 1). Rose, J. N., W. R. Fitch & P. G. Russell. 4292 (XX 6). Rose, J. N. & R. Hay. 6292 (IV 3). 1988, 23029 (III 35). THE APOCYNACEAE. IV 403 Rose, J. N., J. H. Painter & J. S. Rose. 9055, 9129 (III 10) ; 9510 (III 11). Rose, J. N. & J. S. Rose. 22592 (III 57). Rothrock, J. T. 146, 369 (III 9); 497, 646 (IV 2). Rothschub, E. 614 (III 81). ); = ps 1); 46); 397 (XXII Rusby, H. H. 2379, 2380 dd. 79); 2385 (III 23, 90); 2386 (III 40); 2387 (III 96); 2392 (VI 1); 2393 (II 4); 2394 (III 39, 73) ; 2526 (V 23); 2694 (III 73). Rusby, H. H. W. Pennell. 121 (XXVI 15) ; 404, 1032 (III 85). Rusb Squires. 20 ` i 3 Se 3); 293 (III 78); 302 (XXVI Së A. 482 (VII 1). Saer, —. s.n. Sagot, P. A. 1 (5 96); 382 (III 78) ; 383 dx 19); a (III 86) ; 387 (XVII 1); 392 (VIII 10); 886 (III 88); 1067 (V 12); 1143 (VIII 10). Ge : Sagra, R. s.n. (XV 2; XVII 1); 0 (II 9). D bic A. de. 2597 (IV 8). Salazar, A. E. s.n. (III 11 Salzmann, P. 318 (III 43); 320 (III 86) Sampaio, A. J. de. 238 (III 96). Samuels, J. A. 439 (V 9); 457 (III 88). andwith, N. Y. 283 (VIII 10) ; 334 (V 40): 405 (V ved 416 (V 38) ; 467 (V 1); 552 (IX Sartorius, K. s.n. qa "d Schaffner, W. 489 (IV 3 Schenck, H. 1875 (V 27); ` 1947 (XXVI 35) ; 2269 (V 27). Schiede, C. J. W. 167 (III 81); 448 (III 11); 488 (XXVI 55); 493 (III Schipp, W. A. s.n. (XXVIII 4; Ad- denda); 56 (III 96); 148 (XVII 1); 360 (V 11); 368 (III 81); 376 (XVII 1); 491 (XVI 2); 953 (VIII 20); (575) 404 S-7 (XXVI 56); 8-47 (II 1); S-388 (XXVI 42). Schlim, L. J. 510 (VI 3). Schomburgk, Rich. 200 (IX 21); 854 (V 15); 1438 (V 38); 1514 (V 9). Sehomburgk, Robt. H. s.n. (XI 3); 22 (II 7); 87 (V 9); 89 (VIII 10, 16); 130 (III 96) ; 183 (III 103) ; 309 (IX 19); 311 (II 1); 329 (XVII 2); 350 (III 87) ; 374 (XXVI 47) ; 383 (III 89) ; 421 (VI 1); 55? (V 9); 599 (VI 1); 608 (V 1); 665, 681 (VI 1) ; 707 (V 38); 713 (XXVI 9); 725 (V 15); 755 (XXVI 47); 782 (V 9); 788 (VIII 10); 800 (VIII 10); 821 (V Schornba ANAL, 755 (XXVI 19); 5398 (XIII 4); (III 43); 5976 (V 3). Schiich, C. G. de. s.n. (XXVI 4). Schultze, A. 193 (III 85); 245 (XXVI 5); 474 (XXVI 26); 605 (XXVI 5389 5404 49). Schunke, C. 389, 391 (III 79). Schwacke, W. 8754, 11854 (V 35). Seemann, O. 95 (III 82); 156, 161 (II I 96). . & E. Seler. 39 (III 3 m F. s.n. (III 25, 64, A 75; IV EE VTT G: XXVI 19, 35); 34 (III 62); 35 (XXVI 2); 36 (V 35); 169 (III 43); 170 (III 96); 211 (V 8); 818 (III 86); 217 (III 43, 96); 219 (V 8); 358 (V 8); 388 (V 27); 396 (V 22); 460 (V 28); 543 (V 8); 550 (XXVI 50); 635 (V 34); 693 (IV 8); 791 (III 43); 998 (III 96); 1313 (III 30); 1406 (XXVI 48); 1473 (V 30); 1493 (III 76); 1542 (V 27); 1656 (III 69); 1658 (III 72); 1662 (III [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN 2 3203 (III Mi am (IV 10); 2 (III 76); 4801 (III 74). E —. 80 (III 100). Setchell, W. A. & C. C. Dobie. s.n. (XIX 4 (X 1); 29? (XVI 1); a 493, 941 (II 9); 1236 (XX 6); 1281 (X 1); 1447 (V 46); 1500 (II 9); 1624 (XVI 1); 2446 (II 9) ; 2590 (XX 6) ; 2737 (XVI 1); 2748 (XVI 1); 2872 (V 46); 2928 (II 10) ; 2951 (XX 6); 2955 (II 10); 3106 (X 1); 7970 (XX 6); 8268 (X 1); 10870 (XX 6); 10383 (II 9); 10428 (XV 2); 11861 (II 9); 11966 (X 1); 12167, 13419 (XX 6 Shannon, R. C. & E. S. Shannon. 28 1). Shafer, J. Ge 302 (X Shannon, W. C. 218, 251 SN 82). Shattuck, O. 129 (XXVI 37). Shaw, —. s.n. (III 72). Sieber, F. W. 92 (III 81); 333 (III Sintenis, P. 273 (V 45); 273C (XVI 1); 1622 (V 45); 1891 (XVI 1); 2457 (XXVI 1); 2866, 5195, 5306, Be R. 21623 (V 12); 21633 XXVI 46); 21771 (VIII = Small, J. K. 3995, 8786 (XVI 1). Small, J. K. & J 5a: Carter. ys 4 (XVI 1); 2655, 2657 (XX 6); 3116, 8605 (XVI 1); 8751 (X 3). Small, J. K., J. J. Carter & G. K. Small. 3545 (XX 6). Small, J. K., C. A. Mosier € G. K. Small. 6582 (X x 6). Small, J. K. & G. V. Nash. s.n. (XX 6). Small, J. K. & P. Wilson. 7756 (XVI De Smith, A. L. 672 (III 12). Smith, C. L. 71 (III 96). Smith, H. H. 165 (XXVI 47); 884 (V 41); 1640 (II 2); 1641, 1642 (II 1); 1643 (XXVIII 2); 1644 (XXVI 14); 1645 (XXVI 8); 1647 (XXVI 26); 1656 (XXVI 53); 1662 (III 81); (576) 1936] WOODSON—STUDIES IN 1904 (III 36); 2412 (XIII 3); 2413 (III 81); 2525 (XXVIII 2); 6663 (III 95). Smith, J. D. 1746 (III 96); 4884, 6657 (III 96) Smith, L. B. 1529 (III 67) ; 2037 (XXI 2) Smith, L. C. 257, 468 (III 33) ; 660 (III 81); 661 (XXVI 55); 691 (III 81). imd E. H. 528 (II 1). Sodir (XXVI 18); 10613, Hus ES E 106/1 (XXVI 27); 107/16 (XXIV 2). Splitgerber, —. s.n. (III 96). . (III 29, 81, 86, 108; IX 1672 2863 (III 108); 3051 (III 82); 3136 (XI 2); 3152 (IX 23); 3305 (VIII 12); 38395 (III 105); 3430 (XXVI 8); 3481 (V 10); 3550 (IX 5); 3610 (III 104); 3718 (XI 1); 4303 (III 101); 4493 (V 5); 4900 (IX 18); 4908 (V 31); 5390 (III 26); 6019, 6484 (XXVI 18); 6485 (II 1 Stahel, —. & —. Gonggrijp. 2862 (IX 23); 6624 (V 12). Stahl, A. 743 (V 45). Standley, P. C. 22349 (XXVI 44); 28542 (XXII 1); 23673 (XXVI 58); 24512 (III 81); 25027 (III 96); 27717, pu (II 1); 28636 (VIII mat 29251 (II 1); 30184 (VIII 20); e 1); 30385 (III 96); 30562, 5160 (IT 1); 41838 (XXVI 57); 556191 CLE 1); vid (VIII 20); nod 56470 (III 8 Steere, W. C. 1471 = 4); 1477 (XX 5); 1918 (XVI 2); 1987 (XX 6). (577 THE APOCYNACEAE. IV 405 Steinbach, J. 2863 (XXVI 28); 3113 (VI 1); 3256, 3331 (XXVI 8); 3772, 6065 (XXVI 59) ; 6717 (V 20); 6813 (XXIV 1); 7272 (XXVI 24); 7295 (V 23); 7301 (VI 1); 7376 (XXVI 28); 7456 (II 5); 7577 (III 96); 7782 (V 23) ; 8020 (V 20) ; 8662 (III 40) ; 9090 (III 23). te Jue —. 8.n. (III 76, 96). Stevens, E. P. s.n. (III 96). run F. L. 538 (III 81); 1006 (III 82). Stewart, J. T. s.n. (VII 1). Stork, —. 270 he 96). Stuckert, T. IV 8). Stuebel, A. E (TH ak 74 (III 85). Sutliffe, E. C. s.n. (XIX 1). Swartz, O. s.n. (III * V 43; XV 1; XXVI 1). Tafalla, —. s.n. (I1). Talbot, H. F. s.n. (III 78). Tamberlik, —. an. (IL 5; III 87). Tate, G. H. H. 188 (III 103) ; 223 (III pit 235 Gong EE 669 (III 60) ; 805 I1 2); 1). Taylor A. ^n Ba SE 9). Taylor, A. H. s.n. E 88). Tejada, R. 249 (X 2). Tessmann, G. 30 vi E 24) ; nes (XXVI 8); 3587 (III 79); (XXVI 1); 3639 (V 10); B pé 33); 4009 (IX 1); 4034 (IX 18); 4110 (IX 1); 4356 (IX 1); 4359 (VI 2); 4413 (V 5); 4461 (V 10); 4805 V 10); 4816 (XXVI 8); 4918 (II 4); 5064 (III 79) ; 5106 (XXVI 45); 5128 (V 10). Tharp, B. C. 3646, 4648 (IV 5). Thieme, C. 517, 622, 5346, 5347 (III 81). Thompson, W. J. 1005 (XVI 1); 7975 III 9). aee 764 (IV 2). n, E. F. en (II 1; de 96, 106). Tuo, T. F. de. 28550 (XXVI 47). Tonduz, A. s.n. (XIII P. ipe (III 81); 2147 (XX 2); ?441 (XXVI ) E 406 57); 8322 (II 1); 8904 (XXVI 57); 9889 (XXVI 44) ; 12711 (XXVI 57); 13355 (V 32); 13686 (II 1); 13945 (XX Toro, R. A. 28 (XX 6); 59 (XXVI 15); 163 (III 81); 391 (III 91) ; 650 (II 1); 928 (III 91) ; 1105 (XXVI 8). Tracey, Mrs. J. A. 54 (III 85); 297 26). Tracy, S. M. sn, 5468 (VII 1); 9185 (XX 6). Trail, J, W. H. 3n. (IIT 86); 517 (VIII 18) ; 519 (II 1); 522 (IIT 86). Trejo, —. 1084 Trelease, W. s.n. (VII 15 351 (IV 3); 353 (IV 2). Tresling, —. 407 (III 48). Triana, J. J. s.n. (III 37, 81, 85, 91, 92; XXVI 54); 1909 (III 26); 1920 (V 41); 1979 (XXVI 54); 3409 (III 93 jp Troll, C. 2687 E ye Tuerckheim, H. 1233 (VIII 20); 1275 (III E p (III 1); 1829 (III 6); 2448 (III 1); 2586 (11 7); 2688 (XVI 1); 2689 (XX 6); 3341 (IX 25); 3476 (V 46); 3605 (II 7); 3636 (V 46); 3924 (III 81); 8244 (III 96); 8369 (III 81); 8540 (II 1)5.8709-CIIL D. Tulleken, —. 90 (V 9). Tw d Eat tee CULTS 4 HE GER V 27); 791 Ule, E. s.n. (III 96) ; 77b (V 18); 350, js 1051 (V 8); 1161 34 3 (V 27); 3829 (III 58); 3835 aa 43); 4285 (XXVI 35); 4288, 4434 (V 28); 4578 (V 27); 4580 (III 96) ; 4845 (V 27) ; 5175 (XI 2; 6604 (XXVI 30); 6650 Cx 20 (III 86); 6988b (III 43) ; 7020 (III 71); (VI 4); 7121 (III 49); 7405 ges 30); 7681 (III 86); 7823 (III 96); [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN 7825 (III 86); 7826 (II 1); 7827 (VIII 9); 7939 (XXVI 47); 7940 (VI 1); 8266 (XXVI 47); 8449 (III 93); 8451 (XXVI 10); 8490 (III 89); 8737 (III 103); 8952 (VIII 11); 9698 (IX 26); 14747 (III 31). Valeur, E. J. 96 (II 7); 175 (XVI 1); 226 (V 46); 444 (XX 6); 978 (XV 2). Vargas, J. M. s.n. (II 1; XXVI 9); 54 (XXVI 14); 109 (III 81). 1) III 43). v 8. 46 (V 30); 178 (III 24); 1769, 1769a (III 42) ; 2738, 4076 (III 24); 5518 (V 23); 7757 (III 55); 9773 (IV 8); 10181 (III 24). Versteeg, G. M. 72 (XXVI 46); 382 (III 48); 415 (V 9); 451 (VI 1); : Wied-Neuwied). s.n. (III 43; V 27, 34); 3131 (III 49). Wackenheim, —. 119 (VIII 10); 254 (III 78). Waeket, M. s.n. eebe XXVI 2). Wagner, M. s.n. Waly, J. F. 8362 E 88). Ward, L. F. s.n. (VII 1). Warming, E. s.n. a y Ars 25, 46, 72, 715 LY: Os; VI 4; X 4; XXI 4 as We 48); 239 (IX 21); 252 (II 1). Warszewicz, J. von. s.n. (XXVI 18, 55). Watson, S. 379b (III 96). 68 (III 78); von Wawra, H. € K. Maly. 241 (XIIT 4); 279 (XXVI 2) ebber, —. 434 (XX 6). Weberbauer, A. 536 (III 23); 1888 (V 5); 36058 (XXVI 8); 4265 (XXVI Se vp (III 38); 4488 (III 79); (I 1); 4740 (III 101); 6231 E d 18); 6243 (II 4); 6270 (III 81); 6498 (III 23); 7128 (III 15); 7131 (III 17). Weddell, H. A. s.n. (III 25; VIII 7). Weir, J. 462 (III 66). (578) 1936] WOODSON—STUDIES IN Went, F. A. F. C. 109 (V 12)). Werkle, C. 69 (I me 71 (III 96). Wettstein, R. von & V. Shiffner. s.n. d 70 Wheeler, H. E. s.n. (VII 1). leri A E. 1142 (III 80). Widgren, J. F. s.n. (V 3; XIII 4; XXVI 2); 51, 52 (V 35); 61 (III 96); 63 (III 74); 370 (III 72); 371 (III 74) ; 372 (III 76) ; 578 (IV 10); 581 P 25); 582 (IV 10); 1325 (XXVI 2). Wilcox, T. E. s.n. (IV 2). Wilder, E. M. py XIX 1). Williams, L. 928, 973 (VIII 13); 1133 (I 1); 2089 (VIII 13) ; 3891 (V 31); 4011 (V 5); 4024 (IX 1); 4310 (XXVI 24); 4385 (IX 14); 4547 (V po R. O. 10966 . iams, uw O. & H. C. Sampson. 11735 (III 81). m R. S. 81 (III 23); 346 (II 4); 362 (XXVI 14); 536 (III 79); 547 (III 96) ; 571 (XXVI 28); 804, 875 (III 79); 940 (XXVI 39); 2427 (III 40) THE APOCYNACEAE. IV 407 Wilkinson, E. H. s.n. (IV 5). Wilson, N. s.n. (V 14). Wilson, P. 214 (V 45); 422 (VIII 20); 7273, 7458, 7595, 8148, 8240 (XX 6). Wislizenus, F. 341 (IV 4). Woodson, R. E. Jr. s.n. (VII 1). Wooton, E. O. s.n. (IV 4 Woodworth, R. H. & P. A. Vestal. 498 (XXVI 8 S Worthen, G. C. s.n. (III 81). Wright, —. s.n. (VIII 20). Wright, C. 398 (V 46); 399 (X 1); 557 (IV 5); 1876 (XV 1); 1377 (XXIII 1); 1662 (II 9); 1664 (V 41); 1665 (IV Ea 2954 (XVII 1). Wright, C., C. C. Parry & H. Brummel. 09 (II 7 409 (II 7). Wullschligel, H. R. s.n. (III 96); 319 (III 86); 320 (III 96); 321 (IX 10); 322 (IX 21); 324 (XVII 1); 326 (XVII 2); 454 (V 9); 1028 (XXVI 20); 1029 (XXVI 46); 1506 (III 87); 1971 (IX 19); 1972 (III W ydler, H. an (XVII 1); 209 (V 45). Yuneker, T. G. 4956 (XXVI 44). INDEX ro Taxonomic ENTITIES Roman type indicates accepted, preéxisting entities; italics indicates synonyms; d bold face indicates new combinations and newly escribed entities. Numerals refer to the parenthetical pagination, when in bold face to the pion eng tions and citation of specimens. Pages (1)-(186) are containe . Mo. Bot. Gard. 20: 605-790. 1933; (187)-(340) in 92: 153-306. 1935; | qi (588) 1936. in 23: 169—391. Allamanda Page CALNATULOR Ben cic thas e iis dí 419 PA ds ee eee Et 18; > NEO AO plumerlaeflora .............. s. eg SE a ER 6; 41 aoutuoba- EE EN 45 EE EE 61; 399 Page ENEE 84; 399 EE 36 EE EK 137 A EEN 170 Prin tet dun. EE 73 Droo On EES 66; 402 "IEN 39 SE 408 ANNALS OF THE MISSOURI BOTANICAL GARDEN Page Page EE 159 Anechites CHIT OU EE 34 adglutinata ............ nnn 451 Claussenat EE 160 COBOS: uae ide eatin wien lhe LEE 72 lappulacea ......... eese. 400; 411 convoluuwlacea EE 83 Thomasiana e I ea RIA IRA Bi OL EE 347; 541; 550 Angadenia ................... 6; Lë EE 148 10; 14; 16; 18; 41; 304; 363; 367 "Nendleri o c eere 154 ye CRUS PERO Lis ati MEUM E EE 159 Amazonica EE E 368 Be Claussen v ede a Vs 159 «Berteric v E a. SE E 159 382 2 401; 405; 407; 419; 548 IAN EE 54; 266; 272 EE EE 307; 368 Tama omms EE 93 ee 337; 368 ME 93 CUDASS] a Aa B. brevipedunculata ......... 93 8 Te cokes pes EE 320; 368 y. microphylla ........o.ooo... 93 LEE 337; 368 a. DEGUNCULATIS a 93 A Geer EN 36 Guia EE 136 Mala E. 337; 369 LTL WEE 159 grandifolia A 320; 369 hispida EE 159 Haeawensis Ne NN ENEE ER 6 B. tomentosa a EE 159 hypoglauca .........0000. 335; 369 WINSKU .......o.... 53; 405; 410 latifoha «osuere 313; 369 leptophuylla e ee een 550 Tindenana 24Gb cess sees RELEY A te ee ieee 56; 420 266; 364; 367; 383; 411; 548 ie 140 E AER E 335; 369 e glabra Siete ie sists EE 140 "HA eae eo EE 334; 369 a. d NEEN 140 pandurata ......... esee. 369; 433 HANEN E 140 IS A g a pees Sars Marae Dr IRAE 140 PVCU EE 136; 370 B. trichantha EE 140 DUO E 370 Ee EE 164 LEE 148; 370 LEE 82 E A eraa eran 6; DEE noha Anos 60 EE 313; 370 palustris EEN 159; 430 sylvestris EEN 331; 370 B. almadensis ........... see. 159 Valenzuelana ............. 342; 370 VEIT wet T 433 -ANSOLODUS a aU» 5; PAVONE 145 AMAZOMICUS ...ooom.moooo.o.o.. 313; 314 EE 136 latifolius ............ 313; 369 SORTEO odor es onara aie 155 ELE EI 172 COON AGUS EE ioe 307 EE 91 i MC EET 320 torosa e ane aa s GE 52 B. grandifolius .............- 320 tubsftOora me 44 distinctus... e eese rer 556 ge 148 A EAEE A ET. 320 B. intermedia oee 151 ELE 328 Anartia hebecarpus EEN 317; 370 BOQOUENSIS A EE 73 glabratus e eae [Vor. 23 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV Page Gs COMENLOCUB. E e alias ere 317 Ge LCT EE 317 En 317 OH 320 SCD ARN ce EE 319 lanoif olus Fis oo odeurs 340 EE RES Ore ate C MORES TE. 556 I'Bertoitete Ba, aaa cse. Hos 322 B. ODIUBUE o... oris a e ree 322 PORUGTUS eris rr un ae 12 nn EE 317 PUNOISCUIOSUS: ¿esoo 320 A E RT 55 LEA | Ae ere 5 IS PRUCOL: eerie siete 313; 314; 370 StadeUneyeri- ies cece EE 5 EE 317 Be EE EE 258 2A COUO E 203; 204 cannabinumv. E 25 var. glaberrimum ....... 230; 257 LUT. Dee 41 paniculatum ` ee een 203; 204 EE EE 242; 448 APONEA EE E 6E IST COT UO" o THERE eere 245 Aselepias rt . CX REA EE 274 Asketanthera 6; 10; 11; 17; 18; 435 ealyeosa . 403; 419; 436; 439; 541 Mugs. capp A OU UU ENSE 406; 438 Ekmaniana EE 412; 439 ae: E AET Eo 439 Selen 416; 437 Beaumontia. . fos. 12 e wa es 562 elandra Ee 6; 449 el 503 DEE 6; 341 bahda ens Fes cite s 344 Cameraria GUYONENSIS Foss cae: ss 286; 303 Rt EE 286 Tamaquariaqs EE 286 Catalpa macroearpa Fars coe cane ates ere 304; 412 Cissampelos ovalfals eee dec eis cuss 406 409 Page Charionma tori on Ao aes FEE es 05 Sí P M ira E ale clas 372 AO EID dU CUM ERE 372 ONE T AA UR Bice ror see MN 372 NOOSE ie TOUT E 125 SCURGCNS e Cee E 372 SUPT COL. EE ewes eu be 372 WETUECULOLO a nen Hare E 372 CODGROCTR EE A Rs 6; 304 paniculata «1. A v da 339 COMBYIOCHEPON E EE na de 410 CLA ODIS: Ee 5; 7; 8; 10; 11; 17; 18; 304; 386 Tee EE SE 331 rte ee cracls ease 386; 387 var. tomentosa ........... Ki 388 EE 2i. MERE COPIE A 387 VAL-VOhUdiB. 14.5.22 di e eei 388 Sage E, le 388 EE, e 388 Cylicadenia DOMO E. osse veio 5 "RA. a nde AP E 331 EE oi tts .5; 7; 41; 95; 96; 197; 352; 551 Aoi eg e ; 95 A A 78 COUNTED, a ia Mas ter ee see 99 DOTTIE 1. cee ss tite sek oe ok 86 A A E A ec 125 androsaemifolia .............. 125 OHNE OD EE, eines 106 WODUPDRFER ~; sisteie ER ere 120 RA 120 A Sa EEN 120 BOULE SE dee 120 AN A AO evt 120 A eet 120 E A eris 120 bela A A eas arn ftre 113 DONOSO 2 ds as ees 112 O A dE AN 126 EE 349 COGOUNGEE Ie tae octet 549 oono e E Ee 72; 95 A e a CURING ana 95; 126 CLA Fyn ican Re ala aaa ats 102 410 Page BEE 72; 99 A E EE 372 RT ROA RS RSS OO AA 109 a. oppositifolia ...oomoo oo... 110 B: fernatifolia PP 110 Gard HOT V LAU A rire EE 124 B: grandiflora ......... ess. 124 genMttanoides CRT Te sie p. glabra ......ooomomo.. 129; 548 p*. longiloba ............. kale TEE EE 129 f$ ovala EE 129 Spe Mere 129 Be COUUTUR EE 128 at. longiloba EE 12 glabra RA AAD ER 105; 548 Harris o els tele ela tee TEE 331 TT Een dip COLO SO Dee B. glabra e cas 125; 399 forma glabra Ao. ee ees forma pilosa. oe ee: oie 124 LIME EM ENS 124 Di cup tia senden: 124 TORWAR 124 $ pubescens EE 124 e. oblongifolia ............ 124 a. rotundifolia ........... 124 TAS hr SLOT cote 124 DUbOSCERS EE 124 Lean f olas cc OS ROO ODO E 75 4ncaris s So TIENS rks cs ese 129 longiloba- eg 549 Lust ett mS ME 95; 98 B- glabra A COE eie EE E 98 EE Moricandiana: EE 101 myriophylum E 7 Ee 125 oblongifolia s o scene ea 104 LOL MENORES e 75 varsteminfoha «ses pais 76 pedwunoularis Loi ic ee es 75 ET EE 121 DULL EE 126 [VoL. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN Page POMNO eres «ip gis err d 129 OUER 75 B- DNEV OVUM E 76 p antermedia rs O 76 y. peduncularis ............. 76 ax Lena OO Fenelon ke cite 75 i Vf UR EBB ARA SHAH DATS 76 Jupuberula rore 76 ESA RAGA A RAGA AHS OF 76 PUC a 132 EE 110 Rosa-campesiris .............. 124 OTA 126 ul dE 122 SONA EE 116 BONONQING EE 131 SEL c easton ce 130 YA) SRR EE 113 elt 115 8Digeltaeflora mot a 132 ALON TU OUR E LER 132 aptendena a EE 103 ENEE 72 mensis qo EU. 100 engt 75 forma pastorum ............ 76 FO pubera es SEM 75 S TE 75 DEELER 126 TEE 100 wropnhalla EE 118 EE 128 VENULOBO ¿o bie aioe Glee 119 1 LON. EE 75 canthostoma as. wie we lon ore 130 forma longiloba ............ 132 OQ RR OATS 130 ISPs ere 30 b. mi Echites . T es 3305575 E 10; 15; 18; 25; 42; s 304; 389; 435; 449 Te EIS dv queror ont: 31; 140; 398 EE adglutinata SEENEN 398; 450 e on cbse E 98; 410; 418; 420; 430; 450 No A e S OO EIS 398 (vet tn TQ OE 125; 399 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 411 Page Page madensis .......... 159; 368; 399 Gütesbüek. or oc ose ee css 372; 403 altesoandens eege ee e 137; 399 Chilensis ...... 384; 403; 409; 417 AMAZONICA ....... ee 313; 368; 399 ohlorantha. ceceno sve ens. ; CONGUE A O OO EE 61; Christophoriana .... ee NA 403 Andrews A ori ses 372; 399 ONE corre eae herr re 159; 403 ENEE Sos Seve wore es 84; 399 OVLOTEO 5p oe RIS TD Ea 403; 541 augusti olia, e ocio ano oie Ee, rk ew es 392 TESSA 36; 170; 399; 402; pee circinalis .................404; 450 annulanis EA 400 ; 481 var. Thomasiana ........c... antennacea ..... 137; 399; pe 402 ISTOU M 33; 402; 404 sedi lige E NOU NE 2400 X OVTMIO A S EAn E TS NR Ludos CENE 283; 284; 400 coalita ..... 402; 404; 413; 423; 452 RE E E OLEI ES A DEER Sane Os EE A E 400 coccinea ....130; 404; 424; 432; 541 E e EE 400 var. g. oe E 131 A EE 61; 400 SENE Loco UM el 307; 368; 404 at Poe ng adno od uot 120; 400 COmO8SG .......e eee eo ro 04 Gir OUtOLacea s ala aaa 120; 400 CORDON, Socr see. f- SC SE tae DEP 185; 400 00876310... ene 71; 404; 417 aGarteilala E 159; 400 convolouacea ..........o... ; EEN ee E E 400 o Ee EE EE 82; 405; 430 Bang. ere ist alas 1; 461 OOGO EE 336; 368; 405 VOLUME class coun ens 372; 382; 401 corymbosa NEE e 4; 252; 405 SE Mors EE 170; 399; 401 CONV cca EE e ; 4 e A 65; 36 Fe A 347 ae b: 401; 405; 407; m 548 NNNM iuo duy REN $ a EE EE crassipes .....10; 15; 398; 405; 541 bicorniculata .............. p^ 401 ea ge NOE E DER T OT TER Eee ee ; 405 Ee, EE 313; 401 A T ETEEIR ETT 319; 368; 405 biflora ....... 4; 378; 401; 407; 416 var. a. angustifolia .......... 319 DignoTuaeflora DE nS oce 401 var. g. grandifolia ...... 319; 369 BUDJA o ee 378; 401 OUSDIIMEME aa EEN Ce 26 DBIGRENCTIA. Lu Ce de 401; 452 CONO ir dr 140; 405 LE Tao ai? 73; 95; 402 CAVAS Acs sa 148; 405 DOEN, PRIEST LETS 137; 402 eyaniphylla .............. 406; 456 brachyloba- aia e 66; 402 densevenulosa ............ 406; 556 BrachysiphoR ............ a 402 E cL EIER CROIRE OR DOO ; 406 brachystachya ............ 14 2 denticulata ..... 406; 421; 494; 500 bracteata ........ 69; 157; 236; 402 TT IAN E 0; 406 grat Wee det 38; 402 GIN ek Gate ns bos Sars 406; 480 E DE 36; 37; 402 UG OCDE E sh lewis seit breves m e JS sn. 267; 268; 406; 417; 419 A n aA ; 402 dünde BEE 6 calycosa ........ 403; 419; 436; 541 A ee 0 OGMPONULGE E dolichopetala ............ 406; 438 Ek 158; 163; Ke 403 domingensis ..............972; 406 E Ee ; 549 ET 6 412 age Dusem 9m dre Hue: 407; 453 EE 396 Eggersit Ee 407; 543 Ehrenberg o onic ees EE 378; 407 SCENE 368; 407 elegantula a A 394 emarginata A. eee eens 68; 407 CIO: Ss ia sansa eine = 68; 407; 540 EECH 07 EE 266; 365; 407 emmer ere 248; 407 FUMANENSS Coes sence 159; 407 B. Claussenti ......ooooooo... 159 adiit EE 109; 110; 407 ONCISCED .......... 407; 426; 429 var. pallidifora KN 427 En DEE EE 408 funiformts ............ 93; 408; 414 galegiformis ......... eee 08 Tabarica E 93 GOTANCTIONG aes eis ene eee 549 Be grandiflora Ee 549 geminata ....oo.ooo momo...» EE 368; 369; 405; 407; 408 glandulosa ............ 88; 408; 539 glaucescens ........... 82; 408; 430 glomerata e E 408 gracilipes ............ 328; 369; 408 EE 139; 408 grandiflora .......... 185; 330; 408 KREE Oe eee ees 183 var. wes Papin sates 185; E 548 Guarantee oae ama 185; 09 Vade in SPORE meyer TEE eis S eta 139 Guianensis ........... eee. 136; 409 ebecarpa c ire ee ere: 317 heterophylla ......... 372; 384; 409 KO 145; 158; 368; 382; 399; 400; 403; 407; 409; 412; 418; 422; 430; 538 B AC MEO ne serene 146; 59 hirtella meee ee 82; 139; 409 HATO ala 140; 409 EE 159; 409 Hoffmannseggiana ............ ee 330; 370; 408; 409; 412; 422 (584) [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN Page IO Oa GO a e 409 SE ae AO a ga ates 409 ALLL ON Garrone O totais oie 0; 461 hypoglauca .......... 335; 369; 410 hypoleuca ..177; 410; 421; 541; 542 TE EE 123; 410; 419; 424 ARTS 410 A AA a e 410 Jamawensis EE 372; 410 NOTE 26; 410 JAEMUMNOTA os. cise snes cise 139; 410 EE 171; 410 TE ATIDINSK EE 53; 410 «Lambertiana cc 183 lanta o E ToU UTE 410 lanuginosa ........... 178; 411; 541 LEE 400; 411 lasiocarpa ........... 146; 411; 430 4 angustifolia ......... eu. 146 DD 146; 430 its O s 411 Laurentiae-disca .......... 411; 461 ————— T 91; 92; 411 pd SE 411; 430; 451 Ee Ee 152; 411 AENA E 364; 411 linearifolia EE 36; 152; 411 COTO eect iets eters 41 56 Ms EE 396 EE 185; 396; 400; 408; 409; 412; 439 LONDON eee ein 12 LOTO EE 334; 412 lutea ....... 317; 370; 406; 412; 424 macrantha .......... eere 3 12 EE 357; 363; 412 MACTOCATDA e 30 12 macrophylla ....... 74; 95; 158; 412 Macrosiphon ............. 179; 412 MACTOSÍOMA 1.1... ee ee eens 380; 413 EE 413 Lu E 383; 413 QUIS OQ TU,» SI Te ieee a © 32; 33; 413 MADUENSIS Lino 13 Maranhamensis ........... 148; 413 Martana EE 98; 413 1936] age EE 413; 452 Maximilianea ......ooo.... 413; 429 Mög proe is 413; 501 membranacea ....... eee. 140; 413 MERCA Fast EE 56; 414 OT WAND AA A Schone c 331 MAONO ONYT Ee eer 140; 414 God are EE AA 140 ANACT ODNYUG= e clon E 93; 414 MAMME E ei 39; 414 MOULESSUNVE E 147; 414 MORTAN. o eee E 60; 414 RE 139; 414 EE E EE 184 TTACOLO EE E 414 myrtifolia seee cise lene ns 38; 414 RETANO EE 72; 414 AAA neo es 334; 369; 412; 414 MOCO RO EE 414 MUU OUTS O ECE 59 Oazacana ....... 82; 408; n4 430 ODUQUA ere cala 396 LEE 15 obovata... s 101; 373; 415; 548 obt OUND: v. Sacre ER, 15 dadi sere cco ete E 415 1a e UT: 415; 428 Eet US TRE rc: 415; 555 OPA a DUE dios a 22; 415 O A o USE DIT, sions, od ee 396 palida ens Ee eee: 27; 415 Dal^t10801-. EE 378; 416 Le DEE S 159 pandurata ...... 369; 377; 416; 433 maniculata. EE 16 Parvi NNA EE EE 416 DASTOFUM EE 75; 416 Pavoni sie: 145; 409; a6; 538 pedunoularis- ee cos ce eee 16 peltata . pt 363; 412; a. 417 etHorto v e AT ruens 372 peltigera oes ces s 362; 416; 423 EE 83; 416 Picarda 11.5 Gee E 416; 437 Hosa e o EE 193; 416 DUR UAL OWE. er EE a 416; 433 (585) 413 Page DWIATOA EE 183; 417 PUC E A e 357; 363; 417 Pohhana EECH ; 417 var. a. angustifolia 128; 417; 548 yar. 8. latifolia AAA 138 portobellensis ............ 12 TACUIT so se pcia ec 136; 2 417 DECANE OU KE 417 Diat mica e 384; 417 UO CT UL ar torre EM e 268; 417 pubescena suno E 72; 384; 417 DUO SLOT EE EE ; 418 Vance TEE EE. 132; 418 TAuloherrwma EE 317 pu niis 319; 368; 369; 405; 418 PURA Ee es 64; 406; 418 pun ibis adc DEI M TR 4; 415; 418; 431; 459 UC HOE On EMT 4; 35; 418 TODO CEU DEEST kein het 418; 451 EE, 158; 418 E E 418; 535 TIONES Tr ie ao Eco ao (a 418 TUPOTU oss 61; 62; 399; 400; 418 Rosd-campestris .......... 124; 419 TOSUH LE EE oes eoe ER 48; 419 Mare REPE SEN. E 38; 419 TUDISOURUE 05 EE rS ien 419 RBügetidha 2... enl. 419; 436 THUEUORENE CS cans TT 419 Lui n cd TE 150; 419 SGOLLLGNE aces eo E SE 419 EE 365; 383; 419 Salto OU, arc graio ads 19 EE 122; 419 Sanctae-Cructs ..ococoscin... 30; 419 Sanctae-Martae ........... 420; 544 sanguinolenta ...ooooo.... ; 451 scabra 48; 370; 401; 402; 403; 406; 413; 418; 420; 422; 424 Ne ee Zeng 420 silio E E 420; 431 semidigydüg cineres d ei 420; 431 808A VAR M1 e nien 303; 420 Sm T AP E E 56; 420 SPOOR TEE Lcd PE 139; 420 414 age spectabilis .......... 385; 420; 441 bot, E cT eI 4; 245; 420 spigeliaeflora ........ 132; En 420 ONG EE E MERE RC 103; 421 Stadelmeyeri ............. 159; 421 Dr TEE 421; 423; 426 stellulifera EE 421 stipellaria cerdo 329 suaveolens ....... 91; 177; 421; 541 subcarnosa- E 21 aubcordata EE 21 aubereota ees 4; 372; 421 et 21; hs Fea ata ....139; 405; 408; 09; ; 413; 414; 420; pe 431 Ers EE ; 421 subspicata ..135; 370; 409; Ee 422 ESUIPRUTEO aa 32; 422; 452 ri LEE 26; 422 SYLVESTER marc see ee 331; 370; 422 symphitocarpa ............ 151; 422 e eg 422 enui E EE 148; 422 tenuiflora ............ 75; 416; 422 thyrsoidea EE 234; 422 tomentosa ........... 158; 422; e [AA AA LE ee 4; 52; 403; i. War, Brownet oa e 52; T05ulosac cee reeset. 38; 52; 422 Lal, EE $2265 27; 403; SC 410; 415; 418; 423 forma puberula ............. var. AU ico Der tropaeolifolia ............. 362; 423 tubiflora .......... 44; 45; 404; 423 tubulosa ii ero 23 binata anat Y deta 91 Luet EE Dies 92 tuxtlensis EE 390; 412 eediana ar. arn eta ote umbellata ASIE 4; 5; 15; 389; 390; 395; lo. a var. crassipes ...... 541 Var. tONGthOnd EE 9 VAr. typica o enea 396 [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN Page NOOT e D SE LN DRE ara 423 CTS YU Re EES 423 Valeneuelana |... eee SURE erts 342; 370; 383; 423; 550 e e WN ADO DOD COO aS 423; 426 VAT. DU DAJGO. o atar loe 27 VAT NOSE KEE 427 var, Sulphurea .......ooomo.».. 27 KU nT ae EE 415; 423 Kauthieri hs rene ea a 423; 452 eene 184; 423 Velloziama ........... 236; 402; 423 OMA EE 128; 424 VONEN OGG Kee 124; 424 Ee SE 86; 424 EE Ee E 369; 370; 399; 401; 424 See Er 48; 424 UarteoWlatin: 1 e er eheu s 424 violacea ........ 408; 413; 424; 429 A O A 180; 181; 424 ucataneonsis EE 393 zanthostoma ......... 130; 424; 432 Zuecariniana ........ eee 31 424 Elytropus ...... 6:515711: 0717511857 888 cOn 384; 403; 409; 417 as aie EH EEA Ie OA 84 EE 384 BOONE ta vie tole 84 EE 385; 441 e EE (0B vec ul GE EE 167 Exothostemon ........ 5; 41; 133; 449 bracteata aes cts assess se Iss 157 OO 529 Qraoue cor ERIT ESTE 139 E rc E eee ae 171 macrophyllum .........o.o.o... 158 mollissimum NNN NEE 147 TEE EE: 378 EE 529 E EE 139 Een ; 6; Luke Ee ees i braehy harynx NS ters pandurata E ee 173; i 377; 416; 417; 432; 433 1936] WOODSON—-STUDIES IN THE APOCYNACEAE. IV Page A erect eve Duis tes BRE 10; 14; 19; 187; 284; 412; 420; 555 BE AX Cer P DE CRT 203 DOUCET OMG. TEE 195; 196 B. pubescens ............ 195; 196 Sdenobasiss Eeer 241; 245 iral GENE OOO m 200 Alezündrii e Ue eS. 248 el TEE 214; 217 JUSTAS EE 226 thamisna Seaan AA 204 OTGCCCONG is. ce re tee cele eae 236 KONDO GE, 224 p bahiensis ¿iia «sos 224 Drevifols 0 NONI eR 207 chiriguenBis docu 5-25. ere 237 E CORO EROR ORE 220; 221 E SC 5 corymbosa ....... 16; 252; 405; 555 decipiens EE 5; 217 WAS ae SC diospytifolia EE 209 CHE 555 GKel aero ade state 211 Re EE ot erent 196; 199 embede. GE EE, 193 CT A O O O T floribunda 14; 16; 248; 251; 407 OT NU UNAG 3. 5 NOSE 254 Ee 222 BADINA 20500 0 ac ee 232; 233 Gardner! EE oe pote A 219; 220; 228; 235 VE SE 197 AKTE 206 E A O UE, vos cists 254 Jeer EE ; 448 leptoearpa ...... 224; 226; 997; 555 var. glaberrima ............. 224 E, E 556 Lusehnathi ¿ccoo aos 195; 196 Wa. stn 242; 243 meridionalis ....oooommo.o corso. 193 A A dose 236 EREM 193 415 Page MOL ado a CR 221 VEEL? LOMOBW EEN 222 VAL REY EE Ee 222 MONAND E 217 MVULTURET UL. sse EE 234 B. glabriuscula .............. 235 ax MACTOPRYUG A sans 236 8200087080. EE 234 a A Ad 231; 233 OOE a ree ev ae AN Poio 235 ODtURIIODE 229055 9 uos sees 227 OUGU LOU EE 555 E 213 A E SN 257 Dr MEER E 230; 257; 258 DOTADAS 71.2009. e vio 249 A one AE 193; 416 Pg gO) TA 251 EE IIR v 96 pubescens ....... 218; 220; SS 235 p. eordata ......cooooooo.... 220 PEGUIOULSISUS 712.5400 9 ec? 239 noschir sete oce 195; 254; 256 TEE SM OON A cen 255 DOUG recae e 210; 211 rotundiuscula ............. se 555 TINI EE TOT ees. 201 Li 201 Sandwithiana ............ 243; 245 Schomburgki ............ 203; 242 B. umbellata ............. 242 EE 217 SUN ET 188; 247 AUNTS Dt ALAR A ....188; 245; 247; 415; 420; 556 eelere SE, eg 223 (GT aaa > 229; 258 pa O O O 233; 422 var. glabriuscula ....... 234; 555 var Ee e 234; 239 umbellata ........... 242; 245; 448 A O = ; 402 Kaell O A 205 O OOO OL 208 NET E O 19 416 Page Galaetophora ..7; 8; 17; 18; 346; 352 LN TEE 34 crassifolia ...... 346; 347; 541; 550 AMIA o 350 EE EE 347 Schomburgkiana .............. 348 Haemadictyon ................ ; 44 acutifolium ..... 401; 410; 411; 460 B. latifolium .. 460 Amagonicum `... 498 GE 481 (8 T6. a tate ee TE e RT 501 bracteotum `... ENEE EN 539 OO 461 calycimum ENNEN 499 Cayennense ..... eee 463 CWCWalis: oe eee 451 CONTOTTUM `... ee 529 denticulatum .......o.oo.ooo.o.oo. 500 CUM Foi sree ein pares eis ee cine 472 Gaudichaudii ............. Ln. 500 glabratum EE 478 glandulosum ............. es 88 grandifiorum `... ec ec cee reeves 331 OM OCNSE A eea n e aR 468 LANAMON Ue et aes oe ei 466 maoromneurwm ....... s 493 marginatum .......... co. 463; 464 megalagrion e NENNEN Ee 501 membranaceum NNN 480 M HAETTEN 529 TEE 478 MUL ONG EE 459 B. sanguinolenta ............ 451 OH 500 DAMA ade 478 KT EE 463 parviflorum ........... e. ; 928 Jede e 418; ae 534 ENUM E 512 solanifolium .............. 431; 454 suberectum ....... ecce 372 tomentellum .............. eeu. 478 LEE 494 EE 459 Warmingii ................ s. 468 [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN Page Haplophyton Kuele E a EE CIO o ee eean ES 403; 541 dE aech 5; 41; 6 EE mieurndto s nere 167 Homaladenia ..... eee 6; 41 ¡OA EE 76 linearifolia ...... eee 76 el 76 peduncularis `... 76 TER Fog oan re 76 EU 76 vinoaeflora. ee a 76 UE 19 Daisies EE 5; 41; 48; 71 E EE forma e. glabra ............. 70 forma y. Glabrescens ........ 70 forma a. Guilleminiana ...... 70 2. Griseo-fuseoa ........... 70 1. Griseo-olwacea ......... 70 forma g. Aaa E 70 BB. obovata E ee 70 aa. Kee ATEM 70 1. griseo-olivacea ....... 70 forma ¿. scabrinervis ........ 70 ONENA OLEO e NEE NEEN NI e 137 DC EE 550 EECH 67 emargwata a sae eee 69 RE 69 3. acutsfolta Eye sce 69 ¢ DID Rs om sa ar WE ES 69 y. Glabrescend ....ooooomooo.o. 69 a. Guilleminiana ............ 69 és griseo-fusea ee EEN NN 69 . griseo-olivacea ......... 69 B SE EENEG 69 Pps EE io ovy ase ss 69 dä. OVALA Bae cies ses aie alee 69 2. griseo-fusca ......... 69 1. griseo-olivacea ....... 69 €. SCODTINCTUNG Doce 69 JONOON 54 glabra e E ae nies 69 1936] Page Guillemimiana ....oomoooooom... 6 ful EE EE 67 E EE 373 MAL Be E 549 E DO s- cre ernie oie ole: eere ser 224 GO: Dee 67 OUUGUANETUNG EE 69 ALLAN MARA vere eae oo ers ote eee 67 EE EE 373 BUDSTACOUG! EE 136 DOUBT ULA: EE EE 136 OOSA 0 EE 143; 404 IAD OL GIA d. 22 99990 002 (ure ve Oy 11; 145 I3 e 293; 542 Boissierii AONO OUO 07; 542; 543 LACON LIS TENE Sis S d 3 4T peninsularis << sca cles en 546 PEIN GIGI: EE 538; 545 x f SOOPER: 420; 544 EE 373 Heen ..6; 17; 18; 440; 443 GE EE 0 442 ES DEE E 441 spectabilis ...... 386; 420; 440; 441 Macrosiphonia 7275999 cle vie 9 ere EE 6; 1; 8; 102 175; 18; 173; 1/4 E Gx, OOO SE 184 Berlandier date e 179 Brachysiphon eras. as «so 176; 402 CHLOTOMTAQU ara peer IE SEM 185 THOR PETIA tae io A 175 hypoleuca wc, ao + cle setis 22.1775 179; 410; 421; 541; 542 poe ae 178; 411; 541 longiflora ..185; 400; 408; 409; 412 var. Guaranti0l .......oooo.. 86 Macrosiphon ...¿........> 179; 412 qe ce ee 180 pedunoularis ...... ern 184 ncm By ae oie OBO Oe 182; 416 ER 417; 548 var. Geer E 183 Var. PINE) OMG one siete mee oles) 548 var typiea cee gees cee ss sale 183 pinifolia ...... ennt 183 forma glabrata ........... 184 417 Page var. intermedia... e Nee 184 forma peduncularis ....... 184 forma solot Ee 184 DT OSUTOUG cs WR tio noit nie 83 IVelame v. cum eI 174; 184; 423 < gojagsensi8 EE E 184 verioulata: os dent unes soothe 182 PANUCTINC OMA eee ees. 183 y. peduncularis .......... eee 183 ADEMAS Wa ais rie s er 183 var. pinifolia ........... 183; 548 EES 181; 424 VAL IMISONUM sss. E EE 181 MERA VAS ONU E TT Maljusta a serene ee 5; 8; 14; 18; 19; 272; 291; 292; 420 A WER 87 amplexieaulis ............» 295; 296 EE dore 283; 400 SS O ER «ors syste 303 LE 274; 275; 282 CADA E VY ee ; 278 dem e EE 284 MAL e Rode IA PME 276; 277 furfuraeea ...... 274; 288; 290; 295 B. grandifolia .......... 289; 291 ghadulfera AA 290; 293 EE E les SE 297 TACO: oed EES 299 guatemalensis CHE 301; 302 CANIBUS. AA 287; 303 EE 303 ER ee tra 296 MOUs EES 303 HOO a IE E E 283; 284 EE, el 279 ME ORE Wires Ii Gres, e cis esate ss 283; 284 r2 E, E ote 275; 280 e 300 ODIA ads 287 A O 286 E E EE KE EEN 301 a A reaa 293 EE E O 303 pubescens ........oooooooo.... 285 erte SE e x.t 277; 278 418 Page A eat ate eeu ee 30 Behombnrpkli se oe 278 sossa soccer cas San ee 303 Tamaquarina aa a ek 273; 274; 286; 290; 291; 293; 295 ge PAROS OU = in inca e 287 By NEE 286 AS O a 287 teirastaohya oS 303 vir o ED DOR EM IEDE 298 Mandevilla ...5; 7; 8; 10; 12; 16; 18; 41; 174; 352; 409; 413; 542; 549 n Pato lan KEE 73 CO Oe Ee 45; 398 Albo-viridis oc E 86 ANCONA EE 170 STEE LCE 41; 84; 399 an Oase E see antennacea 137; 139; Ce 400; 402 apoc olia? mE BC we i nohis E "400; 549 attenuata sce tk ces iso 157 auriculata c cee teen eee E 160 ener 89 eil Benthamii ..... 8; 40; 170; 399; 401 OgOtensijs Ee 73; 402 BoWiana EE 137 boliviensis iis 5085 109; 112 brachyloba eere ; 66; tenta oet 144; 157; 402 bracteosa cc oe ct 138; ps re calaca tons oon) ites oe = callista agai te 9; 43; 96 CAU OO E pt cercophy UE ie eis Dn E Pur p OR 130; 133; 404; 424; 432; 541; 549 GOTT 55 congesta ........ 9; 43; 71; 404; 417 conyolvulacead 2.42... e PE Soa ete T ln TEE 99; Ge 2 (590) [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN cuspidata ....... 126; 127; 129; 340 Gentiouldiqu © e e ae CT V es Ook oes 168 Donnell -Smithit: oon ses is cee 46 equatorialis EE erecta 51; 57; 68; 71; 382; 407; 423 OES ge eee ain, ess 102; 103 e CAE EIUS 154; 550 E EE UE 54; 266; 272 Praca era A fragrans o 12; 407 funiformis ..... 9; 43; ae 408; 414 var SEH EE glabrae SA 105; 167; 548 glandulosa ....... 88; 340; 408; 539 DEE eie apes hirsuta 22... 8; 163; 147; 15 368; 382; 399; 400; 403; 407; ae 412; 418; 421; 422; 430; 555 Kee 123; 127; 129; 410; 419; 424 r. glabra ; labra ee ; 399 VAr AAs e PUE ; 949 immaculata a Jamesoni Te ceo oe eee eee 62 j&vilansia AA 171; 410 ere te KEE 53; 405; 410 ONE auer E E e e Ee ..146; 147; 411; 430; 555 laxa eae ee 42; 91; 411; 421 SE E 152; 411; 550 O 129 UNC ee ore rience 152 Loesneriana. A A 86; 87 lucida o be oes eee Luetzelburgii ............... 9; 94 SOUCHE 89 LAND eee eerie als 98; 413 Var RlaDLa e ec. 549 Waray, PCAs TS SHOOT XII mencana ya eer 56; 414; e mollissima EE 144; p as montana 85 e CNN VAT. peruviana NEESS EE e 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 419 g Page Moricandiana ............ 101; 103 ae oes. eer eee ee EE 63; 421 Var DARIO eo ros 102 a EA ers 41; VUE INDICA a 101; 102 135; e 370; 409; 417; 422; 549 Motitziana os cc's 4909092. 164 BUPEr E eder ds Mueller ....... 1 e renes 113 a ARA AN inyriophyllum ¿0er incoado 78 Symphitocarpa .... 2 xn 151; 422 acana 40; 82; 408; 415; 430 MYIHLE -- EE hee ees 45; 49; 51 ODIONGILOUA: «0.5.9 e ; 99 o eio cu bios re vate PAVOL Bie Siete a IET Ae folge eei 75; 416; y Pavon 24-2 145; 409; 416; 538 FOMENTO I. A one se e hotels eere E ost eects ete E ae 121 FPR SUL es Bac este cs et en Pentülandidng 05 ad 51; 67 VALE MOST E ooo. exe avers 160 DeomHxta EE 11 VR V GILCONG e esie PHI 2352121. E dor 550 torosa ...16; 40; 52; 249; 403; 422 platydactylu a. cee RES 47; 48 SEKR, Ok Ee wena na ET EE 165 tubiflora .......... 44; 47; 404; 423 DOLOSMA Tot. IEEE E 173; 433 A 72522579 N er oie sere 91 pulchra 4.002505. v. 548; 553 LU 167 pycnantha ......... 57; 64; 406; 418 dE A E 118 TIPA 405.202 99 61; 399; 400; 418 pf) 1.25 oro EX ras 167 ROSATI crite aire S © sates. ote 48; 419 velutina ....127; 129; 130; 424; 541 TURNOS E A ens 150; 419 var. angustifolia ...417; 548; 549 Lt 160 VAIO EE 128; 548 Ha. were aha E 153 ee ee 128 INCITA. de cnx Wee ies ass poc 164 MI EE eee eae 173; 433 ELE 122; 419 TU eee ET eee e BAnderi eet 116; 117 veraguasensis .......... 86; e 424 Ree E vice A EEN Sg 114; a oo cese 143; 144; 165; Sc 148; 370; 383; 401; 402; 403; VARIA ......... EN 405; 413; 418; 420; 422; 424; 549 Marsdenia Schlimi E POS MENOS 155 E EE 413; 549 Schwmanniana ........ eere 327 ¡MOROS oso bes ela E A o O ates 59 : 6j 8; 10; 16; 18; hd Sa 415 Below 0. 115; 117; 118 GCAO C. or eX vs. 413 SOPLO 40000 0228. 45; 50; 51 ABOVE A Ere RA ; 84 Sin ing. un 198: 133; 418; 420 A A A 40; 171 SEELEN re. 108; 421 eo HSS O ON E o 19 EE ; 399; 402; 404; 411; 549 SUGUEOLERE LT otv UNS. 91; 92; 421 callada UNA toca sy us 30; 401 SEELEN vo. cc. 40; 168; 421 lup C TCU ONT sitter hss Aa subeordata te as ateo leads o AN E ate 40; 52 (ERT A ee ; 412 o EE 33; 402 subsagittata ...... 40; 139; AER EC 1 3; 144; 400; 405; 408; 409; ITT AA T 27 410; 413; 414; 419; 420; 421; 431 Guayaquilensis ............ 40; 140 "m MORTE usd 49 CUMMINGS Bios ei kote esis 41; 136 Page hastata oo oos ak van a pers 40; 140 hutoa note hs aa aac eh 40; 140 Mritellula. oi cats NEEN EE 40; 82 JO DUC ORM EE jasminiflora ......o...o.o..». 40; 140 lanceolata `... 35; 36 linearifolia ENNEN ENN 36 Mansoana gege 25; 32; 413 EE 39; 414 myrtifolia ..o..oooomomomom.m..o.» e ENEE EE 35; 36; ae "418 TOROS E EE 3 4: 19 anctae-Crucis ............ ; 42 ubcarnosd senseo eaa 40; 168 sulphurea EEN 32; 33 surinamensis AN 7 Orosa EE 2 trifida 26; 29; 31; 40; 403; 406; 410; 415; 418; 419; 422; 423; 549 Mior ade Ma, rosca oror a e ;41 acuminata EE 549 GtroVtolacea SIE 120 EE ENEE 549 OR 99 A ee re 110 hirsutula T ers 98 EE BE Moricandiana ................ 102 ROGUE 549 EE 110 NM UL 115 splendens... esee seres 103 urophylla- EE 118 EE ee 6; 304; 449 Blanchett EE brachystachyus `... 149 EE 34 OI 452 CUYAD EE 149 e EE 550 EE 93 funtfOTMis ......... ENNEN 93 te 328 HEES 93 leptopyhilus ........ooomooo.o.. 152 [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN Page Mexicane aioe eh EPI RA 529 uge Kee 93 MUD Fo. LT EG 151 symphitocarpus .......... ese 151 fenollus EE 334 tenuicaulis ..... ccc cece eee 149 UOT RICO EE 148 Molopanthera paniculata. «cursos xus oer RR 257 Neobracea AAA 6; 8; 18; 341 angustifolia ege peas nas AMEN o sees sense 342; 344 Valenzuelana ................. A 342; 370; 383; 423; e d O Neriandra suberecta «02... ee ccc eens cece 372 erium sarmentosum .............L. E 372 pa inp Bee hase died CUR ; 904; 332 uu 367; ns 556 Ba SO O T ANA EEE 314 AMAZONICA 6... cece ccccs veces 313 angustifolia `... 334 a TE 9; 329 OH 30 caudigera NNN EEN NN 556 INIA oa EE 307; 308; 309; 311; 368; 404 Cordata Ee Cordigera ...........o.ooooo.o. 316 COACE A E 337 EE euspidata EE 126; 340 Dusendschoenti .........00 cee ee ONMOSG wean nes sais a 331 Tunigcora A 308 E ECH EG 336; 368; 369; 405; 407; 408 glandulosa ........ lesu. 340 D DEE 338 OO 328 acilipes ....... 305; 328; 368; 408 gracia occ ccc cece wesc eee ER EE 331 Se EEN Sol 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 421 Page Page Hoffmannseggiana ............ giganteus 155. sare ee EE, de 360 "D c He 05; 308; 330; 332; EECHER SOS 370; 408; 409; 410; 412; 422; 557 s ios CPM: e MERE 356 hypogiauca E 335; ; 410 MAVAE LOLE 1.225 vix ge « 354; 355 nb "DUE EC 315 peltatus 359; 356; 363; 412; 416; 417 e A EE 325 stemmadeniiflorus ............. 355 lauretiana de cs ssc 3s fess 323; 325 Perictenia EE, up A 339; 340; 547 BLADES EE 329 (UCLA ae Faces RUE 334 Periploc E, en A 252 517; 319; 370; 406; 412; 424; 556 Prestonia .............. b; 8; 10; IIBEIOCUS EE EE 311 12; 14; 16; 18; 292; 448; 516; 558 nitida. uas 334; 369; 412; 414 tota SE 401; 410; 411; 460 B. CORMAN TEE 334 Vat, TOMSBUING: EN 461 y. angustifolia ....o.oooo..... 334 TE ee E E 334 398; 411; 418; 420; 423; 430; 450 'Perrotteti- va ce oc 2 525 A Partem 323 AmAMIUOhnBIS EE ss eU a 31 BOEDA or EEN 337; 369 Aa 05 fora en 8 polyneurs EE 338; 383 SE 400; 430; 481; 483 puncticdlosa ads 24 23 EE o E 521 311; 319; 368; 369; 405; 418; 556 rn EE 526 Sege TEE, 23 DONNE LU E 488 hunn Strachan 326; ee 547 adm ee T ESI 523; 526 SPORE e Coi ROV CODE 32 COMMONER Ed e BEE, de e 519 spoliata: cas CURT xr €? "321 Wier yr 413; 422; 423; 452; 454 T RR UNE. 313 EE, ss ae 503 FRE A D SSS 333; 558 o AA UT EET PETI 524 Surinamensis 5-5. 950. coto 310 Gi Age Au dE oleo 6; 456 iki in ME E. 331 denticulata .......... 406; 421; 500 VEITUCOSS™. oA EE hari PA Se eee 480 368; 309; 370; 399; 401; 424; 556 Praiss AE 558 H EE ECG 318 DENM AAA EE ERIS Y 407; 453 forma MD CD 318 CCUGOOT CRUS Ais cat sies lata 478 forma ovalifolia ............. 318 ihe MORAL A Ee 494 var. tomentosa forma. scandens 318 OXBOTLH UD o T 468; 472; 474 Odontostigm Up ee eT Aie 484 aleotitamam 2 SEN re ee 401 Gaudichaudá ................. 500 Orthechites gia orba. d LIRE eee 478; 494 GORGDM S etd ro scare etr Riese se 265 Reen EE 538 LST 19; 247 EE 472 Lie sii ic hese tetas tate pte 67 gegen GEN AAAI EE 511 MUD Ee E 252 EE E c 482 floribunda S si No Lets 248 ates O QU 456; 457 teptocanpa A a 224 EE EE 145; 523; 538 aspicOla d c E err roe 45 imas AI E LUE Ue Lt 22 Peltastes ...... 7; 14; 17; 18; 353; 363 E A A 532 COLOMA o e 56 "pam SOEUR UR ICE EUMD 514 [Vor. 23 422 ANNALS OF THE MISSOURI BOTANICAL GARDEN Page Page laeta. EE 501 d'Or Oe 449; 519 lagoensis A sec O 468 trifüida 4... 486; 492; 494; 497 lanata eer nee 519 vallis c istic ects 507 Tanglasset a E 538; 545 WAN Kee EE 495 latifolia rr rer eres 519 EE 468; 473; 476 BEE 489 versicolor.) 5-2 EE 504 TIindleyana- ERAN A 499 TT A a 478 TO 460 Prestontopeis ...... + s=onon... 6 longtiuba S. eoe ee 532 ln EE 72 LULERCENS A earn ee are 519 LL EE 72 maecrocarpa ses re nies 512 Se 72 MACTONCUIA asec srs ee sie 493 EE 72 rginata |^ on E ADOS 408 Rhabdadenia Ee marginata: ae es centre ee nets 464 ; 7; 10; 11; 14; 15; 18; 367; 377 Meg agros cee 413; 501 DADA e 73; 382 megalagrion .....oomooo mo. moro. 501 CULOTU A ONE 365; 382 EE HE 529; 532 biflona os 15; 378; 401; na 416 ollis: EE 478 OOM DESIG oS 82 mucronata. cess ae 527 EE nk 383 Muelleri Geer EE 535 OO 78 OS 502 A UE 366; 383; 405 pachyphylla ......... 456; 464; 466 ren GEO 378 EE 431; 52 latifolia was: scissors ee 381 parvifolia 3.220.500 468; 474; 478 EE d BEE 381 pere OO OO EID OE TRO ID On 506 Date 73; 383 perplexa o aos 476; 539 Lindeniana .266; 364; 367; 383; 411 perwiana. o E 539 B. E EE 364 Phenax 6-61 4. 484; 486 das. dr ET E 334 piumierifolis m ee 497 macrantha een e 78 portobellensis see 417; 512 macrostomas EE 380; 413 purpuríssata EE 483; 559 EE 383 quinquangularis .415; 418; 431; 459 OR 381 Raedelii n.. 418; 431; 534; 538 NOVO EE EE 378 NOOO 4 DALLAS il 378 Trotundifolia cota csia to 490 Ohlii- daa ata 378; 381 uba OOOO SN 509 HE 381 Sohieadenia oe EE 512 KE E 381 Schumanniana . 22... ees 53 VAN, VOlUOUG tei, ier sates 381 EE e 522 POEMO EE 383 CA Cova eee ey e eee tes 529 SaGgraet EE 366; 383 mult Ne EM I M Wrightiana .......... 342; 383; 550 IU E vO 456; 465 Rhaptocarpus .................0; 449 els SE 430; 431; 450; 454 DEER 452 BUER EE 533 EEN 452 SUTINAMENSIS eo co. 518 EE 480 LODAGENSI AS leas eins eee 472 MOTA TELA EE 452 1936] WOODSON—STUDIES IN THE APOCYNACEAE. IV 423 Page age EhodocalyX 2. csc eret rmn Süpecomat rs ston cts 6; 14; 17; 18; 361 Ne Ge 7258s 14, I5 I5; no 539 maerocalyz ..............997; 363 CUY COSUL EE 541 HEEN 357; 363 TEE 541 UG Rca ranas senate 357; 363 El EE 130; 541 Daraboltca A E 357; 363 CTS SUID MS T EE 348; 541 E 357; 363 OH OSSUDES eden sacs RAS SE 398; 541 poltigera cy. vee ae 362; 416; 423 OUNEU, OUUS: E E SE 541 DIACGIO oor T PICS ES 357; 363 RUDOLCUCUS Pare ole 177; 541 PUNA ce ced 357; 363 VONUGUNOSUS: OO E 178; 541 EVECIOS EE EE 357; 363 EE 131; 541 Streptotrachelus ..............0; 542 rotundifolius ........ 407; 539; 540 UNG LES EEN 539; 545 SUAVCOLENS onoo eei on 114 041^ o A mE 5; 187 SE EE 131; 542 Tabernaemontana ...4; 398; 420; 431 mcr e 5; 273 ap eres eed ca een. eee OUR EE 282; 291 COSI OIGe RI DET I are see EE 282 DE XOU DUET A NUS 283 ECK 396 EE 303 grace EE 297 Salpineteg .... o <...ons.s 350; 352 JT ONAN OM E 303 Guide eset ee actin 352 KEE 283 El 351 OOP CG rs POET eee Msc 286 Secondatiae E TINE EECH 303 16; 18; 19; 258; 262; 265; E 284 LEE e 268 ar Von SUMUS AC E M CE 283 ar A IO E 303 EECH ee E 259; 262 mea EE ave a TE EE 259 ; 10; 17; 18; 42; 214; 425; 449 ar. paraguariensis ......o.o.o.. 259 mino: A com Fr EP 3 81 2 DOS OUO LU. EI EIE 259 DIOS IDEE esc Se, a cee dre 427 CLG OT AS. EE 266; 268 CONAGUA aig erre sis eiui. 82; 430 IDI {ol oem duos 259; SE 265 COTTUGUATO EENEG 430; 455 ARAS Se: 2 A NA 427 floribunda - 9-5 259; 263; 264; 265 glaucescena coL eu 82; 430 Talsasq EE 263 laatacat nd oo EE 146; 430 Gacdnert.s o5 E done 263 leptolüba- eats aie 29 c ege 430; 451 $ (30601800). EE 263 OD DEAN T eee res 146; 430 y. petiolaris En 263 EE 415; Mare 258; 265 paldiftora EEN 160; 427 DEE 261; 262 an aie tipo tec cere ens = miele 160; 430 Schlimisna ai. eee ee ees 262 DAROCA Eer 431; 528 stan Gate a EE 55; 266 quinquangularis .......... 431; 459 Stemmadenia TCU CUA O c's caret E 431; 535 e E 401 SCOURTAHONG A EET T 140; 431 grandiflora .........«.........<. 303 genAdAgs ma eine wie eis So ties i> 431 guatemalensis- actrees lace: 301 golami o 2 er ; 454 obovata var. EE 29 StellAriBo ege 421; ez 426 424 Page slenantha 7 s E 426 tonuioula ne a 431; 455 COMENTO A tia 160; 431 violacea ....407; 413; 424; i; 429 EA OSE) SOG ao 131; 431 Thenardia 5; 10; 14; 17; 18; 423; 443 OO DOE 242; 448 ribunda eee ee 444; 445 Galeottiana cn wana. eee 447 gonolobioides ............ 446; 448 Mr 212; 448 BUCEO ONE io 445 Md m ETE 444 ellata SN ; 448 TLhwrsanthug ee 5; 187; 420 COLON EE 203 adenobasis ............ eee 241 af nis E E 200 e BEE 242 Benthamiana. ....... SEN NEEN 204 DALLAS 236 Brasutensts i. vr see I 224 OORT E 245 corymbiferus ns es EL 242 COFYMDOZUB NEIRA SE 252 le MO TEE 226 OM EE 268 dtospurifolius 2... ses E 209 embeltoides caos ie 193 Gardner Vo ogc oc ne ae ae 199 e EE 228 HA 19 EE SE EE 206 Re 212 ca AS EA NOR Lc 224 ohnat EEN 195 macrophy BARRENS IIS 242 DEEST 194 DEE SE E E 234 [Vor. 23, 1936] ANNALS OF THE MISSOURI BOTANICAL GARDEN Page MUTAONUR US eee Ee tee ee 254 EE 193 HE 218 POPULAR 268 MUDESCENS cra Ee 218 PUTES OT mas BARN OGRA AOA ASAT 555 EE 6 BA OOO ern aoe 201 fae, e E 203; 204 Ek 245 ee GE ET E OUS 208 Tintinnabularia ee 559; 562 EE tr 561 P UR SE 18; 19; 266 EE EE E 303 IMM 19; 266; 267; 268; 270; 272; 406; 417; 419; 422 autoen 267 SLOT rere E ; 266; 272 Urechites ...... 6; 10; 14; 16; 19; 370 AMC a E e 373 A 375 dolteantha NO 373 e DEE 372 KarwinskW .......... eee. To 433 EE 16 371; E 376; mn 383; 399; 401; 403; 406; 409; 410; 414; 415; 421 var. am. hy eti alia RA 373 ING INU anche Gan Daa oA Goss 373 EE EE 373 suberecta m vcr 372 pepglabrata o oso See reas 372 y. rotundifolia .............. 372 Vinea lutea ...... 372; 382; 383; 399 ut 401; 403; 406; 409; 410; 414; 421 sternutatoria Se ehre ere le Dee e eis AAR] BO KE ee 303 N. B. As the final section of ‘‘The American Genera of Echitoideae’’ goes to press the type specimens of Sessé & Mociiio’s ambiguous species of Echites are Museum of Natural History, C be included within the present account. be published in the near future. (596) received from ke through the courtesy of Mr. Paul C. Standley, of the Field hicago. e poorly understood en- esult these tities may now be assigned definite status in many instances, although too late to A subsequent paper concerning them will [Vor. 23, 1936] 426 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 1 Allomarkgrafia ovalis (Mgf.) Woods. Habit (x 34) and dissection of calyx and reproductive organs (x 10). Fig. 1. Anther, side view. Fig. 2. Anther, ventral view. Fig. 3. Dissected calyx, showing internal squamellae; nectaries, ovary, style, and stigma. (598) a ie ‘ot. T Ae ae = WAS = ANN. Mo. Bot. GARD., VOL. 23, 1936 A EE PEA S] o 6 m " € peu ae EON NAT RES E A EY AS LII ARSAL a FP. LON PI QUOIOXY NDREWS—SE A [Vor. 23, 1936] 446 ANNALS OF THE MISSOURI BOTANICAL GARDEN EXPLANATION OF PLATE PLATE 21 Sequoioxylon Pearsallii, n. sp. Fig. 6. Transverse section showing gradual transition. Fig. 7. Transverse section showing abrupt transition. "Fig. 8. Strand tracheids in late summer wood. Fig. 9. Portion of same more highly magnified. PLATE 21 3, 1936 Di ANN. Mo. Bor. GARD., VOL. e E ween ` Ri anie AT Kar Bes Sg eS I EI Qt. B." ee Se ml Tm vea - t^ M ume = See? Sg O OS - w^ z Fax * " x 2 m - m: = eme ez. eg era K y y ANDREWS—SEQUOIOXYLON PEARSALLII FIELD AND HERBARIUM STUDIES, IV! LOUIS O. WILLIAMS i Washington University Fellow in Botany Picea pungens (Parry) Engelm., Gard. Chron. N. S. 11: 334. 1879. During the past three preceding summers occasion was had to study the distribution of the Blue Spruce in Wyoming, of which there is probably a great deal more than is generally realized. It extends, so far as was observed, along the Snake River, from Jackson Lake, in Teton County, at least to the Gorge of the river in Lincoln County, as well as up all of the tributary creeks and rivers. Along the Snake River, some fifty miles, it is the predominant conifer and is exceedingly abundant, but along the tributaries which usually gain elevation rather rapidly it is of less importanee. In Sublette County it was found around Fremont Lake, Half Moon Lake, and other of the lakes in that vicinity. It is also to be found along the Green River but there it is not common. It does not ascend the river as far as the Green River Lakes; whether or not it goes farther down the river than a point due west of Pinedale is not known. The tree is also to be found, in this county, along the Hoback River whieh is a tributary of the Snake River. In Fremont County it was first observed near Dubois on the Wind River and extends up that river from there for several miles, and is also on some of the tributary streams. How generally it is distributed along the other creeks or how far it may extend down the east side of the Wind River Mountains remains to be investigated. Acquaintance with the tree in Colorado does not quite give one an adequate idea of the species as it occurs in northwestern Wyoming. At first sight it may not be realized that it is the Blue Spruce. It is almost entirely lacking in any “blue cast." Again it is to be found only along the streams or at most no great distance from the water. The size attained is much 1 Issued September 1, 1936. (447) y [VoL. 23 448 ANNALS OF THE MISSOURI BOTANICAL GARDEN greater than any the author has seen in Colorado. Trees four to five feet in diameter at shoulder height are not uncommon; one measured just under six feet in diameter. The height is in proportion. The altitudinal limits seem to be rather definite. It is rarely found above 7500 feet altitude and so far as observed not below about 5700 feet. The best development is between 6000-7000 feet in Teton County. In this region the Blue Spruce is often associated with the Engelmann Spruce (Picea Engelmannu Parry). One tree, in such an association, observed above Dubois had cones which seemed to be intermediate both in size and in shape of the cone scales between the two species, but in- clining slightly more toward those of the Blue Spruce. Cones observed from a tree on the grounds of the Lee Ranger Station, near Wilson, showed a large range of variation, particularly in shape of the cone scales but cones were not so small as on the tree near Dubois. Salix Tweedyi (Bebb) Ball, Bot. Gaz. 40: 377. 1905. This willow was found in abundance along Cascade Creek, one of the cold glacial streams in Grand Teton National Park. During the past summer it was found again in the Big Horn Mountains of Wyoming where the original collection was made. From Dr. Carleton R. Ball, who has kindly determined the ma- terial, comes the information that sets of this species have not been distributed among herbaria. My numbers 1133, 1668, and 1686 show it in various stages of development. Specimens were collected in a spruce bog near Bald Moun- tain, elev. 9500 ft., Big Horn Co., Aug. 29, 1935, Williams 2511. Salix cascadensis Ckll., Muhlenbergia 3: 9. 1907. This plant is quite common in Grand Teton National Park, between 9500 and 11,000 feet altitude, where it covers acres of ground. My numbers 915 and 1352, determined by Dr. Ball, illustrate it. Salix arctica Pallas, Fl. Ross. I. 2: 86. 1784-1788. This species is not mentioned in Rydberg’s ‘Flora of the Rocky Mountains and Adjacent Plains’ or in the Coulter and 1936] WILLIAMS—FIELD AND HERBARIUM STUDIES, IV 449 Nelson ‘Manual of Rocky Mountain Botany.’ It is found oc- casionally in Grand Teton National Park. My number 1708, determined by Dr. Ball, represents it. Populus taccamahacca Miller, Gard. Dict. ed. 8. Populus No. 6. 1768. Both the broad- and narrow-leaved forms are found in Teton County, Wyoming. The narrow-leaved form occurs along the streams in the valley. The broad-leaved form, which Dr. Alfred Rehder has kindly determined and which he informs me was not previously in the collection of the Arnold Arboretum from Wyoming, is to be found in the canyons of the Teton range. Xerophyllum tenax (Pursh) Nutt., Gen. 1: 235. 1818. This genus seems not to have been reported from the state of Wyoming. Two specimens are at hand, both sent to the author for determination and both from Teton County (slope north of Jackson Lake, July 11, 1932, W. B. Sheppard; four miles west of Cascade Creek on the road to Ashton, Idaho, along Reclama- tion Road, June 25, 1934, Mrs. J. W. Orlob 1633b ). Professor Nelson, in Coulter and Nelson's ‘Manual of Rocky Mountain Botany,’ p. 118. 1909, states the range as: ‘‘Montana, possibly Wyoming, and west to Oregon." In a letter he informs me that there are no specimens in the Rocky Mountain Herbarium collected in Wyoming. Paeonia Brownii Dougl. in Hook., Fl. Bor. Am. 1: 27. 1829. The published floras covering the Rocky Mountain region do not give this genus as occurring in Wyoming, yet several speci- mens from the northwestern part of the state have been in existence since 1860. More recently several collections have been made from the same area, of which the following may be cited: Jackson's Hole, on Snake River, June 18, 1860, F. V. Hayden; on Henry's Fork, June 19, 1860, F. V. Hayden; Lake Fork, June 22, 1860, F. V. Hayden; gravel flats, Jackson's Hole, Aug. 3, 1920, Payson & Payson 2169; near Jenny Lake, Grand Teton National Park, 1931, Mrs. A. C. Lyon; gravel flats near Jenny Lake, Sept. 2, 1933, Williams 1434. VoL. 23 450 ANNALS OF THE MISSOURI BOTANICAL GARDEN The Hayden specimens were sent to Dr. Engelmann for de- termination but were not included in the published list of the plants of that expedition. They bear no annotations by Dr. Engelmann. Aquilegia Jonesii Parry, Am. Nat. 8: 211. 1874. This rare columbine was found again in the Big Horn Moun- tains. It seems to inhabit only the caleareous rock slides. My number 2358, July 5, 1935, represents it. Roots and seeds were sent to the Cheyenne Horticultural Field Station at Cheyenne. It will be of interest to see if it sur- vives in cultivation. Ranunculus jovis A. Nels., Bull. Torr. Bot. Club 27: 261. 1900. While the rocky flats of Grand Teton National Park were in most places still covered with two feet of snow and the tempera- ture was below the freezing point each night this little plant was thriving. Indeed it had nearly matured its seed before the snow was gone from the flats. It is of interest to find this at an elevation as low as 7000 feet, since it is usually to be found in the alpine regions. (Williams & Pierson 1074, April 24, 1933, distributed as R. glaberrimus Hook.?). Thermopsis rhombifolia (Nutt.) Rich., var. annulocarpa (A. Nels.), comb. nov. T. annulocarpa A. Nels., Bull. Torr. Bot. Club 26: 239. 1899. Professor Nelson in Coulter and Nelson's, ‘Manual of Rocky Mountain Botany,’ p. 271. 1909, referred this to T. rhombifolia as a synonym. However, it seems to merit varietal rank even though the characters by which it is distinguished are superficial. There seems to be another closely related variety of the species occurring on the western border of the range. Whether it represents an undescribed variety or is T. arenosa A. Nels. will have to await a study of the type of that species. Dalea Grayi (Vail), comb. nov. D. laevigata Gray, Pl. Wright. 2: 38. 1853, non Sesse & Moc. 1832. Parosela Grayi Vail, Bull. Torr. Bot. Club 24: 14. 1897. Thornbera Grayi Rydb., N. Am. Fl. 24: 119. 1920. 1936] WILLIAMS—FIELD AND HERBARIUM STUDIES, IV 451 Dalea Thompsonae (Vail), comb. nov. Parosela Thompsonae Vail, Bull. Torr. Bot. Club 24: 18. 1897. Among a fine collection of plants received from Mr. Bertrand Harrison for determination is a specimen which seems, ex char., to be referable to this species. If the determination is correct it represents a considerable extension of range. The specimen bears the following data: dry sandy wash, exces- sively alkaline soil, Henry Mountains, Vanadium Mine, Gar- field Co., Utah, May 20, 1934, Harrison 7520. Hoffmanseggia tenella B. C. Tharp & L. O. Williams, n. sp.? Slender perennial herb, 8-15 cm. tall; stem proper short, unbranched or nearly so, terminated by a few-flowered simple raceme; leaves bipinnate with 3-7 pinnae, as long as or exceed- ing the inflorescence, sparingly soft-pubescent, the petioles 5-13 em. long; the pinnae with 5-6 pairs of pinnules, the pinnules sessile on the rachis or nearly so, 2-4 mm. long, 1-2 mm. broad, oblique, glabrous on the upper surface, sparingly pubescent on the lower surface and margins; stipules small, scarious, 1-2 mm. long, adnate to the petiole; inflorescence not exceeding the leaves, usually 3-5-flowered, each pedicel subtended by a short scarious bract ; calyx about 4 mm. long, the lobes linear-oblong, obtuse, slightly naviculate, finely but densely pubescent ; petals obovate, attenuated into a very short claw, 3-4 mm. long, 1.5-2 mm. wide; filaments free, with a few short hairs; mature legume 12-15 mm. long, 4-6 mm. wide, straight, finely and rather densely pubescent but not glandular; seeds 2-4. Texas: Robstown to Alice, Nueces Co., Nov. 22, 1931, Mrs. F. E. Clements 128b (Herb. Univ. Texas, TYPE; fragment and photograph of type in Herb. Mo. Bot. Gard.). This species seems to have its nearest ally in H. drepano- carpa Gray, from which it differs in several aspects. The ? Hoffmanseggia tenella B. C. Tharp € L. O. Williams, n. sp., herba perennis grac- ilis, 8-15 em. alta, simplice aut sparse ramosa; inflorescentiis paucifloris terminal- ibus; foliis bipinnatis, 3—7 pinnis, pubescentibus vel fere glabris, 5-6 paribus foliolarum; stipulis parvis, 1-2 mm. longis; calyce fere 4 mm. longo, pubescente, lobis lineari-oblongis; petalis obovatis, 3-4 mm. longis, 1.5-2 mm. latis; legumine 12-15 mm. longo, 4-6 mm. lato, recto, pubescente, sine glandulis. [VoL. 23 452 ANNALS OF THE MISSOURI BOTANICAL GARDEN leaves of the latter have 7-11 pinnae, those of ours have 3-7. In A. drepanocarpa the inflorescence usually exceeds the leaves and has several flowers, while in our plant it rarely if ever ex- ceeds the leaves and the flowers are fewer. The legume of that species is 25-40 mm. long and is strongly falcate, while that of ours is about half or less that length and straight. Our plant is noticeably more slender than H. drepanocarpa and has fewer stems to the root. Dryas Drummondii Richards., var. tomentosa (Farr), comb. nov. D. tomentosa Farr, Ottawa Nat. 20: 110. 1906. That this can be maintained as a distinct species on its rather meagre characters is doubtful. Juzepezuk, in Bull. Jard. Bot. URSS. 28: 311. 1929, places it in a new section, Nothodryas, of Dryas, along with D. Drummondi Richards. and D. grandis Juz. With the former, at least, it is closely related, but the latter has not been seen. It is of interest to note that Juzepezuk, l.c. p. 325, describes a new species, D. Hookerianum from Rocky Mountain material. The writer is unable to find specific or even varietal differences between available material, which he cites, and European ma- terial of D. octopetala L. Zauschneria Garrettii A. Nels., Proc. Biol. Soc. Wash. 20: 36. 1907. Z. latifolia var. Garrettii Hilend, Am. Jour. Bot. 16: 66. 1929. Finding this species above Bradley Lake in Grand Teton Na- tional Park came as a distinct surprise to the author. How- ever, on looking up the distribution of the species several collections were found which were out of the range, ‘‘in mountains of Utah and southern Wyoming,’’ given by Miss Hilend in her revision of the genus. The only specimen cited for Wyoming in that revision is from west-central, not south- ern, Wyoming. The following specimens may be cited: Wyomine: hills east of Afton, Aug. 8, 1923, Payson € Arm- strong 3771; ledges above Bradley Lake, Grand Teton National Park, Aug. 14, 1933, Williams 1403; mountains west of Cody, 1936] WILLIAMS—FIELD AND HERBARIUM STUDIES, IV 453 Park Co., July, 1905, Worthley; mountain top, Holm Lodge, about 40 miles west of Cody, Aug. 26, 1922, von Schrenk. Rhododendron Warrenii Macbr., Contr. Gray Herb. N. S. No. 56, p. 55. 1918. Azaleastrum Warren A. Nels., Bot. Gaz. 56: 67. 1913. A consideration of the type and two subsequent collections of this species from Colorado, contained in the Rocky Mountain Herbarium and kindly loaned the author for study, raises a question concerning the taxonomic status of the species. The collection on which the species is based is rather meagre ; however, the description given for it is accurate. It compares very favorably with the abundant material at hand of R. albi- florum Hook., Fl. Bor. Am. 2: 43. 1834 ( Azaleastrum albi- florum Rydb., Mem. N. Y. Bot. Gard. 1: 297. 1900), from the northwest in a similar stage of development. An excellent collection, mountains due west of Walden, July 20, 1930, Leonard Johnson, from the same region, perhaps type locality, leaves little doubt that the plants are the same as those from the Northwest. The several hundred miles between the Colorado station and the nearest known station in Montana raises an interesting question in distribution. Dr. Rydberg gave the distribution and range of Azaleastrum Warrenii, in the ‘Fl. Ry. Mts. and Adj. Plains,’ p. 640. 1917, as * Mountain slopes: Colorado." However, so far as it is known, it seems to occur only at the station at which it was first collected and near by, not nearly as widely distributed as Dr. Rydberg’s note would indicate. The three specimens in the Rocky Mountain Herbarium are all from Jackson County. Nemophila petrophila n. sp.? Low annual, 4-13 em. tall; cotyledonary leaves persistent, op- posite, obovate to oblanceolate, entire, 1-2 em. long, 4-6 mm. 3 Nemophila petrophila n. sp., annua humilis 4-13 em. alta; foliis eotyledonium oppositis, obovatis vel oblanceolatis, 1-2 cm. longis, 4-6 em. latis, integris, infra glabris; foliis eaulium ovatis, 1-2 cm. longis, pinnatis, lobis ovatis, integris, utrinque strigosis; floribus axillaribus; calyce fere ad basin diviso, lobis lineari- lanceolatis, 3-5 mm. longis, ciliatis; appendicibus in sinu 0.5-1.5 mm. longis; corolla fere 2 mm. longa. [Vor. 23 454 ANNALS OF THE MISSOURI BOTANICAL GARDEN broad, glabrous below, the petioles joined and sheathing the stem; eauline leaves opposite or rarely alternate, ovate in out- line, 1-2 em. long, 3-5-pinnate, the lobes ovate, entire, sparingly strigose on both surfaces; flowers usually one from the axils of the upper leaves; calyx 3 mm. long in flower, about 5 mm. long in fruit, divided almost to the base, the lobes linear- lanceolate, long-ciliate, otherwise glabrous or nearly so; re- flexed appendages in the sinuses of the calyx lobes 0.5-1.5 mm. long, ciliate ; corolla campanulate or apparently tubular, about 2 mm. long, shorter than the calyx-lobes, destitute of any ap- pendages within, lobes ovate, about half of the length of the corolla; stamens attached near the base of the corolla by very slender filaments, barely reaching the orifice of the corolla; style 0.5-0.75 mm. long, enlarged and lobed at the apex but not divided ; ovules two on each fleshy placenta, only one maturing ; mature capsule round, 3-4 mm. in diameter, sparingly pubes- cent; seed round, 2.5-3 mm. in diameter, roughened or scarred at the apex, otherwise smooth, dull brick-red, solitary, filling the capsule. Wyomine: rocky flats under Pinus contorta, Double Dia- mond Ranch, Grand Teton National Park, June 3, 1935, Wil- liams 2172, TYPE; rocky open flats near Sensenbach's ranch, Grand Teton National Park, June 8, 1933, Williams 1094; Jack- son's Hole, on Snake River, June 12, 1860, Hayden; marly soil, Jackson's Hole, June 14, 1860, Hayden; gravelly soil, Jack- son's Hole, June 12, 1860, Hayden; rich marly hills, Jackson's Hole, in the valley of the Snake River, Hayden. All specimens cited in Herb. Mo. Bot. Gard. This species seems to have its nearest allies in N. parviflora Dougl. and the closely related entities of that species as treated by Brand in 'Pflanzenreich, Heft 59, IV. 251, pp. 54-55. 1913. It is quite common on the sagebrush flats in the region cited. The other species of the genus, N. breviflora Gray, which occurs in the mountains and in the same region as N. petrophila but along the moist swales and ereek banks in the shade is quite distinet. Specimens of the proposed species were first collected by Hayden 76 years ago, but it seems not to have been found again until recently. 1936] WILLIAMS—FIELD AND HERBARIUM STUDIES, IV 455 t aridus Rydb., Mem. N. Y. Bot. Gard. 1: 348. 1900. Apparently the first known collections of this species for Wyoming were made by the author during the summer of 1935. Dr. F. W. Pennell, in his treatment of the genus (Contr. U. S. Nat. Herb. 20: 313-381. 1920), which covers Wyoming, did not include it. Rydberg in ‘Fl. Ry. Mts. and Adj. Plains’ gives the range as ‘‘Montana.’’ The plant is quite abundant in the Big Horn Mountains. The two following collections are to be referred here: dry hillsides, lower Ten Sleep Canyon, Wa- shakie Co., July 3, 1935, Williams 2321; dry western slopes of the Big Horn Mountains, ten miles east of Kane, Big Horn Co., July 5, 1935, Williams 2348. Penstemon Caryi Pennell, Contr. U. S. Nat. Herb. 20: 354. 1920. i Excellent specimens of this rare species were secured in the Big Horn Mountains, where the type was collected. Dry west- ern slopes of the Big Horn Mountains, ten miles east of Kane, Big Horn Co., July 5, 1935, Williams 2349. Pedicularis cystopteridifolia Rydb., Mem. N. Y. Bot. Gard. 1: 365. 1900. This rare and seldom collected species was found to be quite abundant in the Big Horn Mountains of Wyoming at elevations of 8000 feet and above, along the road between Dayton and Kane. In the field it shows a striking contrast to its near allies, P. scopulorum Gray, and P. Halla Rydb. Number 2355 of my collections represents it. Downingia brachyantha (Rydb.) Nels. & Macbr., Bot. Gaz. 55: 382. 1913. Moist clay ditch banks, Evanston, Uinta Co., Wyoming, June 21, 1984, Harrison & Larsen 7933. This seems to represent the first reported collection of this genus for Wyoming. Microseris nigrescens Henderson, Bull. Torr. Bot. Club 27: 348. 1900. Dr. S. F. Blake, who has kindly determined my Compositae (exeept Senecio) collected during the past summer, informs me that this has been rarely collected. It is not uncommon in [Vor. 23, 1936} 456 ANNALS OF THE MISSOURI BOTANICAL GARDEN moist meadows in the Big Horn Mountains and is represented by my number 2339, collected in moist meadows near Powder River Pass, Johnson Co., elev. 9000 ft., July 4, 1935. Senecio Harbourii Rydb., Bull. Torr. Bot. Club 33: 158. 1906. Dry western slopes of the Big Horn Mountains, 10-15 miles east of Kane, Big Horn County, elev. 8000 ft., July 5, 1935, Williams 2351. Dr. J. M. Greenman, who has kindly determined my Senecios, tells me that this seems to be the first recognized collection of this species in Wyoming. It extends the range of the species some 300 miles northward. Senecio spartioides Torr. & Gray, var. Fremontii (Torr. & Gray) Greenman, comb. nov. Senecio filifolius Nutt. B Fremontii Torr. € Gray, Fl. N. Am. 2: 444. 1843. Sandy hills near Hat Creek, Niobrara Co., Wyoming, Wil- liams s.n. This interesting variety of S. spartioides, although previ- ously eollected in Wyoming, seems not to have been reported hitherto for the state. Specimens of all collections mentioned in this paper are to be found in the Herbarium of the Missouri Botanical Garden unless otherwise noted. Sets of my own collections have been or will be distributed to several American and European herbaria. THE SPECIES PROBLEM IN IRIS EDGAR ANDERSON Geneticist to the Missouri Botanical Garden Professor of Botany in the Henry Shaw School of Botany of Washington University I. INTRODUCTION As a biological phenomenon the species problem is worthy of serious study as an end in itself, and not as a mere corollary to work in some other field. It is, to be sure, a problem so funda- mentally important that it touches many such fields. Workers in any one of these are humanly prone to regard the evidence from that field as all important and its techniques as all suffi- cient (particularly if they are themselves unacquainted with other aspects of the problem). When, however, one takes up the problem, as a problem, and studies it from the diverse view- points of genetics, taxonomy, cytology, and biometry, he real- izes that he not only needs most of the existing techniques but that he must devise new ones as well. Tris versicolor and Iris virginica were chosen for such a study since they customarily grow in colonies containing many individual plants; a peculiarity which facilitates the location and study of large numbers of individuals. A preliminary analysis of the problem (’28) and a discussion of certain points connected with the distribution of these species (’33) have already appeared. The following series of papers constitutes a final comprehensive report. The central core of information is an analysis of a precise morphological census of the two species (section IV). For the interpretation of this morpho- logical data it has been necessary to undertake correlated investigations in cytology, taxonomy, glacial geology, and ge- netics. A technical taxonomic treatment of these irises, to- gether with the related Iris setosa, is assembled in section II, although material indirectly of taxonomic interest will be found in sections III and V. The phylogenetic relationship of Iris versicolor to Iris virginica has proved to be somewhat ex- ANN. Mo. Bor. GARD., Vor. 23, 1936. (457) [Vor. 23 458 ANNALS OF THE MISSOURI BOTANICAL GARDEN ceptional though it is by no means unique among the higher plants. The case has been presented in detail in section III as an example of reticulate relationship. Finally in section V the general problem of evolution in the genus Iris is discussed in the light of all the above information. The major portion of these investigations has been carried out at the Missouri Botanieal Garden and at the Arnold Ar- boretum of Harvard University. A fellowship from the Na- tional Research Council enabled me to acquire eytologieal and statistical techniques for continuing the work. During this time I was a guest of the John Innes Horticultural Institution and of the Rothamsted Experiment Station. A two-months’ leave of absence from Harvard University in 1932 made it pos- sible to study with Dr. Sewall Wright at the University of Chi- cago. Dr. Wright, Prof. J. B. S. Haldane, and Dr. R. A. Fisher have greatly furthered the final analysis of the data, though they are in no way responsible for the imperfections of the work or of its presentation. To the above individuals and in- stitutions grateful acknowledgment is made for these excep- tional opportunities. I am indebted to the University of Chicago Press and to Dr. J. Paul Goode for the base maps used in the second paper of this series and to Mr. Fred A. Barkley for figs. 2 and 13. BIBLIOGRAPHY Anderson, Edgar (728). The problem of species in the northern blue flags, Iris versicolor L. and Iris virginica L. Ann. Mo. Bot. Gard. 15: 241-332. ———— — ——, (733). The distribution of Iris versicolor in relation to the post- glacial Ge Lakes. Rhodora 35: 154—160. 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 459 IL Tue Taxonomy OF THE NORTHERN BLUE Fraaos As a matter of convenience the essential facts in regard to the nomenclature, classification, and distribution of Iris versi- color and its relatives are summarized below. Such a segre- gation of the taxonomic aspects of the problem is essentially artificial and is dictated largely by practical considerations. Material of considerable taxonomic significance will be found throughout the other papers of this series. Attention is called in particular to pages 476 to 480, 495 to 496, and 501 to 506. For the loan of material for study the author is indebted to the Curators of the following herbaria: Gray Herbarium, University of Wisconsin, Missouri Botanical Garden, United States National Herbarium, University of Pennsylvania, and the Canadian National Herbarium. KEY TO THE NORTHERN AND SUB-ARCTIC BLUE FLAGS A. Seeds with a conspicuous raphe; petals setose, less than 2 em. long. B. Stem short, usually unbranched; natives of eastern North Americ E, uu I. setosa var. pss BB. Stem various, often branched; natives of Asia and western North meriea. C. Braets me exceeded by the pedicels; stem usually branched; natives OF celitral Alaska.. ::2...2.2.436.29 92^ fore e I. setosa var. interior CC. Bracts eli the s natives of Asia and the northwestern tof North America; do. se 00 eee Eres eve eres es I. setosa AA. Seeds e an inconspicuous Bes or none; petals laminate, more than long B. Seeds D- Geer sometimes with an inconspicuous raphe; surface of icose, regularly pitted; valves of the mature seed capsule eden Ai slightly, if at all; sepals minutely papillate at pi of blade; outermost bracts of the inflorescence darker and somewhat vernicose along their MarginsS........oooooooooooooooooo.. f; Ref BB. Seeds round or D-shaped, without a raphe; surface of seed not verni- cose, pitting irregular; valves of the mature seed capsule strongly reflexed; sepals macroscopically pubescent at base of d outer- most bracts of the inflorescence with undifferentiated margin C. Seed capsules globose or subglobose; natives of the yes sea- T Rnard Do ae e abn oe ea ele RS ess . virginica CC. Seed capsules at least twice as long as broad; natives of the Missis- MONA TT ET I. virginica var. Shrevei Issued August 10, 1936. [VoL. 23 460 ANNALS OF THE MISSOURI BOTANICAL GARDEN [nis sETOSA Iris setosa Pall. ex Link in Spreng., Sehrad. u. Link, Jahrb. d. Gewachskunde 1*: 71. 1820. Iris arctica Eastwood in Bot. Gaz. 33: 132, fig. 2. 1902. From coastal Alaska to the Lower Lena River and south- wards to Japan. Perennial from a superficial or underground rhizome; rhizome stout, thickly clothed with the fibrous remains of old leaves; leaves narrowly ensiform to linear, 9-65 em. long, 0.5— \ > Eë / L AAN NPY / / "P, 4 \ \ 7 / / ^ d j d 4 / x \ E? < : ) > ANAI N S / , D i : P v J D \ l x vA H ~ Le E EE, LTA ET Ds dä, . pom y y \ . N J DA f 2 T" / Ñ E [E EN Ay ae / : A \ y] | \ c L Map. 1. Range of Iris setosa (open circles), I. setosa var. canadensis (small solid circles), and J. setosa var. interior (large solid circles). Cross hatching shows extent of maximum Pleistocene glaciatio 1.7 em. wide; stem slender to stout, 1-5 dm. high, unbranched or with one or two secondary branches, the latter not exceeding the main axis; upper cauline leaves seldom equalling the in- florescence; inflorescence a compact, 1-3-, mostly 2-, flowered fascicle ; bracts of the inflorescence foliaceous to scarious, 3.5-8 cm. long; pedicels slender, exceeding the bracts or exceeded by them; sepals 4-6 cm. long; haft broad, the margin undulate; ade 3-5 em. wide, glabrous even at the base, dark blue-violet (occasionally wine-colored) with dark veins on a lighter ground-color ; petals small, setose; ovary short, 1-2 em. long in 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 461 the flower, conspicuously three-angled, inflated at anthesis; capsule short-cylindric to ovate, symmetrical, highly vernicose within, often persisting on the plant for one or two years; seeds small, D-shaped, with a conspicuous raphe, highly vernicose over minute regular pitting; chromosomes 38 (2n). [RIS SETOSA VAR. CANADENSIS Iris setosa Pall. var. canadensis Foster in Rhodora 5: 158. V j —9 Fig. l. Diagrams to seale of (a) Iris setosa from the Aleutian Islands and Alaskan peninsula; (b) I. setosa from northern coastal Alaska; (e) I. setosa from southern coastal Alaska; (d) I. setosa var Wei? nët var. canadensis. The diagrams are drawn to scale 1/10) fro ments and enumerations of the herbarium material. Each ger Google average numbers and sizes for all the available material. Iris Hookeri Penny in Steud. Nomencl. ed. 2, 1: 822. 1840. From Labrador, around the Gulf of St. Lawrence; up the St. Lawrence River to Riviere du Loup, Quebec, and along the coast to Washington Co., Maine. This typical preglacial relict differs from the type only in its generally smaller size and lesser variability. As was pointed out by Dykes,' the smallest of the Alaskan specimens are in- distinguishable morphologically from I. setosa var. canadensis. 1 Dykes, W. R., The genus Iris. p. 94. 1913. [VoL. 23 462 ANNALS OF THE MISSOURI BOTANICAL GARDEN The averages of the plants of the two regions are strikingly dif- ferent, however, as can be seen from fig. 1. Its greater con- servatism (i.e. its lesser variability) is a general characteristic of the relict species and varieties around the Gulf of the St. Lawrence, as has been pointed out by Fernald.? This point is discussed at greater length in another section of the paper (see below, pp. 495—496). [RIS SETOSA VAR. INTERIOR Iris setosa Pall. var. interior, var. nov. Ab specie bracteis scariaceis vel crasse chartaceis non foli- aceis rubicundiusculis saepe minoribus quam pedicellis differt. Bracts scarious to thickly chartaceous, not foliaceous, some- what florid, often exceeded by the pedicels. Upper Yukon valley of Alaska, merging into the type in the lower valley and along the western coast. ALASKA: Fort Gibbon, frequent E the Yukon eg Tanana valleys in lakes and along small diag July 4, 1905, Heideman 62 (US TYPE) ; same Mam Aug. 10, 1905, Heideman 98 (US) ; wall: drained gully, See July 4 Har rington 37 (US); Rampart, July 24, 1901 [fruit], Jones 63 (US); alt. 150 m., vicinity of Fairbanks, Aug. 31, 1928 [fruit], Mexia 2302 (MBG); Fairbanks, July 25, 1931, Anderson 1221 (US); Fairbanks, June, 1927, Palmer 1783 (US) The characters which distinguish Iris setosa var. interior from the type have been found to characterize all the available herbarium material from interior Alaska. "Transitional forms are to be found in the region where this great interior valley meets the coast. The following specimens represent such transitional forms: ALASKA: Ft. St. Michaels, Norton Sound, 1865-66, Bannister s.n. (US); moist grassy places, shade of alders, 16 miles west of Nome City, Aug. 5, 1900 [fruit], Flett 1560 (US); on the Yukon River, between Andreafski and Anvik, July 16—18, 1889, Russell s.n. (US). According to glacial geologists,’ this large region remained unglaciated during the Pleistocene, and there, if anywhere, we ? Fernald, M. L. Persistence of plants in unglaciated areas of boreal America. Mem. Am. Acad. Arts & Sei. 15: 244. 1925. * Capps, S. R. Glaciation in Alaska. U. S. Dept. Inter., Geol. Surv. Prof. Paper 170—A. 1931. 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 463 might hope to find living irises most similar to those Iris setosae which must in preglacial times have extended across northern North America. A number of facts have been found which support this hypothesis and they are discussed below (p. 480). While the irises of coastal Alaska are probably not varietally distinet from the type* (which is from Asia) there are minor geographical differences to be noted, when one compiles careful averages for such regions as the Arctic coast, the Alaskan peninsula and Aleutian Islands, and the southern Alaskan coast. Such averages have been prepared from all the avail- able herbarium material and the results are presented graphi- cally, to scale, in fig. 1, along with similar averages for Iris setosa var. canadensis and Iris setosa var. interior. TRIS VERSICOLOR Iris versicolor L. Sp. Pl. ed. 1, 39. 1753. Perennial from a superficial or underground rhizome; rhi- zome stout, clothed with the fibrous remains of old leaves; leaves narrowly ensiform, 1-8 dm. long, 1-3 cm. wide, green to grayish-green ; stem stout to slender, 2-6 dm. high, with one or two secondary branches, the latter seldom equalling the main axis; upper cauline leaves seldom equalling the inflorescence; inflorescence a compact, 2-4-flowered fascicle; bracts of the inflorescence thickly chartaceous to scarious, 3-6 cm. long, the margins so heavily vernicose as to be much darker im color; pedicels slender, some of those in each fascicle usually longer than the subtending bracts; sepals 4—7 cm. long, mostly 1.4 times the length of the petals in living material; blade 2-4 em. wide, variable in color in different plants, mostly violet-blue to blue-violet, the veins slightly darker than the ground-color, minutely papillate at the base, forming at most a dull greenish- yellow spot in living material; petals 2-5 em. long, 0.5-2 cm. wide; ovary 1-2 cm. long in the flower, obscurely three-sided, slightly inflated at anthesis; capsule short-cylindric, mostly symmetrical, somewhat verrucose without, delicately vernicose *Hultén, Erie. Flora of Kamtchatka and the adjacent islands. Kungl. Svenska Vetenskapsakad. Handl. III. 5': 255-256. 1927. (Vout. 23 464 ANNALS OF THE MISSOURI BOTANICAL GARDEN within, usually persisting into the first winter; seeds D-shaped, often showing a poorly developed raphe, surface regularly pitted, vernicose; chromosomes 106—108 (2n). From Labrador to Winnipeg and southward to central Wis- consin, northeastern Ohio, and northern Virginia. Morphologically, Iris versicolor is much closer to Iris vir- ginica than to Iris setosa, though in every character by which it differs from Iris virginica it departs in the direction of Iris law | Map 2. Range of Iris versicolor. setosa. This peculiar intermediacy is discussed at length in the following section of this paper (pp. 478-480). TRIS VIRGINICA Iris virginica L. Sp. Pl. ed. 1, 39. 1753. Iris carolina Radius, Naturforsch. Ges. Leipzig Schrift. 1: 158. pl.3. 1822. Iris carolimiana Wats. in Gray's Manual, ed. 6, 514. 1890. Iris georgiana Britton, in Britton € Brown, Ilust. Fl., ed. 2, 1: 537, pl. 1330. 1913. Perennial from a superficial or underground rhizome ; rhi- zome stout, clothed with the fibrous remains of old leaves; leaves ensiform, 2-9 dm. long, 1-6 cm. wide, green; stem stout, coarse, 3-10 dm. high, mostly with one secondary branch, the latter usually subequal to the main axis; upper cauline leaves usually exceeding the inflorescence; inflorescence a compact, 1-4-flowered fascicle; bracts of the inflorescence coarsely and 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 465 thickly chartaceous, 4-8 em. long, margins undifferentiated ; pedicels stout, shorter than the bracts; sepals 4-8 cm. long, mostly 1.2 times the length of the petals in living material; haft narrow with a straight margin; blade 1.5-4 cm. wide, blue to Ek; res Map 3. Range of Iris virginica (large cireles), and of J. virginica var. Shrevei (small circles). violet-blue and violet, veins scarcely darker than the ground- color, thick pubescence of fine hairs at base, forming a bright yellow signal patch in living specimens; petals 3-7 cm. long, 1-3 em. wide; ovary 2-4 cm. long in the flower, terete or ob- scurely three-sided, not inflated; capsule spherical to long- [Vor. 23 466 ANNALS OF THE MISSOURI BOTANICAL GARDEN eylindrie, mostly asymmetrical, suberose and coarsely verru- cose without, never vernicose within, seldom persisting until the seeds are fully ripe, valves of the capsule reflexed in well- ripened specimens; seeds round or D-shaped, without a trace of a raphe, surface suberose, irregularly pitted; chromosomes 70-72 (2n). From Virginia southward along the Atlantic coast. Although well provided with distinguishing characteristics, Iris versicolor and Iris virginica seem to be under a special curse so far as their recognition in the herbarium is concerned. The shapes of the petals and sepals, the interior glandulosity of the calyx-tube and its shape—any one of these characters is sufficient for accurate specific delimitation. Unfortunately, Tris virginica differs also in texture and substance; its flowers, though larger, have less permanency. As a result they wilt very rapidly; even though carefully pressed when fresh, they have so little substance that the resulting specimens are too thin and fragile to be preserved intact. While the pressed flowers of Iris versicolor are none too accurate in their reflec- tion of the original condition of the perianth, they are far superior to those of Iris virginica. The latter are so badly pre- served that it is almost impossible to use them, even by boiling them up. Perianth dimensions from herbarium material are completely unreliable in these species, and for that reason have been largely omitted from the keys and descriptions. The second most useful set of characters are those provided by the seed capsule and the seed. Here again the characteristic lack of permanency in the capsular walls of Iris virginica has been a great hindrance. With the exception of special collec- tions made by the writer and by other recent students of Ameri- can irises, herbarium specimens of Iris virginica seed capsules simply do not exist. The reason is not far to seek. The coarse stems of this species are neither durable nor stout, and they are not held above the leaves as in Iris versicolor. Iris vir- ginica furthermore prefers slightly damper situations. In nature, therefore, the less lignified stems and seed-pods of Iris virginica, choked by iris leaves and other rank swamp vegeta- 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 467 tion, begin to rot long before the seeds are mature, particularly in the southern part of its range. By September it is not at all unusual to find the black, deliquescent, half-rotten seed-pods lying flat upon the ground, their corky brown seeds spilling out from the capsular remnants. A special collection of seeds and seed-pods has accordingly been brought together in the herba- rium of the Missouri Botanical Garden as a permanent record, and the writer will be grateful for further material of either seeds or seed capsules, no matter how unattractive the partly decayed state of the latter. It is unfortunate that mature capsular material is so difficult to obtain since it displays a curious and striking character, to which Small and Alexander* have called attention. The valves of the capsule in Iris virginica are strongly reflexed as in the related European species, Iris pseudacorus L. Fortunately other characters can be found. Of these the most generally useful in the herbarium are the bracts of the inflorescence (the spathe-valves). Those of Iris virginica have the texture of coarse paper or thin cardboard. They may or may not be streaked with the fine chestnut lines caused by resinous deposits, but if so the streaking will be uniform throughout the bracts. In Iris versicolor, however, the bracts not only are of a finer, yet more durable texture, but the lignifi- . eation is intensified towards the edge so that the margins are often deep chestnut and are distinetly vernicose. In well-preserved specimens the pubescence at the base of the blade of the sepal is a useful charaeter. In specimens of Iris versicolor it appears under the hand-lens as a minutely papillate area. In Iris virginica the hairs are larger, more overlapping, and are often conspicuously straw-colored. In the field Iris virginica is readily identified by the larger, broader petals, the bright yellow pubescent spot on the sepal, and the spongy glandular inner surface of the calyx-tube with its sickish sweet fragrance. Reasons for attaching the Linnean name Iris virginica to * Small, J. K. and Alexander E. J. Botanical interpretation of the Iridaeeous plants of the Gulf States. N. Y. Bot. Gard. Contr. 327: 356. 1931. [Vor. 23 468 ANNALS OF THE MISSOURI BOTANICAL GARDEN this species have been detailed elsewhere? and need not be re- peated here, other than to state that the type is in existence and has been examined. "The suggestion has since been made* that material from the presumable type locality may throw some doubt on this opinion. "Through the kindness of Dr. T. W. Whitaker, I was able to obtain irises from Nesting, Gloucester County, Virginia, which is in the same general vieinity. Of these plants one or two answered the description given above; others showed signs of hybridization with Iris versicolor and were practically sterile. Reference to the distribution maps of Iris versicolor and Iris virginica (maps 2 and 3) will show that very region as the actual boundary zone of the two species. Because of this fact collections from or near the probable type locality are not so definitive as they otherwise might be. Iris virginica is centered upon the Ozark-Appalachian land- mass, an area which has been available for continuous plant oc- cupancy since very ancient times. It would be strange indeed if no geographical differences were to be found within such a species, particularly in the case of the area along the Atlantic seaboard. Such differences are, however, rather difficult to find. At flowering time, I have been able to detect for the plants of the Atlantic seaboard only slightly narrower perianth seg- ments, a larger average flower size (fig. 12), and an inflores- cence which is characteristically somewhat less branched. The seed capsules, however, though variable, are distinct. Well- developed capsules from the upper Mississippi Valley are much longer than broad, while those from the coastal plain are practically spherical, as well as possessing larger, corkier seeds. Unfortunately the difficulty of collecting fruiting speci- mens of iris in the southern swamps (see above, pages 466—467 ) makes exact delimitation of the areas occupied by these cap- sular types a matter of the future. Throughout the upper Mississippi Valley and Great Lakes region it has been a com- paratively simple matter to determine, and I am accordingly * Anderson, Edgar. The problem of species in the northern blue flag, Iris versi- color L. and Iris virginica L. Ann. Mo. Bot. Gard. 15: 241-332. 1928. * Small and Alexander. loc. cit. p. 356. 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 469 using Dr. Small’s name Iris Shrevei in a varietal sense for these elongate-capsuled Iris virginicae. When adequate ma- terial is available for study it will be possible to determine the exact geographical relationships of this and probable other varieties of Iris virginica. Until sucha time it has seemed pru- dent to recognize only this one variety, and to postpone for the present the precise delimitation of the typical and other pos- sible varieties. TRIS VIRGINICA VAR. SHREVEI Iris virginica L. var. Shrevei (Small), comb. nov. Iris Shrevei Small, Addisonia 12: 13-14, pl. 391. 1927. Mississippi Valley and Great Lakes region from southern Minnesota and southern Ontario, southwards to Texas and Alabama. The exact boundaries of its junction with the type as yet unknown and perhaps complex. ANN. Mo. Bot. GARD., VOL. 23, 1936 PLATE 22 Seeds x 10: s, Iris selosa; ve, I. versicolor; vi, I. virginica var, Shrevei ever ANDERSON—THE SPECIES PROBLEM IN IRIS 471 III. Tur PHyLocenetic RELATIONSHIP OF [RIS VERSICOLOR AND [RIS VIRGINICA The northern blue flags, Iris versicolor and Iris virgimca, were originally chosen for study because they were known to be closely related and preliminary investigation had shown that in any one locality they varied markedly from plant to plant. It was accordingly planned to study the minutae of variation so intensively in these two species that one might demonstrate the way in which one species had evolved from the other, or from some common ancestor. It seemed at the beginning of the work that here was splendid material for il- lustrating the way in which individual differences merge into racial, racial into varietal, and varietal into specific. A con- fident beginning was made with this end in view: five years of hard work showed that Iris versicolor might vary greatly and that Iris virginica might vary greatly but that each re- mained itself. They were of different fabrics. One might compare them to two old English villages, one in a sand- stone region and the other in limestone. In each village there would be no two houses alike but all the houses in one village would be made of limestone, all those in the other made of sandstone. The conclusion was reached that closely related though these irises might be, variation within either species was of quite another order of magnitude from the hiatus be- tween them (Anderson, ’28). The variation within could never be compounded into the variation between. The two species were made of two different materials. If one of these species was not derived from the other through the slow accumulation of minor differences, in what other manner could it have originated? Fortunately, at about the same time that these detailed studies of variation came to an impasse there were published a number of accounts dealing with another way in which species might originate in the higher plants: amphidiploidy. This phenomenon may well be illus- trated by the case of Primula kewensis (Newton and Pellew, 29). Primula kewensis originated as a highly sterile hybrid between P. floribunda and P. verticillata (fig. 2). Kept alive by [Vor. 23 472 ANNALS OF THE MISSOURI BOTANICAL GARDEN (FF) (vv) (ff vv) FAB Fig. 2. Primula floribunda (ft), P. verticillata (vv), and their amphidiploid hybrid P. kewensis (ffvv). Drawn from herbarium specimens collected in the greenhouses of the John nnes Horticultural Institution. 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 473 vegetative reproduction, it has on several occasions produced fertile flowers. The progeny from these exceptional flowers have furthermore bred true, or substantially so, and the hy- brids are today grown commercially by means of this fertile strain. Cytological examination has demonstrated that this fertile, true-breeding hybrid has 18 pairs of chromosomes, while the sterile hybrid and each of the parental species has 9 pairs. The fertile hybrid evidently originated when an excep- tional nuclear division in the inflorescence of the sterile hy- brid was not accompanied by a cell division and a sector arose in which the entire chromosome complement had been dupli- cated. On this hypothesis Primula floribunda might be dia- grammed as 9F + 9F; P. verticillata as 9V + 9V ; the sterile hy- brid as 9V + 9F; and the fertile hybrid as 9V + 9V + 9F + 9F. The original hybrid was sterile because the two sets of chromo- somes (V and F) were too unlike to pair and produce fertile gametes. Doubling the number resulted in two sets of V’s and two sets of F’s so that pairing could proceed regularly, pro- ducing a fertile, true-breeding hybrid, or amphidiploid. Amphidiploidy, the production of fertile, true-breeding hy- brids by doubling of the chromosome number, is now known to be a fairly common phenomenon among the higher plants (Winge, ’32). It has occurred under controlled conditions in the experimental plots of many investigators. More than 24 such cases are now on record including several among floristi- cally indigenous species (Müntzing, "20. ’32; Clausen, '83). It has apparently occurred in the development of the cultivated irises (Randolph, ’34). Amphidiploidy is largely confined to the flowering plants and is foremost among several factors which make specific relationships among the higher plants more intricate and more various than they are among the higher animals (Anderson, 731). If our two blue flags did not originate by the slow accumula- tion of individual differences, the most likely explanation of their fundamental divergence is that one or both of them came into existence suddenly through amphidiploidy. A simple hy- pothesis immediately suggested itself. [Vor. 23 474 ANNALS OF THE MISSOURI BOTANICAL GARDEN Iris versicolor is geographically and morphologically inter- mediate between Iris virginica and the Arctic blue flag, Iris setosa. Might it be possible that Iris versicolor is an amphi- diploid hybrid between these two species, a hybrid which oc- curred in pre-glacial or inter-glacial time? Startling as such an hypothesis seemed, it found confirmation in facts from such diverse fields as geographical distribution, cytology, morphol- ogy, and genetics. The hypothesis was used, with complete success, to predict the presence in central Alaska of a pre- viously unrecognized variety of Iris setosa. It orients a num- ber of facts which are either puzzling or meaningless on any other hypothesis. For all practical purposes it may be taken as proved though it is capable of still further tests. The facts which support this hypothesis may be grouped under several different heads: 1. Genetics.—Although they have been placed in different sub-sections of the genus, Iris virginica and Iris setosa are at least partially fertile inter se. It is difficult in this latitude to bring both species into flower at the same time, but on one oc- casion it was possible to do so and two pollinations were made. From these two erosses of Iris virginica x Iris setosa were ob- tained two seed-pods well-filled with seeds, but with shrunken endosperms. None of them germinated, but it seems likely that if the eross could be repeated in quantity a few viable seeds could be obtained. It should be pointed out in passing that the most successful amphidiploids so far obtained have been between plants which are ordinarily quite sterile with one another. As was first pointed out by Darlington (’28, pp. 244—245), the more inter-sterile the two parents of an amphi- diploid, the more fertile and true-breeding is the resulting progeny. 2. Cytology.—The cytological investigation has been ham- pered by the high chromosome numbers of Iris virginica and Iris versicolor, the highest known in the genus. This makes the determination of chromosome number and configuration rather diffieult, and I am happy to report that my own counts have been completely confirmed by several other investigators, 1936] . ANDERSON—THE SPECIES PROBLEM IN IRIS 475 principally by Randolph (’34). Iris virginica has 70 to 72 chromosomes [2n], Iris setosa has 38, and Iris versicolor has just what we would expect if it is an amphidiploid hybrid of the e ag po e ex ee P LI Cp e "A p Fig. 3. Above: smear of early metaphase ect of Iris vir- inica var. Shrevei from Wiere muth, Michigan, somewhat distorted by pressure mera-lucida drawing (made at bench level at x 2280, Med to x 1140). Below: smear of PMC. of Iris virginica var. Shrevei from Frankenmuth, Michigan. Late diakinesis, upper and lower hemi- spheres drawn separately. [Vor. 23 476 ANNALS OF THE MISSOURI BOTANICAL GARDEN two, 108 chromosomes. Iris virginica shows occasional multi- valent association, mainly in fours, and very strong secondary association (fig. 3). These facts would suggest that it is itself an ancient amphidiploid hybrid of two species each with 36+ chromosomes. Iris versicolor has occasional multivalents ; hexavalents such as the one illustrated in fig. 4 are not un- common. The cytological facts therefore are in complete agreement with our hypothesis, and they go even farther by suggesting that Iris virginica is a set of two genoms and Iris versicolor the component of three genoms. > ge Bu e Ta 9 +S d Fig. 4. Aceto-carmine smear of PMC, of Iris versicolor from Connecticut Lakes, N. H. Camera-lucida oti ruin at bench level x 2280, reduced to x 1140). 3. Geographical evidence.—The three species of irises which we are considering have strikingly different distributions in North America, and the distribution of each is characteristic of many of the plants with which it is found. The significance of these areas has been pointed out by Fernald (31). Of the region about which the Iris virginica is centered he says: ““Temperate eastern North America has, then, an extensive area (the southern Appalachian Upland) in which land-plants have had an opportunity to spread since the advent of the Angiosperms.”’ 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 477 The other putative parent species, Iris setosa, also comes from a region (again we quote Fernald) ‘‘which apparently retained [its] present distinctive flora through at least the last glaciation.?' Iris versicolor, which we are presuming to derive from these two ancient species, inhabits a more youthful region, one char- acterized by Fernald (loc. cit., p. 28) as ‘‘the vast region of Canada and the Northern States which has become available for wholesale occupation by plants only since the decay of the Wisconsin ice, within the last few thousand years."' The geographical facts, therefore, point to Iris virginica as an ancient southern species and to Iris setosa and its variety canadensis as being certainly pre-glacial. Iris versicolor, our putative hybrid, is either late pre-glacial or inter-glacial. The present distributions of the species would suggest that the original hybridization (or hybridizations) took place in the interior of the continent, perhaps in the general region of the present-day Great Lakes. As will be shown below there are morphological reasons for believing that the Iris setosa which entered into the cross was not the depauperate remnant which lingered on around the Gulf of St. Lawrence. The actual an- cestor is rather to be sought among the hordes of Iris setosa which must have occupied the interior of the continent before the glacial period. 4. Habitat.—In its habitat preferences Iris versicolor is like- wise intermediate. It grows in situations more moist than those preferred by Iris setosa and a little drier than those in which Iris virginica is found. The three species are not found growing together in nature at the present time but Iris versi- color is found with each of the others. Around the Gulf of St. Lawrence the marked preference of Iris setosa var. canadensis for drier situations has been noted by a number of investi- gators. In Michigan and Ontario, where Iris versicolor and Iris virginica are growing together, it can be seen that Iris versicolor will continue to flower and fruit in spots so dry that Iris virginica only persists vegetatively. It is not uncommon to find Iris virginica growing luxuriantly in marshes where [Vor. 23, 1936] 478 ANNALS OF THE MISSOURI BOTANICAL GARDEN there is standing water over its roots for several months dur- ing the growing season. Iris versicolor may tolerate such a situation but does not welcome it. 5. Morphology.—Before considering these data in detail it may be well to point out again that, according to our hy- pothesis, Iris versicolor was formed by the summation of Iris setosa with 38 chromosomes and Iris virgimica with 70. In other words it received two doses of Iris virginica but only one of Iris setosa. We should expect therefore to find Iris versi- color in an intermediate position morphologically but much closer to Iris virginica than to Iris setosa. Such does actually prove to be the case. e % % Outline drawings of petal and sepal from plants of Jris setosa (left), I. versicolor (center), and I. virginica var. Shrevei (right). In the living plant the size and dimensions of the petals and sepals are among the best diagnostic characters for these three species (as indeed for most species of Iris.) It will be seen that Iris setosa differs from Iris virginica in having a shorter, broader sepal and a much smaller and narrower petal. Iris versicolor, as our hypothesis demands, has differences in this direction. These significant dimensions are presented dia- grammatically in fig. 5. These differences in proportion of sepal and petal are so absolutely in accord with the theoretical demands that, given any two of the three species, it is possible to derive the average proportions of the other by statistical prediction (pl. 23). e sepals of Iris virginica bear a bright yellow pubescent patch, the hairs of which are clearly visible to the naked eye. No such patch exists in Iris setosa, and the epidermal cells are ANN. Mo. Bot, GARD., VOL. 23, 1936 PLATE 23 Photograph of three-dimensional model showing the precise geometrical relation- ship in mee pu sepal size and proportions of Iris virginica (left), I. versicolor enin M) EAT setosa (right). In the dics is ) placed t i of the SEN bety ween the two putative parents, EE their chromosomal contribu- tions to the hybrid are in the oe ratio of 2 to 1. Measurements combined by method illustrated in figure 8. ANDERSON—THE SPECIES PROBLEM IN IRIS 479 seen to be barely papillate when examined with a strong hand- lens. Iris versicolor is intermediate; the bearding is clearly visible only with a hand-lens and the color is at best a kind of greenish-yellow. The blade of the sepal in Iris setosa is broad and slightly undulate, in Iris virginica it is narrow and straight; Iris versicolor presents an intermediate condition. The ovary of Iris setosa inflates rapidly after fertilization so that in the ripening capsule there is a wide space between the walls and the seeds. In Iris virginica the walls are stretched tightly over the seeds, in Iris versicolor the condition is inter- mediate though nearer to that of Iris virginica. In Iris setosa the capsules and stem are so strongly lignified that they sometimes persist for over two years, and it is cus- tomary to find last year's seed stalks among this year's flowers. In Iris virginica the capsule and stem, though much larger, are poorly lignified. In the humid swamps of the south they fall over and the capsule usually is more or less disintegrated by the time the seed is ripe. Here again Iris versicolor is inter- mediate; the eapsules usually persist well into the winter but are seldom found the second season. Iris setosa bears seeds which are unique in the genus. They are small, heavily vernicose, and with a conspicuous raphe down one side. Iris virginica bears large, spongy seeds which may be either round or D-shaped. When the above working hypothesis was first considered, one possible objection seemed to be the fact that Iris versicolor was without a raphe. Subse- quent examination of the seeds of Iris versicolor shows that it does have the shadow of one on nearly every seed and had even been illustrated as having one (though without comment) in Dykes’ plate of Iris seeds in his monograph of the genus (713). (See pl. 22.) When minute comparisons were made in this way, character by character it was found that there were, however, at least three characters in which Iris versicolor was not intermed- iate between Iris virginica and Iris setosa var. canadensis. The theory demanded an Iris setosa with several flowering branches, with pedicels longer than the bracts, and with bracts not greenish but brown and subscarious. Since the demands of [Vor. 23 480 ANNALS OF THE MISSOURI BOTANICAL GARDEN the theory were met so minutely on all the other characters it seemed possible that these three represented details in which the raee of Iris setosa entering into the original hybridization differed from Iris setosa var. canadensis. Such a race might conceivably have present-day relatives living among the Iris setosa which is so widely spread in Alaska and Asia, since the species is notoriously variable there. Herbarium material was accordingly consulted. The first few specimens examined proved most disappointing. They were from localities along the arctic coast and they had none of the three desired qualities. Their bracts were long and green, completely eclipsing the pedicels, and the plants were unbranched. Farther down in the pile, however, was a plant which had not only long pedicels, but scarious bracts and a branched inflorescence, the very com- bination desired. Farther on was another and eight specimens in all were found. When their distribution was plotted it was found that all came from central interior Alaska and represented, in fact, the only specimens from that region. They have accordingly been described above in the taxonomic section of these papers as a new variety, Iris setosa var. interior. Further search unearthed the even more significant fact that Iris setosa var. interior grows in that part of Alaska which was adjacent to the edge of the continental ice-sheet (map 1). If representatives of the pre-glacial races of Iris setosa which must once have covered much of northern Canada are to be sought anywhere today, glacial geologists would suggest this very region (Capps, '31). In other words, we not only found the variety demanded by the theory but we found it in exactly the region which the theory would suggest as most likely. The comparison of the three species can be closed therefore with the presentation of diagrams to scale of Iris virginica, Iris versicolor, and Iris setosa var. interior. It will be seen that in size, node number, leaf length, number of branches, length of pedicels, and length of bracts the demands of the hy- pothesis are exactly met (fig. 6). Though it is capable of still more rigorous tests, the theory that Iris versicolor is a pre-glacial, or inter-glacial, amphi- 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 481 diploid hybrid between Iris virginica and Iris setosa var. in- terior would seem to be well established as a working hy- pothesis. Ultimately it should be possible, as in the ease of the European Galeopsis Tetrahit (Müntzing, "201. to re-synthesize the species from its two constituents. i. SETOSA VAR, INTERIOR I. VERSICOLOR L VIRGINICA VAR. SHRE VEI Fig. 6. Diagrams to scale of Iris setosa var. interior, I. versicolor, and virginica var. Shrevei. The diagrams represent precise Plot of all the available herbarium material. The hypothesis also gives an explanation to several curious facts which had previously been most puzzling. The first might be called the one-way relationship between Iris virginica and Tris versicolor, or so I have attempted to phrase an impression received from long-continued study of variation within these twospecies. It seemed that the relation of Iris versicolor to Iris virginica was quite different from that of Iris virginica to Iris versicolor, or to state it somewhat less mystically, that Iris versicolor often reminded one of Iris virginica, but Iris vir- [VoL. 23 482 ANNALS OF THE MISSOURI BOTANICAL GARDEN ginica never reminded one of Iris versicolor. If on the above hypothesis Jris versicolor is indeed Iris virginica plus some- thing else, then the relationship should be different in one direc- tion from what it is in the other. Another puzzling fact had been the frequency of albinos. Iris virginica and Iris versicolor by any ordinary standard were unusually variable species yet pure albinos were exceed- ingly rare. In spite of prolonged search and inquiry I found only three in Iris virginica while in Iris versicolor I have found no pure albino without a trace of blue, and only one case has been reported in the literature (Fernald, '36). Yet albinos are common in many species of Iris, as, for instance in [ris mis- souriensis; why then should they be absent from our common blue flags? Why should they shun this particular species which by any other standard is peculiarly variable in flower color? On the above hypothesis this is exactly what one might predict. If Iris virginica is made up by the summation of two ancient species, albinism, being recessive, cannot appear until it has occurred in each of the constituent sets. In Iris versicolor it cannot show itself until it appears in these and also in the set of chromosomes derived from Iris setosa. "This means that if the original frequency of albinism in the basic species had been, say one in every 5000, that we should find it 1n Iris vir- ginica once in every 25,000,000 and in Iris versicolor only once in 125,000,000,000. The infrequency of albinism in Iris vir- ginica and its even greater rarity in Iris versicolor is therefore in strict accord with theoretical expectations. SUMMARY 1. The absolute morphological discontinuity previously dis- covered between the closely related Iris versicolor and Iris vir- ginica is explained by the following hypothesis: Iris versicolor originated suddenly as a fertile, true-breeding hybrid (an amphidiploid) between the southern Iris virginica and the sub- arctic Iris setosa, in pre-glacial or inter-glacial times. 2. Since the former has 70 chromosomes and the latter 38 we should expect to find Iris versicolor generally intermediate be- 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 483 tween these species though much closer to Iris virginica. This is found to be the case. 3. A consideration of one or two minor exceptions to this generalization led to the discovery of a previously unrecog- nized variety from central Alaska, Iris setosa var. interior. 4. Since it harmonizes so many otherwise incoherent facts from cytology, morphology, geographical distribution, ge- netics, and geology, the theory is taken to be well established asa HUE hypothesis. 9. The theory also explains two phenomena which had pre- viously seemed incomprehensible: (1) the ‘‘one-way’’ mor- phological relationship between Iris virginica and Iris versi- color, (2) the infrequency of albinos in Iris virgi nica and their even greater rarity in Iris versicolor, a species otherwise un- usually variable in flower color. Itis shown that both of these results are to be expected on the basis of the above hypothesis. BIBLIOGRAPHY Anderson, Edgar (’28). The problem of species in the northern blue flags, Iris versicolor L. and Iris virginica L. Ann. Mo. Bot. Gard. 15: 241-332. (’31). Internal factors affecting discontinuity between species. d Am. Nat. 65: 144—148. Capps, S. R. (731). Glaciation in Alaska. U. S. Dept. Inter. Geol. Surv. Prof. Paper 170-A. DI J. Eo Cytological evidence for the hybrid origin of Pentstemon us Keck. Nk : 65-76. Dartington C. D. (’28). Studies in Prunus, I and II. Jour. Gen. 19: 213-256. pl.1 Dykes, vi E) (713). The genus Iris. Fernald, M. L. (31). Specific segregations and identities in some floras of eastern North America and the Old World. Rhodo 3 63. * ei Albino Iris versicolor. Dia. 38: 52. Müntzing, A. (730). oo to a genetic monograph of the genus Galeopsis. Hereditas m 185-3 —, (33). E genetic investigations on synthetic Caleopsis Tetrahit. Ibid. 16: 105-154. pr W. C. F., and a Pellew (’29). Primula kewensis and its derivatives. r. Gen, 20: 405-4 eng L. (788) E Bos numbers in native ED and introduced species and cultivated varieties of Iris. Am. Iris Soc. Bull. 52: 61-66. Winge, 6. (732). On the origin of constant species-hybrids. Se Bot. Tidskr. 62: 107-122. ANDERSON—THE SPECIES PROBLEM IN IRIS 485 IV. [NTRA-SPECIFIC DIFFERENTIATION IN THE NORTHERN Brvr FLAGS As has been related in the previous paper of this series, de- tailed studies of variation were made in two species of Iris with the aim of demonstrating the way in which one had arisen from the other. Though they failed to produce any such evi- dence (and have as a matter of fact led to a very different hypothesis of the phylogenetic relationship between these two partieular species), the data may still be used to examine the way in whieh evolution has proceeded and is proceeding to build up differences within these species. Many of the actual data have been presented in full in a previous communication (Anderson, '28). Since that time, however, the interpretation of this evidence has been profoundly affected by information derived from other fields of study. The following paper is largely concerned with the results of a detailed morphological census of two species of Iris in eastern North America, I. versicolor and I. virginica; some attention has also been paid to the related glacial relict, Jris setosa var. canadensis. The results of such a census may be presented individual by individual or they may be grouped and averaged in various ways. In the following census the colony is recog- nized as a vegetational and evolutionary unit of major im- portance, so far as irises are concerned. Throughout most of the region in which they are found today, I. versicolor and I. virginica grow in small colonies of from one to several thousand individuals. Single individuals usually cover several square feet and send up several flowering stalks each year. In exceptional cases one individual may by vegetative reproduc- tion cover a much larger area, and in rare instances a colony of several acres may be composed genetically of but one plant. With a little study the recognition of individual plants is not at all difficult. The sea of blue-purple flowers which at first glance seems so uniform resolves itself into a little community with quite as much divergence between the various members as is found in human communities. One plant will have brown spots on the sepals of each flower, the next one will be without [Vor. 23, 1936] 486 ANNALS OF THE MISSOURI BOTANICAL GARDEN the spots but will claim attention by the extraordinary size of its petals. Another will have flowers which are a very light blue, still another will have flowers which are almost wine colored, another will have deep notches in the petals. And just as in a village each man has a hand with characteristically different proportions from all other men yet has his left and right hands built on almost the same pattern, so it is in a swamp full of irises. The petals and sepals of the different flowers on a plant will have substantially the same proportions, but these proportions will vary tremendously from plant to plant. This point is illustrated in pl. 24 where three flowers are shown from each of six plants. These photographs were taken with identical illumination and exposure and were developed and printed uniformly. The differences in shade are due to dif- ferences in the flowers themselves; plant no. 2 had flowers of a very light blue and they have photographed almost white; plant no 6 had a great deal of red in with the blue and it is much darker in the picture. It will be noticed that even such a tenuous character as the carriage of the petals and sepals is based upon inherent factors; note, for instance, the floppy aspect of all three flowers of no 5, the contrasted horizontal sepals and upright petals of no. 1, the undulate sepal margins of no. 3. The number of such colonies is enormous; a rough approxi- mation has been made by observing the numbers to be seen from the highway along various routes and converting these figures into number of colonies per square mile. The method seems to be reasonably accurate, since it yields consistent re- sults when different trips are made through the same territory. According to this method, the average frequency of Iris col- onies per 100 square miles is 120 in northern Michigan, 350 in southern Michigan, 170 in northern Illinois, 30 in southern Mis- souri, and 5 in Alabama and Mississippi. Colonies are par- ticularly frequent north of the terminal moraine where an un- even glacial topography produces many small swampy areas favorable for the growth of Iris. Figure 7 illustrates a rep- resentative area of 50 square miles within this region. It 2393. T, suy Jo Sjue[d xis ot YoRa S419MOp IBA DOMB.NA 1240 UU NM WO] "IINOSSIJA XNOIS sep 9:38].10J ANN. Mo, 30T. GARD., VOL. 2; PLATE 24 ANDERSON—THE SPECIES PROBLEM IN IRIS 487 EJ] ` d, y 3 = 7 3 | - si D 2 Le > op j D S E i 7; ig E ¡LAR PZE Tete S E L E? A o h JA ee `Z e) SE KI (ES || 8 M 880) + Wow vog Y o dpa CENTRAAL | eh f D Ce i eo SS a qj f, :) T : Representative region of 50 square ie in south- ern Michigan. SE vc da by Iris virginica var. Shrevet shown in solid blae [VoL. 23 488 ANNALS OF THE MISSOURI BOTANICAL GARDEN demonstrates how numerous are the Iris colonies and how isolated they are from one another. It is a point of some theoretical importance (see below, pp. 495—496) that the colonies are probably distributed in much the same way that they were before the land was cleared but that Fig. 8. Diagram illustrating how petal length and width, and bim length and width are combined to form an ideograph. Above, Iris virginica ; below, 1. versicolor 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 489 the numbers of individuals in the colonies are greater. The deforestation and pasturing of swampy areas have increased many fold the area available for these irises throughout the northern states. Turning the land over to pasture is partic- ularly helpful to them since grass, their worst competitor, is kept down by the livestock. In most cases large colonies now numbering thousands of plants are probably the descendants of a much smaller number which were growing in that area be- fore the land was cleared. The census has to do with four measurements; length and width of sepal, length and width of petal. Since such measure- ments are of greatest significance when their interrelations with each other are understood, the results are presented graphically in a way which makes it possible to convey these relationships simultaneously. Figure 8 shows how the four measurements of each flower can be built up into a simple black- and-white diagram. This diagram or ‘‘ideograph’’ is es- sentially a white petal superposed upon a diag tie black sepal. Figure 9 presents ideographs for 20 plants of Iris versicolor and 20 of Iris setosa var. canadensis which were growing together in a pasture near Île Verte, Quebec (Ander- son, '35). It demonstrates how such ideographs may be used to present a large amount of data in a small space. Figure 9 is a graphieal summary of four measurements and six propor- tions on each of 40 plants. Itis,in other words, a simultaneous presentation of 400 separate facts. The precise comparison of such colonies can be carried farther by the production of average ideographs for the whole colony, utilizing the average petal length, the average petal width, the average sepal length, and average sepal width (fig. 9, central ideographs). The data for colonies are presented in this way in figs. 10 and 11. The colonies are arranged by species and subspecies and within these categories are placed roughly according to geo- graphical position from south to north. A study of these fig- ures yields the following conclusions : 1. There is little or no regional differentiation in shape within any of the subspecies. No general characteristics can be recognized for the irises from southern Michigan or from VOL. 490 ANNALS OF THE MISSOURI BOTANICAL GARDEN NARA ANN nfinnnnn nnnnnnn Fig. Ideographs t E poe of Iris versicolor and 20 of I. setosa var. gen ensis from fle Jine Averages of entire eolony (50 for each species) shown in eeh dun mes 23 1936] 491 ANDERSON—THE SPECIES PROBLEM IN IRIS L261 Hm QOO Ni 9261 "HAN QOO NI? 228 HOIN HALOIANYINY d 4 HOM | 32131^831N32 HIN 1 IVY INOOHIS OIHO 390148 Avg ON ITUASAVIA 928 HOW HLOWNIYN Y Yd MIN vOYVWHY HOIN JINIEMVA “wi NOSIQvIN LC 6261 TU NYITNA b. NOLONING YS Viv HINOYBNIM "SSIN NOSHIWP e26! OW xXNOIS dd YNVISINOT SE HDi 370/840 "SIM ITUAIIOU VA 4261 OW XNOIS Ud ON €3y2!^ 3 VNAJINNOS Wed A311TvA AX NOLNYLS "MOI HOTT PZ HOW 321145110 ginica and I. vir- I: f Iris vir 1es 0 Average ideographs for 31 colon of > > > = = E? SV o EI wei E Ké Pig "es > [VoL. 23 492 ANNALS OF THE MISSOURI BOTANICAL GARDEN the Mississippi Valley, or from the Cumberlands, or from the prairies. As will be shown below, this conclusion is reinforced by the computation of regional averages. 2. There is a very slight trend in size. Iris virginica var. Shrevei reaches its greatest development in the Cumberlands of Kentucky and Tennessee. Northward or southward it be- comes somewhat smaller on the average. Iris versicolor, on the contrary, is largest in the north and becomes smaller ONT ONT TRURO NS r Y 2 I] o 3 9 + n ALBERTON TIMAGAGI MIDOLEBURY VT. DUXBURY MASS CLARENDON VT. o 2 E > z « > [*] z VI "I =| > CONEWANGO N.Y HUBBARDSV. NY LIVERPOOL PA HARMONSBG PA Fig. ll. Average ideographs for 18 colonies of Iris versicolor. towards its southern limits. Transplants from these various areas have kept their same relative sizes when grown together in the experimental garden. 3. Colony averages are fairly consistent from year to year in those cases where measurements could be made in different years. Allied to this fact is the experimental evidence that the flowers of plants collected and grown together in the garden maintained their characteristic size, shape, color, and color pattern. In several cases divisions of the same plant have been grown and studied in Boston, St. Louis, and Schoolcraft, 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 493 Michigan. For these reasons the differences which distinguish the individual plants and thereby create the peculiarities of the colonies are thought to be largely inherent. An even stronger proof is the fact that progeny tests of several indi- viduals produced evidence for the heritability of various in- dividual peculiarities. 4. There are striking differences between colony averages, even for the same region. The colony averages of figs. 10 and ONTARIO H MICHIGAN pag Q < 3 f 5 2 9 = 5 w o CAROLINA Fig. 12. Regional averages for Iris versicolor (above), I. vir- ginica var. Shrevei (below, left) and I. virginica (below, right). 11 are particularly instructive when compared with the re- gional averages of fig. 12. It will be noted that though there are differences between the regions they are slight, and they have no evident geographical trend other than the slight one in size already referred to. The regional differences are in- deed so slight that the variation in these irises might seem to be without any phylogenetic significance. If attention is shifted from the regions to the colonies, the evolutionary significance of the variation is more manifest. Each little colony is a more or less independent evolutionary unit and has evolved a more or less distinctive combination of characters. All that is neces- sary for the production of a regional variety is the isolation of any one of these colonies. Were some succession of droughts and floods to exterminate the great bulk of either species, leav- ing only two or three colonies persisting in different parts of [VoL. 23 494 ANNALS OF THE MISSOURI BOTANICAL GARDEN its present range, and were the area to be repopulated from these centers we should then have the formation of geographi- cal varieties. The majority of the colonies as they exist at the present have achieved sufficient individuality to be rated as incipient varieties were they only to reproduce their several types over larger areas. For an actual difference of this mag- nitude we may compare Iris virginica of the Atlantic seaboard and Iris virginica var. Shrevei of the Mississippi Valley. For an inconceivably long time the irises of the seaboard have been somewhat isolated from their relatives in the Mississippi Val- ley. It is not surprising then that though we can find no out- standing differences in sepal and petal proportion within the interior of the continent, there is a slight difference between the irises from the interior and those from the seaboard. Iris virginica has flowers which are distinctly larger and somewhat narrower. The difference is a minor one as compared to the distinct hiatus between I. versicolor and I. virginica, but it is reinforced when we study such technical characters as the shape of the seed capsules and the size of the seed. Particularly significant is the fact that the difference be- tween I. virginica and I. virginica var. Shrevei is of about the same order of magnitude as the differences between colonies of I. virginica var. Shrevei. It would indeed be possible to find two swamps in the same township in southern Michigan whose iris populations have as great an average difference as that between Iris virginica of the Atlantic Coastal plain and Iris virginica var. Shrevei. But in this latter case the difference, slight though it 1s, characterizes a whole region and has super- imposed upon it the varying pattern of colony differences in each region. An evolutionary factor of basic importance in our common blue flags, therefore, is the rapid accumulation of minor dif- ferences in the little colonies into which the species are divided. Nearly every colony carries within itself the potentialities of a variety or a subspecies. The conditions under which these irises exist seldom release these potentialities. Many colonies arise, develop a distinctive type, and pass on with little or no influence on the main evolutionary stream. It should be re- 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 495 membered that the territory in which they are growing pos- sesses very few geographical or climatic barriers. But the potentiality is there and when opportunity allows a colony to play a larger role it is ready to do so. Then the peculiarities evolved in one colony, or a few colonies, might come to char- acterize all the colonies of a region. By mathematical deduction from the known facts of geneties, Wright (731) has produced a generalized theory of evolution. From the standpoint of pure theory he finds that evolution will proceed most effectively neither in a large inter-breeding popu- lation nor in a very small one but ina large population ‘‘divided and sub-divided into partially isolated local races of small size." Under such conditions he predicts ‘‘a continually shift- ing differentiation among the latter which inevitably brings about an indefinitely continuing, irreversible, adaptive, and much more rapid evolution of the species."' The irises of this study present just such a picture. They are divided into partially isolated small colonies [‘‘ Local races’’] which before the land was cleared were probably even smaller. These colonies differ from one another and from the mean of the species to a degree which is almost of varietal magnitude. Their differences are inherent and to all appearances are largely non-adaptive. Isolation, in dividing the species up into these smaller units, has made possible their several diver- gencies. Were isolation to be made complete, as by another glacial period, the phylogenetic potentialities of the survivors would be released and what had been colonial peculiarities might become varietal differences. The variation within [ris setosa var. canadensis seems par- ticularly instructive in the light of its recent history. It is typieal of those species whose once continuous range aeross northern North America was reduced to the northeastern and northwestern edges of the continent by the Pleistocene ice. In Alaska a large central region was left unglaciated ; around the Gulf of St. Lawrence, on the other hand, the plant refuges in glacial times were little more than rocky nunatacks rising above the ice. The results on the two sets of irises are just what a geneticist might predict. Even from the few specimens [Vor. 23 496 ANNALS OF THE MISSOURI BOTANICAL GARDEN which are available in herbaria one can see that the Iris setosae of Alaska are a varied assemblage. They include one well- marked variety (described above) and several fairly well- marked regional variants. The irises of eastern Canada present a very different picture. Figure 9 gives some slight indication of their lesser variability from plant to plant. Com- pared to the millions of irises which might well have continued to live in Alaska during the ice age, those of the St. Lawrence region were a mere handful. From that handful must have descended the millions upon millions of irises which now carpet the meadows and shores of that region in early summer (An- derson, '35). Compared with our other American blue flags they are a singularly invariable lot. Graphical comparisons are made in fig. 9. They are furthermore the smallest, much smaller than any other recognizable type of Iris setosa (note fig. 1). They are short, seldom branching, with small leaves and few nodes. This conservatism of Iris setosa var. canadensis is distine- tive of most of the glacial endemics (or near endemics) of the region around the Gulf of St. Lawrence. In one of his classic contributions to the subject, Fernald (’29) has aptly char- acterized them as ‘‘already waning types, too old, or too con- servative to spread into closely adjacent and virgin soils.’’ In the case of Iris setosa var. canadensis the invariability can- not be a direct effect of time, for the highly variable irises of Alaska are quite as aged. It is more probably, as Professor Fernald has suggested, an innate conservatism ; a conservatism founded genetically upon the fact that these irises are descend- ants of a small and highly selected stock. Hard times removed from the region all the luxuriant types which may once have existed there. When the ice age was over the immediate area was repeopled from the few plucky survivors. Their descend- ants, Iris setosa var. canadensis, bear the scars of the glacial period, so to speak, in their conservatism; an innate invari- ability which, on the one hand, gives them a greater uniformity, and on the other, prevents their adapting themselves readily to other environments. In the light of its probable history, it is not surprising to find 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 497 no regional differentiation in Iris versicolor. If, as seems probable, it originated in interglacial times, it is a compara- tively young species. It is furthermore inhabiting a region which is extremely youthful floristically, most of it not becom- ing available for plant occupancy until the last retreat of the Pleistocene ice. Having only recently moved into most of the territory it now occupies, Iris versicolor has had as yet little or no opportunity to develop geographical races within the species. Much the same argument can be made for the northern flank of Iris virginica var. Shrevei. It is in this part of its range, thanks to irregular glacial topography, that it occurs most fre- quently and is therefore most easily studied. Within this area there is little or no evidence of geographical races. Were it possible to study Iris virginica and Iris virginica var. Shrevei in the same detailed manner on the older lands they occupy in the south, it is more than probable that considerable geograph- ieal variation would be found. "There are indications of such differentiation in the few samples from these areas which are availablein herbaria. A detailed statistical census would prob- ably reveal still more. SUMMARY 1. Most of the northern blue flags occur in more or less iso- lated colonies of from a few to many thousand individuals. The average frequency of such colonies per 100 square miles was found to vary from 350 in southern Michigan to 5 in Ala- bama and Mississippi. 2. From a statistical study of 60 colonies it is shown that there is little regional differentiation within any of the sub- species. Even within the same region, however, there are pro- nounced differences between the colonies. 3. The colony is an important evolutionary unit in these irises. Through its isolation each colony develops a distinctive type which is of potential phylogenetic importance. 4. It is shown that the differences between geographical varieties in these irises are of about the same order as differ- ences between colonies. Many of the colonies carry the po- [Vor. 23 498 ANNALS OF THE MISSOURI BOTANICAL GARDEN tentialities of a variety or a subspecies, though these are seldom released by the conditions under which they exist. 9. The conservatism of Iris setosa var. canadensis is dis- cussed in the light of its history as a glacial relict. Its com- parative invariability and poor colonizing ability are probably innate as suggested by Fernald. They result genetically from the fact that this subspecies has descended from what in glacial times must have been a small population living under adverse conditions. 6. The lack of geographical differentiation within Iris versi- color and the northern colonies of Iris virginica var. Shrevei is correlated with their occupancy of a floristically youthful ter- ritory. Iris virginica var. Shrevei would probably reveal greater regional differences if it could be studied on floris- tically older lands in the south. BIBLIOGRAPHY Anderson, Edgar (’28). The problem of species in the northern blue flags, Iris versicolor L. and Iris virginica L. Ann. Mo. Bot. Gard. 15: 241—332. , (735). The irises of the Gaspé Peninsula. Am. Iris Soc. Bull. 59: Fernald, M. L. (’29). Some relationships of the floras of the northern hemisphere. Intern. Cong. Pl. Sci. Proc. 2: 1487-1507. Wright, S. (7/31). Evolution in Mendelian populations. Genetics 16: 97-159, 1936] ANDERSON— THE SPECIES PROBLEM IN IRIS 499 V. Tue EVOLUTIONARY PATTERNS OF THE GENUS [RIS Students of the species problem are faced with a curious dilemma. By the nature of that problem they are forced to confine their attention to the details of a few species; yet if they do so exclusively they will be unable to interpret their results in general terms. "The details of evolution vary from genus to genus and even from species to species; when one diseusses the evolutionary patterns discerned by intensive work on one or two species he must attempt to determine to what extent those patterns are general characteristics of most species and to what extent they are special features of those few. Throughout the following discussion an attempt will be made to apply the information derived from these three species of Iris to the entire genus. It is as yet too early to consider the larger problem of speciation in the higher plants in the light of these results, except in the most general way. In addition to the correction factor for the peculiarities of the germ-plasm one must also allow for the peculiarities of the region in which the studies were made. Speciation is a func- tion of the region under observation. If a region is without pronounced barriers, speciation for most of the organisms in that area will be simpler than in an area with a complex system of barriers and partial barriers. It will also be affected by the age of the region. In one floristically young, geographical dif- ferentiation within species will be less intense than in regions which have been continuously available for occupancy for a long period of time. It should therefore be kept in mind dur- ing the following discussion that these studies were, for the most part, carried on in an exceedingly youthful region floris- tically and one in which there are no geographical barriers of any great importance. A number of processes of evolutionary significance are tak- ing place in these irises, all of which affect the ground-plan of the genus. Three of them, hybridization, amphidiploidy, and colonial differentiation, have produced effects which were readily perceptible by the methods used in this study. A gen- eral summary of the results is presented graphically in fig. 13. [Vor. 23, 1936] 500 ANNALS OF THE MISSOURI BOTANICAL GARDEN The fundamental pattern is composed of the three species, which can be represented as great compound cables, stretched from the past into the future. For long eras these cables are praetieally parallel; even the detailed methods of this investi- gation have produced little or no evidence for the ultimate deri- vation of these separate cables from a common source. Examining these cables more closely it is apparent that they are made up of smaller cords, the colonies. These cords like- wise maintain their individuality for considerable time. Such a division of the species into small colonial units is probably characteristic not only of these three species but of a good part of the genus Iris. Certainly the other species with which I am personally acquainted in the field, I. fulva, I. foliosa, I. pris- matica, I. foetidissima, I. pseudacorus, and I. missouriensis, grow in much the same sort of more or less 1solated colonies. Within this small group of irises there are three apparent cases where colonial differences have developed into geograph- ical varieties. Very anciently the Iris virginicae of the At- lantic seaboard became slightly differentiated from those of the Mississippi Valley. As has been shown above, this differ- ence is of about the same order of magnitude as that between colonies and is therefore most easily interpreted as due to some ancient geographical change which reduced one or both of these regions to one or afew colonies. By some such process also Iris setosa var. interior diverged from the Iris setosae of coastal Alaska and of northeastern Asia. More recently, dur- ing glacial times, Iris setosa var. canadensis was reduced to a highly inbred remnant (see pp. 495-496). At the present time, therefore, Iris virginica is composed of two slightly divergent sub-cables, Iris setosa of three, and the youthful Iris versicolor is as yet but a single cable. Sup- posedly by the successive compounding of such divergencies, subspecifie differences might be built up into differences of specific magnitude. There is little evidence in this study to contradict such a theory, but it should be pointed out that there is none to support it. In the opinion of the author the theory that geographical varieties are potential species is a debatable one. SPECIES PROBLEM IN IRIS 501 AN DERSON—THE 1936] 3 l E ava dag ED 3 \ 1DADAYS 4DA vojurbaa Su E q = a ED ——— A HE | ipiis ux gege E ERR PE EZ GE EIS EATEN eee GE e Ss — ram showing phylogenetie relationships in the north- Further discussion in the text. Fig. 13. Diag ern blue flags. ANDERSON—THE SPECIES PROBLEM IN IRIS 503 Observation and experiment have shown (Anderson, ’28) that in Iris versicolor and Iris virginica the colonies them- selves are more or less divided into inbred lines, so that in these species we can speak of the colony ‘‘cords’’ as being made up of pure-line ‘‘threads.’’ Since cross-pollination occurs oc- easionally the threads maintain their identity for only a few generations. This detail of the evolutionary pattern will vary greatly from Iris species to Iris species. There are very prob- ably species in which inbreeding is more severe; there are certainly self-sterile species in which there is complete out- erossing. In the former the threads would be longer, in the latter there would be no recognizable threads at all within the colony cords (Anderson, loc. cit. pp. 308-310). The great ground pattern of the cables is a simple one, with two exceptions. Very occasionally by amphidiploidy a single strand runs out from one trunk line to another and at an inter- mediate point a whole new cable arises. An earlier paper of this series presented detailed evidence for the amphidiploid origin of Iris versicolor. The secondary pairing and multiple association characteristic of the pollen mother-cells of Iris virginica strongly suggest that it is itself the result of an ancient amphidiploid hybridization, perhaps between a species somewhat like Iris tripetala and one related to Iris hexagona. Such occasional interweaving of phylogenetic lines is appar- ently characteristic of most of the genus Iris. Simonet (734) reports chromosome numbers for the section Pogoniris which suggest amphidiploidie relationships and has presented cyto- logical proof (735) for the occurrence of amphidiploidy in cultivated irises. As Randolph has shown (loc. cit., p. 65), there is experimental evidence that in Iris, species with differ- ing ehromosome numbers eross more readily than in many other genera. He also presents some evidence for the func- tioning of unreduced gametes in Iris. Both of these conditions favor amphidiploidy. The presence of even occasional amphi- diploidy within a genus will so complieate the phylogenetie re- lationships that it will be impossible to divide and subdivide it naturally into sections and subsections. We would therefore predict for the genus Iris that while there might be evidence of [Vor. 23 504 ANNALS OF THE MISSOURI BOTANICAL GARDEN groups of related species, it would be impossible to arrange all these groups in a clear-cut natural system. Such is actually the case. The latest monographer of the genus has the following to say about the difficulties of subdividing the section Apogon, numbering some 50-100 species and native to Europe, Asia, and North America: ‘‘It seems unfortunately impossible to select any one character or set of characters to form a guide through the maze of species. . . . The classification given be- low is therefore admittedly unsatisfactory partly because some of the species seem to stand by themselves and to have little or no affinity to any others?" [Dykes, '13]. He then proceeds to divide the section into fifteen sub-groups. On the theory out- lined above Iris versicolor is an amphidiploid hybrid between Iris virginica of his group X and Iris setosa of group XIV. A few relationships of this sort would produce exactly the dif- ficulties which Dykes describes. The other tangle in the evolutionary pattern of these irises is provided by hybridization. Though such tangles are a char- acteristic feature of the evolutionary pattern in the genus Iris, they form in this group of species a small knot of minor conse- quence. Iris virginica and Iris setosa are now geographically isolated, though they still in part occupy the same river system. Iris setosa var. canadensis and Iris versicolor grow together throughout the range of the former, but there is very great numerical isolation between them (38 vs. 108 chromosomes). Hybrids are occasionally found, but they are very rare. The only hybridization of any consequence within the group at the present time is between Iris versicolor and Iris virginica. Even in this case, there is geographical isolation between the bulk of the two species. Along the eastern seaboard, where the two species have apparently been longest in contact, they are both quite rare. It is only around the Great Lakes, where they are both exceedingly common, that hybridization is at all com- mon. Even here there is partial isolation for the blooming periods of the two species barely overlap. At the northern end of the southern peninsula of Michigan, the two species have evi- dently been closely associated throughout much of post-glacial time (Anderson '33) and their areas of contact, due to the 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 21 23 25 27 29 31 33 35 37 39 41 43 45 47 Fig. 1 Frequency distribution for the index, sepal — length + ET eppidum I width, ed 27 SE e— from the foll ing eray es: Pardee , Wisconsin (Iris a var. Shr “a Billings Bridge: Ot xiii "Canada (1. versioolor) ; “Bagadine, Mich. igan yaa St. Ignace, Michigan (hybrid 505 [Vor. 23 506 ANNALS OF THE MISSOURI BOTANICAL GARDEN peculiarities of the various post-glacial lakes, are extremely involved. Even in this area the great majority of colonies are composed exclusively of one species or the other. On the north shore of Lake Michigan, in an area with a sim- ilar post-glacial history, two hybrid colonies were studied in some detail. One was located three miles west of St. Ignace and has since been partially obliterated by relocation of the state road. At this location there were to be found apparently normal [ris versicolor and Iris virginica and a large number of peeuliar intermediates. Few of these resembled the first- generation and second-generation hybrids which had been raised in the experimental plots. Most of them, on the other hand, were very similar to back crosses which had been made between the first-generation hybrids and the two species. This is borne out by fig. 14 where representative colonies of each species are contrasted with these two hybrid colonies, using the following index as a basis for comparison: sepal length/petal length + sepal width/petal width. Figure 14 also shows the very different condition which was tered at the other hybrid colony which was studied in detail. It was located just west of Engadine, Michigan, and was composed entirely of plants which closely resembled the artificial Fı hybrids of the breeding plot. Since they also presented the characteristic vigor of such hybrids it is probable that they were for the most part such hybrids, the parental species having been extermi- nated by their vigorous offspring. It is not impossible, how- ever, that they may represent some new balanced combination of chromosomes. In crosses between such complex amphi- diploids as Iris versicolor and Iris virginica various new poly- ploid types are not at all unlikely. Unfortunately, the large numbers of chromosomes and the very short period during which the reduction division can be studied (only a few days out of the year) render this rather unprofitable material for such examination. One can summarize the effect of hybridization between Iris versicolor and Iris virginica by saying that it does occur very occasionally. Its only effect so far has been a slight blurring of the two species along the zone of contact. It does, however, 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 507 provide a means by which new polyploid hybrid species might eventually arise. Were there fewer technical diffieulties involved a detailed analysis of hybridization in these irises would be well worth while, since hybridization is a characteristie part of the phylo- genetic pattern throughout the genus. The remarkable phylo- genetic tangle of species and hybrids discovered in the Mis- sissippi delta by Dr. J. K. Small is by no means exceptional. Much the same situation exists among the Pogoniris species around the Mediterranean ; the irises of the Spuria group ap- parently behave in the same way in western Asia; it was per- haps from some such complex that the Japanese irises had their beginnings. Because hybridization is apparently such an important phylogenetic factor in the genus Iris, the hybrids of the Missis- sippi delta deserve careful analytical study, genetically, taxo- nomically, and cytologically. It is not enough to prove that hybridization is taking place. Much more important is the determination of its exact role among the Louisiana irises. The effects of hybridization are various according to the peculiarities of the germ-plasm upon which it is operating and the external conditions under which it takes place. It may re- sult in new amphidiploid species such as Iris versicolor. It may produce intermediate swarms which obliterate previous specific boundaries. It may increase the variability of one of the parental species by introducing a small proportion of germ- plasm from the other (Anderson € Woodson, ’35, p. 37). Possibly all of these processes are taking place among the Louisiana irises. It should not be difficult to determine their relative importance and ultimate phylogenetic effects. These irises possess a number of technical advantages for such a study. They have comparatively low chromosome numbers; they are easy to cultivate in experimental gardens; they pos- sess conspicuous specific differences. They are, as Viosca has said (735), ‘‘uniquely adapted for the study of experimental evolution?! and one may join with him in predicting a ‘‘bright future for them in the study of biology." [VoL. 23 508 ANNALS OF THE MISSOURI BOTANICAL GARDEN To summarize: the evolutionary patterns of the blue flags are fairly typical of the genus as a whole. Seen in a greatly foreshortened view they can be represented graphically as in fig. 13. They form a somewhat tree-like system of cables, the ground pattern occasionally made more complex by amphi- diploid cross-connections. The pattern is complicated at one point by inter-specific hybridization. The cables themselves are divided into distinctive cords, the colonies. Much the same set of patterns would characterize the entire genus Iris. The complex knots due to interspecific hybridiza- tion would be larger and more significant in certain other por- tions of the genus; in certain sections the cross-connections of amphidiploidy would be lacking altogether; in a few sections they might be somewhat commoner. All in all, however, fig. 13 may be taken as a fairly accurate representation of phylo- genetic relationships for the whole genus. This general sim- ilarity of evolutionary patterns within the genus rests upon the fact that the germ-plasm of any species of the genus Iris is not a vague generalized germ-plasm. It is Iris germ-plasm. It has a number of inherent characteristics which affect specia- tion and which cause the details of speciation, to be similar throughout (Anderson, ’31). It is a genus in which amphi- diploidy occasionally occurs, in which every species has a strong development of vegetative propagation, in which inter- specific fertility is the rule. It is, on the other hand, a genus which is ecologically conservative. Species after species exhibits rather precise demands as to habitat and seems unable to produce variants adapted to more or less moisture, more or less shade, more or less acidity. In the genus Aquilegia every species which I have studied ex- tensively is richly provided with ecotypes; sun-forms, shade- forms, types inherently adapted to sour, wet swamps, to sunny, limestone cliffs, to sand dunes, and to woodlands. In Iris versi- color and Iris virginica, in spite of intensive search, I have been unable to recognize a single ecological variant. Like many genera of the Liliiflorae, the order to which they belong, they are curiously unamenable to ecological differentiation. This conservatism, since it characterizes the whole genus, must rest 1936] ANDERSON—THE SPECIES PROBLEM IN IRIS 509 upon some inherent property of Iris germ-plasm. What that property may be we cannot, as yet, even hazarda guess. It may be that the germ-plasm of Aquilegia varies in a way that is unknown or uncommon to the germ-plasm of Iris. It may be that due to its ontogenetical organization the genus Iris, though possessing the same basie kinds of germinal variation as the genus Aquilegia, will nevertheless not respond with variants equipped for other habitats. BIBLIOGRAPH Y Anderson, Edgar ('28). The 221 of species in the northern blue flags, Iris versicolor L. and Iris virginica L. Ann. Mo. Bot. Gard. 15: 241-332 , (31). Internal pst affeeting diseontinuity tm species. Am. Nat. 65: 144—148. , (733). The distribution of Iris versicolor in relation to the post- glaeial Fe Lakes. Rhodora 35: 154-160. , and Robert E. Woodson (735). The species of Tradescantia in- digenous to the United States. Arn. Arb. Harv. Univ. Contr. 9: 132 pp. Dykes, W. R. (713). The genus zs NEE L. F. (’34). Chromosome numbers in native American and introduced species and cultivated varieties of Iris. Am. Iris Soc. Bull. 52: 61-66. Simonet, Mare (734). Nouvelles recherches eytologiques et génétiques chez les Iris. Theses Fac. Sci. Paris, No. 15: ———————, (785). Conjugaison Weeer "ES des chromosomes. Bull. Biol. France et Belg. 69: 178-212. Viosea, Perey Jr. (’35). The irises of southeastern Louisiana. Am. Iris Soc. Bull. 57: HYBRIDIZATION IN AMERICAN TRADESCANTIAS EDGAR ANDERSON Geneticist to the Missouri Botanical Garden Professor of Botany in the Henry Shaw School of Botany of Washington University FOREWORD Widely different opinions are held by present-day biologists as to the evolutionary importance of hybridization between species. One of the main reasons for the disagreement seems to be the lack of summarized and codified data bearing upon the problem. This series of papers is not an attempt either to prove or disprove the importance of hybridization. Itis rather an attempt to gather pertinent data from which to analyze the eff ects of hybridization. The genus Tradescantia has been chosen for two main reasons: (1) Previous taxonomic, cytological, and genetical studies (Anderson and Woodson, '35; Anderson and Sax, '36; Anderson, ’36) have indicated the main outlines of the situa- tion in this genus; (2) Tradescantia for purely technical rea- sons is a better object for such investigations than are many other genera of the higher plants, chief among its assets be- ing the facility with which it can be studied cytologically and the ease with which most of the species can be brought into cultivation. The main object of the investigation, as stated above, has been to procure data from which the importance and the evolu- tionary rôle of hybridization could be estimated and demon- strated. It had previously been experimentally determined that most of the species allied to T. virginiana could be hybrid- ized and that their hybrid progeny were semi-fertile. Similar hybrids had been found in the field and had been subjected to detailed morphological and cytological analysis, but such work was purely preliminary. It merely demonstrated the occur- rence of hybridization. What was needed was a method which would determine the dynamies of hybridization in natural populations. In mathematical terms it was necessary to ascer- tain the frequencies as well as the range of variation ( Anderson ANN. Mo. Bor. GARD., VOL. 23, 1936. (511) [Vor. 23 512 ANNALS OF THE MISSOURI BOTANICAL GARDEN and Turrill, ’35). That is to say, that we needed to measure and record the effect of hybridization upon the entire popula- tion in which it was occurring. Such a method has been evolved and is outlined in section I, while its application to a partieular case is demonstrated in section II. I. A METHOD ror MEASURING SPECIES HYBRIDS Biology employs two methods of measurement, the quantita- tive and the qualitative. Each has its advantages and dis- advantages. T'wo of the main objections to purely quantitative methods are: (1) quantitative scales are arbitrary and may not bear a direct relation to the phenomenon under investiga- tion, (2) the observational basis of purely quantitative methods is too narrow for many biologieal problems. It may be well to amplify these dicta. (1) The chief purposes of the quantitative method are to record the variation of certain phenomena and from this record to analyze the forces producing the variation. If a scale has been chosen which is an accurate reflection of the main under- lying factors, then the analysis will be a comparatively simple matter. But if the scale used is not in harmony with these underlying causes then it may be difficult, or impossible, to analyze the data. In other words, the data are then no more than a record; a qualitative measure, if objectively defined, would have been quite as efficient. In a paper of fundamental biological significance Wright has demonstrated (’26) how even as simple a transformation as a percentage scale can ob- scure the interpretation of quantitative data. In many biologi- cal problems the customary quantitative scales in which the original observations are recorded are even more inefficient and misleading. The fact that centimeters are equally spaced on the ruler does not mean that variations of a centimeter are always of equal biological magnitude when we use that ruler to measure plants and animals. Purely quantitative measures, lengths, areas, weights, etc., are superior to qualitative units only when they reflect more or less directly the changes in the underlying factors. (2) As has been pointed out by Minot (’11), the observa- 1936] ANDERSON—HYBRIDIZATION IN TRADESCANTIAS 513 tional basis of purely quantitative methods is often too narrow for studying biological phenomena. A set of weights, or lengths, is too insignificant an observation to yield sufficient data for exploring a problem. Biology has advanced most rapidly when appropriate qualitative measures have been de- veloped and used with precision. In Genetics, for example, the fundamental data are qualitative. Once obtained they are treated with such precision that most geneticists probably think of their work as purely quantitative. But the funda- mental categories, ‘‘vestigial’’ vs. ‘‘non-vestigial,’’ ““scute”” vs. ‘‘non-seute,’’ ‘‘forked”’ vs. **non-forked,"' ete., are quite as qualitative as the fundamental categories of taxonomy. It is because of this fact that they are a broad enough tool to yield useful information about such a complex phenomenon as he- redity. If the methods of Drosophila genetics were purely quantitative the flies would not be classified in qualitative categories, but their wing lengths, eye diameters, etc., would be laboriously measured. Imagine the difficulties of conducting a Drosophila experiment involving two or three wing mutants in which the only available data were the lengths and breadths of the wings! Genetics has been able to advance because it was willing to take the Mendelian recessive (a qualitative unit about whose ultimate significance relatively little was known) and to use that unknown but recognizable entity as a basic unit. For the study of the species problem a similar combination of qualitative and quantitative methods seems desirable. Few of the differences between species are of the simple sort which are readily amenable to quantitative treatment (Anderson and Whitaker, '34). This is not to say that species do not differ quantitatively. They do, just as do the wing mutants of Droso- phila, but in both cases it is more efficient to use qualitative categories. Purely qualitative methods, however, have their own dis- advantages. They are often said to be more subjective, but this point is open to argument. Certainly the gene differ- ences by which Drosophilae are scored are quite as objective as quantitative measurements. A very real disadvantage in purely qualitative work is that it is not commensurate. One cannot with purely qualitative methods make an accurate com- [Vor. 23 514 ANNALS OF THE MISSOURI BOTANICAL GARDEN parison of the Iris plants of one meadow with those of another. What we need for the species problem is a method whose funda- mental observations are based upon the qualitative categories of taxonomy but which treats these categories in such a way that they can be used for comparison and analysis. This has been done below in a fairly simple fashion by constructing an index which is the resultant of all the qualitative characters which are readily available. In the simplest possible case the method operates as follows: Given two species, A and B, the hybridization between which is the object of investigation. Given n readily ascertainable dif- ferences between A and B (as, for instance, position of inflo- rescence, nature and distribution of pubescence, number of nodes to the stem, distribution of stomata, etc.). For each of these characters an individual plant can be scored as ‘a’ (like species A); as ‘b’ (like species B), or as ‘i’ (intermediate, preferably the actual F; if that has been obtained). In partic- ular cases it will be possible to define two or more intermediate grades for the character in question. A hybrid population will be scored somewhat as follows in a case where six distinguish- ing characters were available: plant no. 1: a-a-a-i-a-a plant no. 2: a-a-i-i-b-i plant no. 3: 1-1-1-1-i-1 plant no. 4: b-i-i-i-b-a and so on for all the plants which are scored. To summarize the population as a whole, all that is necessary is to turn this qualitative scoring into a quantitative index. An index run- ning from complete ‘‘A-ness’’ to complete ‘‘B-ness’’ can be made by arbitrarily giving every ‘a’ the value of 0, every ‘b’ the value of 2, and every ʻi’ the value of 1. This will produce a scale with 2n + 1 divisions (in this case 13, 1. e., 0 to 12 inclu- sive. An individual of species ‘A’ would have an index value of 0 (0+0+0+0+0+0), while an individual of species ‘B’ would have a value of 12. Plant no. 1 in the above example will have an index value of 1; plant no. 2 a value of 5; plant no. 3 a value of 6; plant no. 4 a value of 7. When the values of each plant 1936] ANDERSON—HYBRIDIZATION IN TRADESCANTIAS 515 have been computed they can be summarized as a frequency distribution. The frequency distribution constitutes a record of the whole population, in so far as these two species and mix- tures between them are concerned. It can be used for the com- parison of different hybrid populations and to discover and analyze the forces which are at work in such populations. The index can be varied to fit the available data in various ways. When, for instance, certain of the categories are thought to be more or less reliable than the others they can be appropriately weighted in combining the index. The method seems capable of producing useful and unique data in regard to the variation within and between species. I have used it extensively in the genus T'radescantia, one case being published in the second part of this paper. I have also applied it successfully to the analysis of hybridization in Amelanchier and Baptisia (un- published). Dr. H. P. Riley has applied it to the hybrid Jris populations of the Mississippi delta with interesting results (Riley, in press). After I had originated the method Dr. Jens Clausen very kindly called my attention to a paper by Raunkiaer (’25) in which an almost identical method of recording the qualitative categories had been applied to a case of hybridization in Crataegus. He had not, however, performed the further step of turning the qualitative record into an index, thus making it possible to summarize the population. When this is done the case proves to be a partieularly interesting one with certain unique features. SUMMARY The advantages and disadvantages of quantitative and qual- itative methods in biology are diseussed. A method of sum- marizing hybrid populations is developed which utilizes quali- tative categories for its initial observations and from these computes a quantitative index value. II. HYBRIDIZATION BETWEEN T. VIRGINIANA AND T. CANALICULATA The method developed in SEcrrox I has already been applied to numerous cases of hybridization in Tradescantia. The fol- i [Vor. 23 516 ANNALS OF THE MISSOURI BOTANICAL GARDEN lowing one has been selected for preliminary demonstration because it has been analyzed in greater detail and because it is a relatively simple case with no such complicating factors as differences in chromosome number. Tradescantia virgimana L. and T. canaliculata Raf. (T. re- flexa Raf.) are the two commonest Tradescantias of eastern North America. Although each can be cultivated under a variety of conditions they are usually found growing naturally in quite different situations; T. canaliculata in the sun, often upon or near rocks or in dry sands; 7’. virginiana in shade or semi-shade. Until white civilization disturbed their natural relationships they seldom or never occupied the same habitat. Even at the present time active hybridization is confined to a very small proportion of either species. The outstanding morphological differences between the two species are as fol- lows (those characters marked with an asterisk * require the use of a good hand-lens) : T. canaliculata Index value 6-8 nodes 0 T. virginiana Index value 2-3 nodes 2 (1) (2) Sepals broadly ovate, apex broadly aeute; Sepals elliptic, apex nar- rowly acute; calyx not in- calyx inflated after an- flated after anthesis 0 thesis 2 (3) Sepals and pedicels pu- Pedicels glabrous; sepals bescent throughout 3 glabrous or barbate 0 (4) * Stomata no more con- *Stomata (and gereent spicuous than the cells mer? much more conspic- of the upper epidermis 1 than the cell of the upper epidermis 0 (5) * Stomatal areas of the * Stomatal areas much upper epidermis nar- broader than the areas rower than the areas without stomata 0 without stomata 1 (6) * Longitudinal dis- * Longitudinal distance tance between stomat shorter than the diameter longer (often several of a stoma 0 imes as lon han diameter of a stoma 1 Total index value 10 0 1936] ANDERSON—HYBRIDIZATION IN TRADESCANTIAS SIT It was a relatively simple matter to turn these characters into an index, according to the method developed in section I. The figures to the right show in each case the index value as- signed to that character. In the case of (3) it was possible to recognize two intermediate grades. If the pedicel was weakly pubescent and the calyx showed the beginnings of a barbate tuft at the apex it was scored as ‘vi’ (virginian a-intermediate) with an index value of 2; if it had a strongly developed barbate tuft with scattered OE on the calyx it was scored as ‘ci’ (canaliculata-intermediate) with an index value of 1. Cate- gories (4), (5), and (6) are given only half the weight of the previous three since they are all different measures of the same quality, the distribution of stomata on the upper epidermis. It is certainly true of Tradescantia, as Wiegand ('35) has observed for Amelanchier, that hybridization is much more frequent in areas greatly disturbed by man than under more natural conditions. For that reason an area was chosen for detailed examination in which hybridization between T'. vir- giniana and T. canaliculata was for the most part taking place under nearly natural conditions. The spot finally chosen is illustrated in Mapl. Itisan area two miles square in the town- ship of Joachim, in Jefferson County, Missouri. After the area had been selected it was visited repeatedly by foot and by automobile in an effort to record the exact distribution of all the Tradeseantias within the four square miles selected. The region is a low plateau, very much dissected by small streams which flow into the near-by Mississippi River. 'The land has been under cultivation for over a hundred years, the flat hill-tops (plateau remnants) and the rich valley bottoms being for the most part in cultivated crops while the steeper hillsides are left as woodlands or as woodland pasture. The underlying rocks are sedimentary and lie in practically the po- sitions in which they were deposited with little faulting, fold- ing, or tilting. All the strata are limestone with the exception of the St. Peter’s Sandstone, here a soft gray-white stone usually exposed as a more or less perpendicular cliff capped with resistant limestone. The stratigraphy is of considerable importance to the prob- [Vor. 23 518 ANNALS OF THE MISSOURI BOTANICAL GARDEN lem in hand, for it is these steep limestone and sandstone cliffs which constitute the natural habitat for T. canaliculata. They are so rocky and dry that not even in the original forest were they fully covered with trees, particularly on southern and western slopes. It will be seen on the map and in fig. 1 that Map I. Distribution of T. virginiana (small closed circles), T. canaliculata (large open circles), and hybrids (black and white). The letters A to F indicate places at which detailed collections were made. The line 2-3 represents the position of the section illustrated in fig. 1. T. canaliculata is confined to this narrow belt just below the hill-tops, while T'. virginiana occurs pretty generally through- out the region on gentle slopes, on hill-tops, and in valley bottoms, but practically always in the shade. In this particular area T'. virginiana was much more common than T. canaliculata. An effort was made to estimate relative frequencies by computing from the map the areas occupied by 1936] ANDERSON—HYBRIDIZATION IN TRADESCANTIAS 519 T. virginiana, T. canaliculata, and the hybrids. These were found to be 16 : 1 : 0.5 respectively. Thirty plants each of T. virginiana and T. canaliculata were scored at the localities marked as ‘F’ and ‘B’ on the map. The pure species, as can be seen from table I and fig. 2, were rela- tively invariable and agreed with scorings made at other points in their range. A few hybrid plants were found along the zone of contact between the two species and large hybrid colonies, consisting of fifty or more individuals, occurred at four places, labeled A, C, D, and E on the map. Detailed studies were made at the first two localities. l. Diagrammatie section (to seale) along line 2-3 in Map I. Figures to the left show feet above sea-level; closed circles represent T. virginiana and open circles T. canaliculata. The line is slightly over 2 miles long. Locality A. At this point the public road starts up onto Sandy Ridge along the hard limestone which caps the St. Peter's Sandstone. The latter forms a series of much- weathered cliffs slightly to the south of the road and while the area is pastured, it has not been greatly disturbed of late years. At various points along the cliff a few plants of T. canaliculata are growing, and as the road swings to the right the cliffs and the ledge become somewhat shadier by reason of the narrow gorge eut back into the hill by an intermittent stream. At the head of the gorge there is an interesting area of a few hundred square feet in which many kinds of habitats and hence many kinds of plants are brought into close juxtaposition. Rock- plants, prairie grasses, woodland and even desert species are found within a few feet of each other. It is, in other words, exactly the situation in which one might logically expect to find hybridization between ecologically isolated species. Trades- [VoL. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN 520 TABLE I SCORING OF INDIVIDUAL PLANTS AT THREE LOCALITIES IN JEFFERSON FURTHER EXPLANATION IN TEXT. COUNTY, MISSOURI. a e n a C a a £V SR RSS C d [S ES FEES $ ZS ES E FO RSPR E Com rd rd Erd rä rd E E E En A m ri rd ri rd ri C C rd rd Ei Erd rd O *H MN NL DO At 9 00 to t- e$ E rd oc ri rä rnm onm rn a gei T on re AA bh bbe be b h ooocoooococoocococoo0o0o0 Mormon ort O Oem On Osmom Ben © ^ Wl) PPP PPh ee rr ba > e HOON OH 00 Den 9 O9 9 Ow ono EF m bem O bin BO mn ERES E ki ES EFFE SS SG SG P Pa E p aA Orem OOOO 00 OO Own OO aA POCO E Pan Piet E bP Oe E SS PZ 3 Ka mi op e . M IM ond e Slsslirrrrtrttrtkttttk ovooovooovovoooooo |SS| ICA O RE PER ppo o 5 AR Ss s a bbbbbbbbbobbbppp g|o220000o00900000000 E RO lic TO + ee 3 ra K S E e bb EE PP bb bb bb bb p E On O On O D Oem sp OD Do. Y Rid orden o bebes S| 8. E e > SIORMASOHAMHAMOrARS EISES OM~ORO *O t- 00 C) C r C1 c5 H M (to t- 00 ; SB ÍA HHH d o! 0 NANNAN CI e $ Fd rir c 6 OW CI ci ci 6&1 Ql 63 6l c3 S Tc n n OI CIO AAAS E Let E 3 = a AA al E SG | llo®o*%e EA ESS ees sis mi Cnace e e e bal E ER le O ommno cuu P M3 15 C3 *O C1 C1 i em e ei + m > = =~ La > E o POPP bb Pr bb bh helo oo 80980889890 LE boo Omi ba gen Om O e e | D P be Poe bib ple baaa pb fll om Omn 6.0 On OO Uni oO OO blosen Din O 0 Oe et e rd o m m RIS Pon Peet PP em. PP ee bb E O O O OO Veet eee OOO UO Dent cn E EE Ee EES HE H sp D D E O)|/ rr b> br b> bh bh be bh bh pp bd occcocoooooooooo00 SSP SSE OSS OSE Doo P g Al h be hn bbb he be > > p Ooooocooooocooo0oo0oon O O O O Venue O O P E Aen sei ttm, ch rPPPPr rr PrP PrP PP bP Pp Pb o OO OO oom Y Dor OU VO OY D ven set sp sp vi sel sei vi sp sei pi vi ven E E SSILANMHHOCRDRMROHAMHNH m Gi c5 AHI t- 00 O CO c EI c5 HID t- wis rA n n A A NA AA A ao weng HUM S 1936] ANDERSON—HYBRIDIZATION IN TRADESCANTIAS B 5123425609 20m Fig. 2. Fre sier? distributions for the virginiana- -canalicu- lata index. Thirty plants each collected at loealit ies A, B, C (two terr emg si F illustrated in Map I; R and 8 at Algon- quin Station, Y at e Él ood, Missouri. Index values shown on the scale at "us inar. of the diagram. Br «presents a — tion of pure 7. canaliculata, F a population of pu giniana ; the rem: Gage hybrid populations. e dud Kc tion in the tex 521 [Vor. 23 522 ANNALS OF THE MISSOURI BOTANICAL GARDEN cantias occur rather generally over the whole area; a few plants of apparently straight 7. virginiana are growing in the shade of the red cedars and a mongrel population of T. vir- giniana x T. canaliculata grows in the sunny and semi-shady situations. The cliff at this point is so steep that it was imprac- ticable to collect plants in quantity from the cliff-sides but all of those growing on the ledge were scored in detail. The re- sults are presented in table I and fig. 2, A. (In presenting the frequency distributions only the first thirty plants have been (nm © » m 0-1 1 2-3| a 1 2 4 4 2/2 Se Se? yel 10. tim si black, spores dark brown, ellipsoid or ES rin A 12 11. Spores broadly ovoid or subspheroid, 8-9.6 x 6.7-7.7 y. .8. M. Parksii (p. 649) 11. Spores ovoid, ellipsoid to fusiform, somewhat dio below, 8-9.6 x 5 A 7. M. rubescens (p. 648) 12. esis less than 4 y broad, 5.5-6.7 x 2-3.7 gu... see eee 9. M. mollis (p. i 12. Spores 4 y or more broad...... esee ttn [Vor. 23 642 ANNALS OF THE MISSOURI BOTANICAL GARDEN 13. Spores truncate, 6-8 x 4-5 y, exospore conspicuous. . 10. M. durissimus (p. 650) 13. Spores not truncate, 7.5-10.3 x 5-7.5 p, exospore not conspicuous........ KEE EE 11. M. variegatus (p. 651) 1. MELANOGASTER AMBIGUUS (Vittadini) Tulasne, Ann. Sci. Nat. Bot. IT. 19: 378. 1843. Octaviania ambigua Vittadini, Monogr. Tuberac. 18. 1831. Hyperrhiza liquaminosa Klotzsch in Dietrich, Fl. Reg. Boruss. [Fl. Kónigr. Preuss.] 7: no. 468. 1839. Argylium liquaminosum Wallroth, Fl. Cryptog. Germ. 2: 874. 1833. Melanogaster Klotzschii Corda, Icones Fung. 5: 23. 1842; Anleit. z. Stud. Myc. 95. 1842. Illustrations: Vittadini, Monogr. Tuberac. pl. 4, f. 7. Type: location unknown to us but specimen so determined in Broome Herb. at British Museum. Argylium liquaminosum from Thüringen, Wallroth. Fruetifieations subspherical to ellipsoidal, 1-3.5 em. in di- ameter, pale olive to olivaceous-brown when fresh, becoming cinnamon-brown to Prout’s brown on drying, surface cottony- furfuraceous, fibrils scanty, inconspicuous; peridium 300-600 y thick when fresh, somewhat thinner in old dry material, hya- line within, yellowish-brown toward surface, prosenchyma- tous, homogeneous, with outer filaments fraying out to form the more or less erect, brown tomentellum at the surface, of large, thick-walled, agglutinated hyphae which often collapse on drying; gleba black with yellowish septa, fundamental tramal tissue of more slender, smaller-celled prosenchyma, cavities filled with spores embedded in a gel; basidia clavate, evanes- cent, mostly 4-spored; spores (8-) 12-16 (-17) x (5.5—) 6-8 (-8.5) », obovoid with more or less acute apex, approaching cit- riform, thick-walled, smooth, very often uniguttulate, dark brown when mature. Under hardwoods. Central and western Europe, New York, Kentucky, Oregon, and New Zealdnd. April to October. The specimen (Zeller Herb. 2338) collected by Dr. Helen M. Gilkey along Greecy Creek, Benton Co., Oregon, differs enough from M. ambiguus for us to refer it here with some doubt. The peridium in the one fructification has an outer white layer 1936] ZELLER & DODGE—MELANOGASTER 643 (50-60 » thick) which darkens when bruised as described for M. aureus. The inner peridium is dark brown (150-170 p thick). The spores are as in M. ambiguus, but the cavities are filled with a meshy growth of hyphae bearing the basidia with no evidence of gelification. GERMANY: Preussen, Neu Schóneberg bei Berlin, F. Seen type of Hyperrhiza liquaminosa (Kew, Paris); Marburg, E. Hesse, 1885 (Farlow). ENGLAND: Bristol, C. E. Broome 355 (Brown Univ.) ; t s Lynn, C. B. Plov- right 508 pr^ Mus., Stanford). FRANCE: L. Quélet (Upsala); Seine, Bois de Boulogne, July 9, 1843, L. Tulasne (Paris). ITALY: near Milano, C. Vittadini (Brit. Mus.). New York: Syracuse, L. H. Pennington (Lloyd Mus. 87, and FS vit KENTUCKY: High Ridge, C. O. e ee ES F. T. McFarland (Zeller). OREGON: Benton County, Alsea Road . Barss icm 681), ES M. Gilkey (Dodge, and Zeller 2338, 2803, 2805) ; a GC , L. M. Boozer (Dodge and Zeller 2285); Kings Valley, S. M. Zeller 8187 (Zeller) ; Zeen yen H. M. Gilkey (Zeller 6816). New ZEALAND: Canterbury, Oxford, G. Archer (Cunningham 576, and Dodge) ; Wellington, G. H. Cunningham, 1928 (Dodge and Zeller); Helen K. Dalrymple (Lloyd Mus. 3, and Dodge). la. var. EURYSPERMUS Zeller & Dodge, Ann. Mo. Bot. Gard. 22: 373. 1935. Type: in Zeller and Dodge Herbaria. Fructifications globose to irregular, up to 3.5 cm. in diameter, snuff-brown to bister, not changing on drying, surface smooth, dull, fibrils few, concolorous, not extending halfway up, not distinctly radicate, adnate-appressed; peridium thick, tough, 320-400 » thick, of dark-brown hyphae with thin walls, some- what vesiculose, variable in size, closely stupose; gleba with white partitions (drying yellowish) and coal-black cavities, gelatinous, cavities up to 3 mm. thick, filled with spores at maturity; septa variable in thickness, hyaline, of small hyphae up to 2.5 » in diameter; basidia with long slender (2.5 ») ped- icels of odd lengths, tips pyriform, 5 x 7 », 4-spored, sterigmata short, up to 3 » long; spores ellipsoid to citriform, black, 10-11 x 7.5-9 » (usual mature spore 11 x 8.5 »); odor of Rhenish wine. In strawberry patch. Oregon and California. March to April. There were 37 fruiting bodies in the type collection. The odor is distinctive. The variety may be distinguished from the [Vor. 23 644 ANNALS OF THE MISSOURI BOTANICAL GARDEN typical species by the large cavities, white septa, the color of peridium, and more abundant spores. OREGON: Polk County, Rickreall, Etta Neiderheiser, type (Zeller 2660, and Dodge) ; Benton County, Corvallis, L. M. Boozer (Zeller 2346), S. M. Zeller 7059 (Zeller). CALIFORNIA: Santa Clara County, Call of the Wild, H. E. Parks 187 (Univ. Cal. and Dodge). 2. MELANOGASTER MACROSPORUS Velenovsky, České Houby, 808. 1922. Type: location unknown to us. Fructifications tuberiform, fleshy, size of a walnut or even larger, brown when young, becoming dark horsechestnut-brown, soft, elastic, surface dull, thickly covered with long black fibrils; peridium very thin, black, not separable; gleba of black cavities with either yellow- ish or whitish septa, when ripe becoming gelatinous, blackish with yellowish veins; spores obovate-ellipsoid, narrower at the base, at- tached with 3—4 small bodies, reddish-brown, smooth, 15-18 p. long; odor very fetid. In leaf humus of broad-leaved trees in summer, around Chuehle, common, in Trnova, Bechovice, and Karlstein. It seems to me that our fungus is M. variegatus Vitt. which is very often deseribed in Ger- many and is quite common in other places. The spores appear ellip- soidal, 6 x 10 y; either this size is wrong or our fungus is a new species. I have examined the spores very carefully from all points of view. —Velenovsky. The following description is based on Oregon material which agrees closely enough with the above description to refer it here temporarily, pending opportunity to study the type: Fructifications up to 12 x 8 x 6 em. (weight 270 g. when fresh), drying russet to Mars brown and black; peridium duplex, outer layer about 140 » thick, composed of loosely woven, periclinal, dark brown, thick-walled hyphae 7 y in diameter, about 50 y thick, of slender, gelified hyphae continuous with the septa; gleba black, mottled by white septa which are 90 y thick, highly gelified; basidia evanescent; spores broadly obovoid, 12-15 x 8-11 », dark brown; odor of spoiled silage or of onions without tear-producing qualities. In duff under conifers, especially hemlock. Oregon. Sep- tember. 1936] ZELLER & DODGE—-MELANOGASTER 645 OREGON: Clackamas County, on trail to Paradise Park from Twin Bridges Forest Camp, S. M. Zeller 7759, 8230 (Dodge and Zeller); Lane County, Belknap Springs, S. M. Zeller $ G. R. Hoerner (Zeller 8241). 3. MELANocasTER intermedius (Berkeley) Zeller & Dodge, comb. nov. Melanogaster ambiguus var. intermedius Berkeley, Ann. & Mag. Nat. Hist. I. 13: 352. 1844. Type: in Broome Herb. at Brit. Mus. and Tulasne Herb. at Paris. Fructifications 2-4.5 em. in diameter, subglobose to reni- form, fibrils plentiful below, at a definite place of attachment, surface even to undulating, tomentulose, very dark reddish- brown when fresh, becoming blackish where bruised, drying Prout's brown to mummy-brown; peridium 125-160 y thick, light yellow to brownish near the exterior, stupose, of hyphae with vesiculose cells, homogeneous, but loosely woven at the surface with some erect hyphae giving the tomentulose char- acter, distinet from the tramal tissue; gleba slaty-black, slightly lined or streaked with thin septa, of hyaline (dilute yel- lowish) gelified hyphae, cavities filled with spores in a gel; spores dark, mostly obovoid, obtuse, smooth, very rarely some- what fusiform, tip subacute, 11-13 x 7.4-8 »; odor strong, of- fensive, suggesting asafoetida. Under hardwoods. England and central California. ENGLAND: Wiltshire, Spye Park, C. E. Broome (Brit. Mus. and Paris). CALIFORNIA: H. E. Parks, 551, 1123, 1124; Santa Clara County, Alma, H. E. Parks 7, Z29, Z328, 405; Saratoga, Boys’ Outing Farm, A. E. Parks 366; Brook- dale, H. E. Parks 2162bis; Guadaloupe Mines, H. E. Parks Z3, Z7, Z25, Z349, 145, 381, 412, 416; highway to Watsonville, H. E. Parks 3307 ; Del Norte County, Cres- cent City, H. E. Parks 2862; Humboldt County, Trinidad, H. E. Parks 4122, 4436, 4443 (all Univ. Cal. and Dodge). 4. MELANOGASTER TUBERIFORMIS Corda in Sturm, Deutschl. FLHIL3IL 1. 1831. Hyperrhiza tuberiformis Rabenhorst, Deutschl. Krypt. Fl. 1: 292-293. 1844. Illustrations: Corda in Sturm, Deutschl. Fl. IIT. 11: pl. 1. Type: Portions in Berkeley Herb. at Kew and Tulasne Herb. at Paris. Fructifications 1.5-7 em. in diameter when fresh, drying 1- [Vor. 23 646 ANNALS OF THE MISSOURI BOTANICAL GARDEN 2 cm. (in material examined), subspherical, bay or reddish- fuscous drying Rood’s brown to Vandyke brown; fibrils rare, black, appressed, basal; peridium stupose, amber to darker, 100-185 » thick ; gleba black with white or slightly yellowish, in- conspicuous septa, prosenchymatous, cavities very irregular; basidia evanescent, usually 8-spored; spores obovoid, some- times pyriform, very dark brown, smooth, 10-11.5 x 5.5-7 y. Under Pinus, Erica, Quercus. Central Europe and western North America. June to August. CZECHOSLOVAKIA: near Praha, Corda, type (Kew and Paris); Bóhmerwald, E. Soehner ?74 (Soehner and Dodge). GERMANY: Spiegelslust, R. Hesse (Farlow). FRANCE: Alsace, Ballon, L. Quélet (Upsala). IDAHO: Priest Lake, C. V. Piper (Lloyd Mus. 6164). OREGON: Benton County, Corvallis, S. M. Zeller 7760 (Zeller). CALIFORNIA: no locality nor collector (Univ, Cal. 552, and Dodge) ; Santa Clara County, Saratoga, H. E. Parks 968 (Univ. Cal. and Dodge); Marin County, Mill Valley, H. W. Harkness 5 (Stanford). 9. MELANOGASTER MICROSPORUS Velenovsky, České Houby, 809. 1922. Type: location unknown to us. In some respects similar to M. rubescens but it is larger, up to 8 em. in diameter, without odor or with weak odor of meat. Spores very small, ellipsoidal, pointed on both ends, 4—5 D, very thick-walled, light yellow. In leaf mould in woods in Tuchomeric. July, 1918, J. Lissal. —Velenovsky. Mattirolo, Beitr. Krypt.-Fl. Schweiz (Contr. Fl. Critt. Sviss.) 8*: 37-39. 1935, described a fungus as M. microsporus without reference to the previous use of the name by Velenov- sky. His description follows: Fructifications irregularly globose, size of a hazel nut, yel- low-umber; peridium 150 y thick, of three layers, the outer forming a tomentum, the middle deeply colored, of slender crowded hyphae, the inner layer hyaline, thicker, pseudoparen- chymatous; gleba chestnut to fuliginous; basidia clavate, 4- spored; spores ellipsoidal, 3-4 x 2 y, similar to those of Rhiz- opogon rubescens. Under chestnut and pine. Rodero. Como Province. 1936] ZELLER & DODGE—MELANOGASTER 647 6. MELANoGAsTER Broomeranus Berkeley in Tulasne, Ann. Sci. Nat. Bot. II. 19: 377. 1843; Ann. & Mag. Nat. Hist. I. 13: 353. 1844. Melanogaster variegatus var. Broomeianus Tulasne, Fung. Hypog. 93. 1851. Tuber moschatum Sowerby, English Fung. pl. 426. 1815.— non al. Illustrations: Tulasne, Ann. Sci. Nat. Bot. II. 19: pl. 18, Fruetifieations subglobose to irregularly lobed, often co- alescing when cespitose, 2-4 em. in diameter; surface ochra- ceous then Dresden brown with a flesh tint, becoming blackish where handled, drying tawny-olive to warm sepia, appressedly tomentulose; brownish fibrils below; peridium variable in thickness, 50-200 » thick (mostly 50-100 4), of stupose pro- senchyma with large vesiculose cells, yellow-amber to darker toward the exterior, extending as fundamental tissue into the trama; gleba sooty-black, gelified, tramal septiments white to creamy, of subgelatinous hyaline hyphae, cavities filled with spores in a gel; spores light brown, narrowly oblong, broadly truncate below, (5-) 6-8.4 (-11) x (2.5-) 3.5-4.0 (4.5) »; odor of coffee grounds. Under hardwood trees. July to February. Europe, Michi- gan, Indiana, and Oregon. Close to M. rubescens in spore characters, but the spores are darker and definitely truncate below. Peanut-shaped spores are seldom found in this species, but even so they retain the truncate character. ExsriccATI: Berkeley, British Fungi, 285. U. S. S. R. [Kurland]: Lesten, F. Bucholtz (Farlow); Kuruka, Ternomorek, F. Bucholtz (Farlow) ; Moskva, Ocfalfsbebe, F. Bucholtz (Farlow) ; Mikhailovskoe, Bucholtz, Aug. 4, 1899 (Farlow); Ucrania, Czernaiev (Upsala). SwEDEN: Upland, Upsala, Karolinaparken, E. P. Fries, 1853 (Upsala); Th. M. Fries, 1877, 1883 (Upsala) ; Seth Lundell, 1930 (Upsala and Dodge) ; Harry Smith, 1930 (Upsala and Dodge) ; Valby, Lidehall, C. J. Cederstróm, 18-10-91 (Upsala) ; Skáne, Ramlósa, E. Melin (Upsala). AUSTRIA: Tirol, Füssen, J. E. Weir 22,291 (Dodge). GERMANY: Würzburg, E. Soehner 1033 (Soehner and Dodge). SWITZERLAND: Neuchatel, P. Konrad (Lloyd Mus. 050, and Dodge); Genéve, Miolan, J. Müller (Lloyd Mus. 05915, Dodge, and Stanford, ex herb. G. Winter). [ Vor. 23 648 ANNALS OF THE MISSOURI BOTANICAL GARDEN Iraty: Como, Rodero, O. Mattirolo, Aug. and Oct. 1900 (Lloyd Mus. 081, 03709, Dodge); near Milano, no collector (Stanford, ex herb. G. Winter) ; sett G. Bresadola (Upsala). FRANCE: Alsace, L. Quélet (Paris); ae Andelys, G. Malengon (Dodge) ; Loiret, Port pres Gien, L. Tulasne, Oct. 3 (Paris) ; Seine-et-Marne, Vincennes, lasne (Paris) ; m L. Mis pri ENGLAND: Wiltshire, Rudloe, C. E. Broome, Oct. 11, 1842 (Brit. Mus., and Brown Univ. 7 T e. Berkeley, dio. Fungi, 285. : Ann Arbor, L. E. Ekvall 19 (Univ. Mich.) ; Saginaw Forest, C. H. Kofman. u (Univ. Mich.) ; School Ge Glen, A. H. Smith (Univ. Mich.). INDIANA: Lafayette, Happy Hollow, H. S. Jackson (Dodge, and Zeller 2806). OREGON: Benton County, Corvallis, S. M. Zeller 1787, 1898, 2242 (Zeller) ; Wren, L. M. Boozer (Zeller 855) ; Lincoln County, Waldport, S. M. Zeller 4996 (Zeller). 7. MELANOGASTER RUBESCENS (Vittadini) Tulasne, Fung. Hypog. 96. 1851. Octaviania rubescens Vittadini, Monogr. Tuberac. 18. 1831. Hyperrhiza rubescens Rabenhorst, Deutschl. Krypt.-Fl. 1: 293. 1844 Bondatia oleodora Lespiault, herb. nom. Melanogaster nauseosus Coker & Couch, Gast. Eastern U. S. & Canada, 39. 1928 Type: in Broome Herb. at Brit. Mus., Berkeley Herb. at Kew, and Tulasne Herb. at Paris. The type of M. nauseosus from Univ. North Carolina Herb. Fructifications subglobose to elongate, irregularly sulcate, 1.5-3 cm. in diameter; surface tomentulose, even to reticu- lately ridged, light nal when fresh, p raw umber or darker; fibrils absent; peridium spongy, stupose, homoge- neous, darker brown toward outer surface, drying 100-200 y thick, distinet from tramal tissues; gleba moist, bluish-white when fresh, reddening and yellow ng when D eut, drying brownish- TM ; septa broad, variable, einnamon-buff ; cavities rounded, not crowded at first filled ah dark claret ionn or darker spore masses, later becoming hollow, due to the rela- tively few spores scattered through the cavity gel; basidia clavate, unevenly distributed on the branching hyphae of the cavities; spores oblong-fusiform to ovoid-ellipsoid, or curved, somewhat irregular (often the shape of the fruit of Arachis hypogaea), 8-9.6 (-11.8) x 5.5-6.5 n, light yellow, pellucid. Southwestern Europe and eastern North America. J uly to November. 1936] ZELLER & DODGE—MELANOGASTER 649 The specimen from Italy collected by Mattirolo differs from the other material of this species in having long narrow spores which measure 11-13 x 3.5-4 y. It is with some hesitation that we have referred this specimen to M. rubescens (Vitt.) Tul. The specimen from Ithaca, in alcohol, shows a peridium 650- 750 » thick, drying as thin as the peridium in the type. IrALY: Como, Rodero, O. Mattirolo, Oct. 1905 (Dodge) ; Lombardia, near Milano, C. ittadini, type (Paris, Kew, and Brit. Mus). Lot-et-Garonne, Nerac, Lespiault (Bondatia oleodora herb nom., Paris). imi York: Ithaca, Beebe Lake, H. H. Whetzel (Atkinson Herb. at Cornell 13679, Dodge, and Zeller 1446). PENNSYLVANIA: Mt. Pocono, Schweinitz 2210 (sub Ehizopogon aestivus in Curtis Herb. at Farlow). NORTH CAROLINA: Gët County, Chapel Hill, J. N. $ E. R. Couch (type of M. nauseosus, Univ. N. C. 8281); Watauga County, Blowing Rock, Coker et al. (Univ. N. C. 5804). ALABAMA: Sepsly, Payne’s Mills, P. P. Payne (Lloyd Mus. 041). 8. MELANOGASTER Parksii Zeller & Dodge, sp. nov. Type: Univ. Cal., Dodge, and Zeller Herbaria. Fructificationes EECH vel subsphaericae, 2-4 cm. diam- etro, fibrillae basi prominentes, brunneae nigrescentes, siccatae * Dresden brown?” vel “Mars brown,’’ obscuriores contusae, superficie aequo vel undulato, tomentuloso vel glabro; perid- ium prosenchymaticum, cellulis magnis spongiosis contextum, dilute luteum, externe obscurius, 300-480 » crassitudine; gleba ‘‘Brussels brown’’ vel ‘‘warm buff”? siccata, locellis in centro fructificationis subcavis; basidia non visa; sporae late obovoideae vel subsphaeroideae, dilute brunneae, sub lente pel- lucidae, 8-9.6 x 6.7-7.7 p. Fructifieations irregular to subspherical, 2-4 cm. in diam- eter, dark brown to black, drying Dresden brown to Mars brown, darker where bruised, surface even or undulating, tomentulose to glabrous, basal fibrils prominent; peridium spongy, prosenchymatous, of large cells, light yellow, darker toward the peridium, 300-480 + thick; gleba Brussels brown [dry], central cavities more or less hollow when dry with whit- ish or warm buff, gelified septa; basidia not seen; spores broadly obovoid to subspheroid, light brown, pellucid, 8-9.6 x 6.7-7.7 »; odor pungent, spreading. [Vor. 23 650 ANNALS OF THE MISSOURI BOTANICAL GARDEN Deeply buried under leaves of Quercus agrifolia. Califor- nia. April. CALIFORNIA: Santa Clara County, Call of the Wild, H. E. Parks 136a (Univ. Cal, Dodge, and Zeller 1412) ; Guadaloupe Mines, H. E. Parks 136, type (Univ. Cal., Dodge, and Zeller 1655). 9. MELANOGASTER MOLLIS Lloyd, Myc. Notes 65: 1047. 1921 (see also p. 1065). Illustrations: Lloyd, Mye. Notes 65: f. 1939. Type: In Lloyd Mus., Dodge, and Zeller Herbaria. Fructifications subglobose, collapsing at maturity, 2.5-3 cm. in diameter, drying Dresden brown to mummy-brown, pruinose to smooth where rubbed, distinet fibrils below ; peridium thin, 80-130 » thick, simplex, stupose, hyaline with brownish ex- terior; gleba in general black, drying hollow, tramal tissues whitish, then gelatinizing, prosenchymatous, basidia-bearing cavities filled with the deliquescent hyphae, basidia, and spores; spores small, dark brown, ellipsoid, smooth, 5.5-6.7 x 2-3.7 »; odor disagreeable, rancid. In black vegetable humus, hypogeous or emersed, among greasewood. Melanogaster mollis Lloyd (except for M. microsporus ) has the smallest spores of any known species. In the type collec- tion the fructifications are extremely mature, with the gleba becoming deliquescent, drying with large hollows. Wyoming: [Fremont County, Meeteetse, 6700 feet alt.], Simon Davis, type (Lloyd Mus. 041, Dodge, and Zeller). PORTUGAL: J. Rick (Lloyd Mus. 05348, and Dodge). 10. MErANOocasTER DURISSIMUS Cooke, Grevillea 7: 94. 1879. Type: India, Chakrata, 8000’ north of Delhi, Baden Powell. Fructifications compressed, suleate, subspherical to reni- form, smooth, up to 4 em. in diameter, drying 2x 1x1 em., very hard, dark fuscous to sepia, cottony-furfuraceous, rubbing glabrous, no fibrils; peridium stupose, 90-115 y thick, amber; gleba blaek with buff-white septa of close gelified pros- * Lloyd did not reeord the data on this specimen, and the late Mrs. E. B. Blackford assured Dodge that she has been unable to locate a duplicate in Simon Davis’ collee- tions but she found notes for Calvatia defodiodis Lloyd from this locality, and it is probable that all of his collections from Wyoming came from the same locality. 1936] ZELLER € DODGE— MELANOGASTER 651 enchyma; basidia evanescent; spores 6-8 x 4-5 p, ellipsoid, very truncate at base, smooth, dark brown, exospore not conspicuous. British India. November to February. Baden Powell, the collector of the type, notes that this species is cooked and eaten by the natives, who report that chewing a little piece keeps off thirst in crossing high passes. The na- tives report it common in the hills of Kangra and in the Upper Chenab valley. BRITISH INDIA: Chakrata, B. Powell, type (?); Simla, J. M. Clark, 7 Feb. 1879 (Brit. Mus. and Kew); J. H. Burhill, Nov. 8, 1902 (Lloyd Mus. 7331, com. E. J. Butler 14945). 11. MELANOGASTER VARIEGATUS (Vittadini) Tulasne, Ann. Sci. Nat. Bot. IT. 19: 377. 1843. Octaviania variegata Vittadini, Monogr. Tuberac. 16. 1831. ?Tuber moschatum Bull. Champ. 1: 79. 1789. Hyperrhiza variegata Rabenhorst, Deutschl. Krypt.-Fl. 1: 293. 1844.—non al. Illustrations: Vittadini, Monogr. Tuberac. pl. 3, f. 4; Bull. Champ. 1: pl. 479. Type: della Pistola and along R. Lambro near Milano, Lom- bardia, Vittadimi. Specimens sent by Vittadini in Sprague Collection at Harvard Univ., in Berkeley Herb. at Kew, and in Paris (com. Ardissone). Fructifications rounded, irregular, the size of a hen’s egg or fist, drying 2 em. in greatest diameter, smooth, ochraceo-fer- rugineous drying Verona brown, fibrils basal, fuscous; perid- ium stupose, 100-160 » thick, amber-yellowish, extending as fundamental tissue into the trama; gleba yellowish-ashy at first, becoming black, with thin yellow-orange septa, cavities subpentagonal; basidia evanescent; spores ellipsoidal, very dark brown, 7.5-10 x 5-7.5 », smooth. Under Quercus and Cistus monspeliensis. Northern Italy, Algeria, and California. August to December. Itis with some hesitation that we have referred Parks Z22 to this species, owing to its texture resembling Hymenogaster, and to the absence of gel filling the cavities. [Vor. 23 652 ANNALS OF THE MISSOURI BOTANICAL GARDEN ITALY: Lombardia, C. Vittadini, type (Paris, com. Ardissone, Kew, and Farlow) ; agro panormilano, Inzenga (det. Tuber Borchii, Paris). ALGERIA: Alger, Durieu de Maisonneuve (Paris). CALIFORNIA: Santa Clara County, Guadaloupe Mines, H. E. Parks 958 (Univ. Cal. and Dodge) ; Saratoga, H. E. Parks Z22 (Univ. Cal. , Dodge, and Zeller) ; Marin County, San Rafael, H. E. Parks 2059 (Univ. Cal). DOUBTFUL SPECIES MELANOGASTER SARCOMELAS (Vittadini) Tulasne, Fung. Hypog. 97. 1851 Octaviania sarcomelas Vittadini, Monogr. Tuberac. 16. 1831. Hyperrhiza sarcomelas Rabenhorst, Deutschl. Krypt.-Fl. 1: 293. 1944 Type: Lombardia, Vittadini, not seen. Fruetifieations small, irregular; peridiúm smooth, without rooting appendages; gleba very black, uniform in color; cells regular, con- tents gelified, filled with spores; spores ovoid, smooth, somewhat larger. Strong odor of ink. Fungus the size and form of a seed of Faba, with a thin peridium; base hardly distinguishable. Cavities rhomboidal or pentagonal, septa ashy, pellucid, opaque in center. When dry contracting into a solid perennial mass.—Vittadini. A careful study of the original descriptions of Vittadini’s species suggests that M. sarcomelas may be a synonym of M. tuberiformis Corda. In spite of searches by Tulasne, Matti- rolo, and Lloyd, the type of this species has not been found. MELANOGASTER ODORATISSIMUS (Vitt.) Tulasne, Fung. Hypog. 95—96. 1851. Octaviama odoratissima Vittadini, Monogr. Tuberac. 19. H yperrhiza odoratissima Rabenhorst, Deutschl. Krypt.-Fl. 1: 293. 1844. Type: two fructifications from oak woods across the Po near Milano, Vittadini, April, probably lost. Fructifications subspherical, sub-smooth, golden-rufous, with basal appendage. Gleba tough, pulpy, rufous-brown with septa; cavities not conspicuous, never empty; spores ovoid, umber. Odor of Hya- cinthus botryoides. Fructifications the size of a chestnut; peridium with rooting fibrils as in M. variegatus, golden at first, then rufous; 1936] ZELLER & DODGE—MELANOGASTER 653 gleba suggesting the pulp of red beets; cavities not very distinct; spores rare.—Vittadini. Hesse, ‘Die Hypogaeen Deutschlands! 1: 64-65, pl. 4, f. 10— 14, pl. 8, f. 8. 1891, has referred here material collected at Michelbach near Marburg, in 1884, but we have been unable to see this material. He describes it as follows: Fructifications spherical, from size of a large hazel-nut to that of a sparrow egg, golden-yellow becoming deep red-brown, smooth, odor of Hyacinthus botryoides; peridium not easily separable from the gleba, composed of gelified hyphae, becom- ing deep reddish-brown; septa at first golden, becoming red- dish; clamp connections prominent in the mycelium of the cavi- ties bearing the basidia, which are mostly 8-spored; spores almost ovoid [figures show them ellipsoid to almost allantoid], brown, size not given, but if magnification of figures is correct spores about 7-8 x 4-5 y. There seems little to distinguish Hesse’s concept of M. odor- atissimus from M. Broomeianus Berk. BULLARDIA INQUINANS Junghuhn, Linnaea 5: 408. 1830. Argylium inquinans Wallroth, Fl. Cryptog. Germ. 2: 873. 1833. Hyperrhiza inguinans Rabenhorst, Deutschl. Krypt.-Fl. 1: 292. 1844. Illustrations: Junghuhn, Linnaea 5: pl. 6, f. 15. Type: Germany: between Rammelburg and Klaus, F. Jung- huhn, not seen. Fructifications up to the size of a walnut, irregular, gibbous, smooth ; peridium thin, sometimes ruptured at maturity ; fibrils sparse, appressed; gleba black; septa white, composed of branched, loosely woven hyphae; basidia scattered throughout the cavities; spores black. Odor sweet, pleasant, almost aromatic; taste very sweet.— Junghuhn. Under Corylus Avellana L. Germany. September. While there seems no doubt that the above description refers to a species of Melanogaster, it is too brief and generalized to refer definitely to a species. Itis probably not M. variegatus, to which it has been generally referred, as that Mediterranean [Vor. 23 654 ANNALS OF THE MISSOURI BOTANICAL GARDEN species is not known north of the Alps. It is much more prob- ably M. tuberiformis or M. Broomeianus. MELANOGASTER TUBEROSUM (hir Polysaccum tuberosum Fries, Syst. Myc. 3: 55. 1829; Lin- naea 9: 695. 1830. Lycoperdoides tuberosum, ferrugineum, arrhizon, pulpa nigra, Micheli, Nova Pl. Gen. 219. 1729. Hyperrhiza tuberosa Fries, Ind. Syst. Myc. 102. 1832; Sum- ma Veg. Scand. 437. 1849. ef. Th. M. Fries, Svensk Bot. Tidskr. 3: 293. 1909. Illustrations: Micheli, Nova Gen. Pl. pl. 98, f. 2. Type: ‘‘In pratis calcareis,”? Malmoe, Sweden, along with Bovista suberosa. Angelin: Fructifications without roots, round, irregular, often oblong, sug- gesting a small tuber of Solanum, the size of a walnut (Juglans), soft when young, becoming hard; peridium coriaceous, ferrugineous, smooth, almost pruinose under a lens, not dehiscent ; gleba with white or yellow septa, cavities very black, large, round or irregular, pro- ducing a black viscid ink with a very foul odor, filled with spores when mature ; no eapillitium.—F ries. Since M. Broomeanus Berk. is the only species of this genus so far found in Sweden, of which we have seen specimens, it is likely that M. tuberosum may be the same fungus. However, we prefer not to reduce M. Broomeanus to synonymy until we have had an opportunity to study the type, if it still survives. MELANOGASTER AUREUS (Vittadini) Tulasne, Fung. Hypog. 97. 1851. Octaviania aurea Vittadini, Monogr. Tuberac. 20. 1831. Hyperrhiza aurea Rabenhorst, Deutschl. Krypt.-Fl. 1: 293. 844 Illustrations: Vittadini, Monogr. Tuberac. pl. 3, f. 14 Type: in forest on hill near La Stradella, near Milano, sum- mer, C. Vittadini, apparently lost. Fruetifieation oblong, reniform, rooting at the base, peridium smooth, at first whitish. Gleba hard at first, golden, with white veins (septa) variegated, then soft, blaekening, somewhat hollow. 1936] ZELLER & DODGE—MELANOGASTER 655 Odor pleasant. Fungus size of a filbert, peridium thin, becoming fuseous at the touch, with a radiciform appendage. Cavities not numerous, rounded or elongate, suggesting the pores of Boletus, golden-shining within. Septa thick, whitish-citrine. Cut gleba be- coming greenish, but at maturity breaking down into a blackish sporiferous pulp.—Vittadini. EXCLUDED SPECIES Melanogaster Wilsonii Lloyd, Mye. Notes 68: 1176, 1923, is Rnuizoroaox Wilsonii. Zeller & Dodge, comb. nov. Melanogaster carolinensis (Bose) DeToni in Sace. Syll. Fung. 7: 167. 1888. Uperhiza Bose, Ges. Naturf. Freunde Berlin Mag. 5: 88, pl. 6,f. 62. 1811. Uperhiza carolinensis Nees ab Esenbeck, Syst. d. Pilze 1: 159. pl. 15, f. 146. 1816. Hyperrhiza caroliniensis Sprengel, Syst. Veg. 4: 416. 1827. Uperrhiza Boscii Schweinitz, Syn. Am. Bor. 255. 1834; Lloyd, Mye. Notes 30: 395. 1908. Curtis states that the Schweinitz specimen is a bad Scleroderma? from Mt. Pocono, Pa. Type: Carolina, Bosc; probably lost. A careful study of the original description and figures sug- gests Scleroderma sp. but in the absence of the type its identity remains doubtful. GENERAL INDEX TO VOLUME XXIII New scientific names of plants and the final members of new combinations are printed in bold face type; synonyms and page numbers having reference to figures and plates, in italics; and previously published names and all other matter, in ordinary type. A Acremonium kiliense, 40 Acrotheca Pedrosoi, 543, 544, 545, 546 547 Akinetes in Gliocladium, 20 Allescheria, 37; Boydii, 38, on various media, 20, 56, 58 Almeida Floriano Paulo de, Morris Moo oe: a Ria of Chrom- mr SE, 472 Anderso Edgar study experimental P" hybridization in the genus vr Hybridization in ias, 511; Th pitting in tracheids d M ee 151 go apr e Studies IV. The American pipa of Hehiteideae, 169! Apo bun An experimental study of hybridization in the genus, 159; re- of e d tests with certain eet rosaemifolium, 159 161, cannabinum, 159, 161, 168; hy- pericifolium, 159; medium, 159, 161, 168; venetum, 163 quites Tai, 4 450 — 562; Calverti, 562; Kohli, Areangelielia and Macowanites, Elas- Arcangelella 602; africana, 614; al- veo : asterosperm 6 5 635; el rii, caudata, 616; Borziana, Ei SN 617; Campbellae, 613; ; eremea, ; Cu rtisii, be 610; Gardneri, 611; gla- brella, 608; krjukowensis, 626, var. michailowskjana, p laevis, 579; luteocarnea, 570; nana, 609; occi- dentalis, 629; pil 61 ; Ravenelii, 622; rosea, 630; scissilis, 624; semi- , 617; sociali ; ' goder stroemii, 630; Stephensii, 613; tas- manica, 624; violacea, 608; vulvaria, Argylium, 640; inquinans, 653; liquam- Arthurella, 559; Corringtoni, 559, 564 AE in Hyphomyeetes, Aspergilleae, conidial production in, 5 Asterogasteraceae Azaleastrum Warrenii, 453 B Baker, Gladys genus Helicogloe Baragwanathia longifolia, 153 Blakeslea RE Thaxter, The effect of environment upon the production = 2A id sporangiola in, 527, E. A study of the Segen, 138; Flareri, 143; Pinoyi, 145 Blood agar, pe m. yiii of Syrin- spora inexorabilis on, 134 go Blood serum; eroi wth di cert ain Hypho es on, - : Syringospora in- n ilis on, Bondatia Miri - Botrytis, 544; spectabilis, 46 544, 545, 550; monophora, Botrytoides 544, 546 Bullardia ki, 653 Bullardia, 640 C Campsotrichum, 546 Candida ep SEKR 145; ins , 142 Carbo nate pie ow of Syringospora inexorabilis, 135 145; desidiosa, 1 The taxonomic entities in this work are indexed on pages 407-424 of this volume of the ANNALS and will not be repeated here. ANN. Mo. Bor. GARD., Vor. 23, 1936. (657) 658 val pei les, 5 - rro t plug, m of certain Hypho- ch See 8, d in Hyphomye bene e 50 Cephalosporium and some ted Hy- phom A c comparative pipa of eo nidial Ce Cephalosporium, 38; Acremonium, 40; various var ; SEN 40, on various media, 20, , 68; sym mbio ticum, 42, on various 8 res in Hyphomycetes, 19 Chondriosomes in riis N inexor- abilis, 137 Chromoblastomyeosis 543 , New los impu of, 543 versi- niea, 474, aside 76; in de Sen I. virginica var. era of, adochytriaceae, New ger 553 vulvarius, Cla Clathrogaster Beccari, 635; Giele 43; Pani we Ge on ious media, 20, 56, 8, var. Ke Zë 44, on various “nie 20, 56, 64, 68, d rosea, rious media, 20, 64, 68; T D n on various GE 66 Colorado, A new Sequoioxylon from, 439 Conehyliastrum, 560 ves ri, 557, 560, 564; Merritti, 561, Conidial kk Ze dig some Hypho- Conia EE in Cephalosporium ome related ee A EE study o Corn-meal pag d S certain Hy- phomyeetes on, 20; of Syringospora inexorabilis 133 odes 556; Petersoni, 557, 564 ryptococcus or Uu o agar: ae owth ge? certain Hy- phomycetes 20, 68; of Syringo- spora herteabilis on, 132 D Dalea Grayi, Gg laevigata, 450; Thompsonae, Dermatitis Geng 543 [Vor. 23 ANNALS OF THE MISSOURI BOTANICAL GARDEN and Morris Moore. gy and cytology Dodge, Carroll W.: Morphology, sin apr of Sy oe (Mon- ilia bless Se te s), 129 d Sanford M. Zeller, Hydnangium por related enera, 565; Sanford M. Zeller and. Elasmomyees, Ee and Ma- ; radicatus, 550, ei Downingia hie dud 455 po parr var. i 52; i52 Ber in tomentosa, erianum, tomentosa, 452 E p of Apocynaceae, The Amer- Arcangeliella, and Ma- ces, 4 0; borneensis, 600; quie 601; Mattirolanus, 600, 602; micha iowskjanus, 627; russu- loides, bei pide ndo ?s owth s hls on inexorabilis Ge Blakeslea trispora on, ez? m. Env mr The e produetion of s Veiis and dis doe in Blakeslea trispora Thax- ter, 527 Evolutionary patterns of the genus Iris, 499 F Field and herbarium studies. IV, 447 Flags, northern blue: intra- ie gins dif- ferentiaton in the, 485; pun ogenetic Fp onships in, 499, 501 ; taxonomy the, 459 Florissant, Colorado, fossil trees at, 439, Fossil n n m effect of, on dër Au types in Blakeslea re Fuchsin, eur effect n on sporangial types 1 Blakesle en trispora, 537 Fungi INR eti and sexuality, 31 G Gautieria, 611 Glioeladium, 46; deliquescens, zi on Muro media, 20, 62, 68; fimbri- on various media, 20, Ee 68 ; penile, 48, 56, 62, 68; roseum, ; Vermoeseni, 4T, on various media, Dy 64, 68 1936] INDEX EE pf effect of various pH gro MM E^ rtain Hy- phomycetes on, 20, of Syringospora inexorabilis 134 Glyeogen in nier pura inexorabilis, EE David. The effect of environ- pon the produetion of spo- Mr and E in Blakeslea i , 927 in | Hyphomyeetes, ng soi, 544, e medium, E of certain e ie eg on, Granules, occurrence of, in certain Hy- mycetes, 35 603; flavus, 625; Gard- FI palih, 620; ? "solidus, 620; seminudus, 617 Gymnomyces vesiculosus, 634 H Heli fasion A pce of the genus, 69 Helico oglo P considerations, 88; pinicola, 89, mor- phology of, 126, T alniviridis, 90, Herbarium studies, Field and, 447 Hetero genge oou of, to origin of Hyphomyce EE EE, 451; te- nella, Hormodendroides, 544, 547, 550; Pe- drosoi, 544 ciens = Kë 550; compactum, 544; edrosoi, 545, 544, ED 547 Hormodendrum compactum , 049 ipd ig t of, on sporangial types ea trispora, 535 e sraon in e Am escantias, 511; er SC genus Apocy- expe erimental study of, 159, erican Trad- s, 474 Hybrids, method for uos UNE specie, CUL LUPTATUM 565 Hydnangium and related genera, 565 Mosa. 574; album, 581; Areheri, ; auran 586; tium, 8; australiense, 625; brisbanense, 626; ndidum, 570; earneum, 575, 582, var. pur- pureum, 584; carotaecolor, 586; ca- 659 oe 597; eereum, 595; citrinum, ; Clelandi, 596; compactum, 580; pisces 578; den nsum, 593; Eis- enii, 578; galathej jum, 614; Gilkeyae, 581; glabrum, essei, 590; 617; microsporium, neo; mistiforme, 8 monticola, ele ee "ang p Ng ion Aer itam of -— of, on sporangial types eg trispora, 539; relation Kë os growth of certain Hyphomycetes, of Syringospora inexorabilis, Ka ces 640; Behrii, 615; Cam bellii, 613; nanus, 609; gaere 608 Hymenogasteraceae, ve Hyperrhiza, 640; aurea, 654; carolini- nsis, 655; UE sid 6 3; liqua- minosa, -— odoratissima, 652; rubescens, 648; sarcomelas, 652; tu- beriformis, 643; pori 654; vari- egata, 651 Hyphomycetes : omparative study of Ac conidial formation in some, 1; conid- atus in, 5; pathogenicity of, 34; physiology of, 20; sexuality in, 31; taxonomy and ’ morphology of, 37 Hipo terrestris, 98 I Intra-speeifie differentiation in the 485 northern blue flags, ide pié orig nary zer s of the genus, 99; intra-specific differentiation in, 485; phy ylogenetic relationships of, 471; The species problem in, 457; LE of, 459 Iris, 457; arcti ica, 460; carolina, 464; caroliniana, 464; fo etidissima, 500; versicolor, 457, 463, - ber , 474, 476, pao rela- 660 ANNALS OF THE tionship of, 471, range of, 464, seeds ; 4, chromosome eupied by, 487, chromosome numbers in, 474, 475, ower of, 486, range of, 465, seeds of, 4 L Lactose agar: growth of certain Hy- homycetes on, 20, of Syringospora oe , 184 Laubert Leu cophichs EEN 596 Light: of, on sporangial eis in Blakes eslea Oz 539; polarized, study of certain Hyphomycetes under, Lycoperdoides tuberosum, ferrugineum, arrhizon, pulpa nigra, 654 Lycoperdon Gunnii, 596 Lycogalopsis, 565 M ae 573; earniea, 573, 574 Macowania, 636 Macowan nia, 636; agaricina, 636 Macowanites, Elasmomyces, Areangeli- ella and, 599 ian 636; agaricinus, oid 636; echinosporus, 601; magnus, Malt extraet agar, oder of pror spora inexorabilis 32 Martellia, 575; Measuring speci es 1 Beete, 575 hybrids, a dod for, 5 Media: effeet of, upon sporangia pro- duction in Blakeslea trispora, 529, 531; growt wth of eertain Hyphomycetes on various, 20, 56, 68, of Syringo- pora inexorabilis on, 132 Medica aspects of certain Hypho my- tes, 34; of Syringospora inexor- abi lis, 129 ee 639; Fu 042, var. sper 643; edi z rosporus rospor Co 46; mollis, is ee 648; odoratis- imus, 652; Parksii, 649; rubescens, 648; sarcomelas, 652; tuber eriformis, 645, 652; tuberosum, Ch variegatus, 651; variegatus var. Broomeianus, 647; Wilsonii, 655 [Vor. 23 MISSOURI BOTANICAL GARDEN Mieroseris Aere 455 Mo d effeet of, on sporangial types Blakeslea trispora, 535 Monilia inexorabilis, 129, 139; mace- doniensis, 141; pu 144 ne dicus een and H N. An- ransitional pitting in ahi "of Psilotum Moore, Morris, Car B dod nd. Morphology, rd E and cytology of Syringospora inexorabili is (Monilia leed 129 aulo di Almeida, Che EE e GE physiology, and cytology of Syringospora we lis (Monilia inexorabilis), 129 Myceloblastanon macedoniense, 141 EAEEREN 138 Mycoderma desidiosum, Zoe GE E 139; di- orpha, 140; Pana An 140; cedoniensis, 141; rc we icillata, 143; zeylanoides, i 138; ma ge 141; trimorpha, 1 e and Floriano Now organisms of N Nemophila ks Auri 454; parviflora, 454; petrophila, Nitrate ae gro iwi of various Hy- phomycetes on, eg Nourishment, effect of, on sporangial types n Blakeslea trispora, 535, 542 Nutrie dy agar and broth growth of i ie i de inexorabilis on, 133, Nu dd relati of, to growth of certain TAG TAIN. 26 O Octaviana, 603; africana, 614 Cae: 575, 602, 608, Ded. africana, 614; veolata, 620 mbigua, 642; pre Zeg 586; Aye ad 575, 631, var, depauperata, 634, var. hololeuca, 634; erospora, 63 32; aurea, 654; pide ra 625; borneensis, 600 ; brunnea, 632; brunneola, 617: carnea, Gei eer ies citrina, 577; mpacta, 569, 573; galatheja, 614; is, 583; monospora, 7; monti- cola, 578; " mutabilis, 591, 632; oc- 1936] INDEX cidentalis, 629; donada 652; j R ; Stillingenri, 630; nic a, 593; Cé 634; variegata, ange se 603, alveolata, 320; Arch- asterosperma, 63 eri, 58 ; brunnea, 632; compacta, 573; Hessei, RE laevis, 579; lanigera, 591; 583; mutabilis , 632; rosea, 630; goe phensii, 613; tu berculata, 593 Oidia in Hyphomyeetes, 20 Oidium albicans, 138 Oxygen relationships, relation of, rowth of certain Hyphomycetes, 25 E Paeonia Brownii Paramoeciella, 568; > Gambia, Parendomyces, Flareri, 1 Parosela Grayi, 450; Pathogenic fungi, 3 34, Pedieularis ao castis 455; Hallii, 450; riam icaule, 49; 558, 564 Thompson, 451 roseum, aT Penstemon aridus, 455; Caryi, 455 yis caen lea growth o of Blakeslea tri- spor 530 ie Wer ius sizes in species of Iris, 8, 489, 490, 491, 492, 493, 505 6 n H mycetes, 1 Hiel 544, 547, GE E 549; Guggenheimia, 544, Phialophora 544, 547, 550; macrospora, 52; verrucosa, 543, 544, Phi genotis relationships; of I dorrego? and Iris vir ginica, 471; of oeynum, 165 Picea pungens, 447 Pigments in certain Hypnomyeetes, 28 Pinkerton, M. Elizabeth. - tive study of coni Cephalosporium and some related Hy- phomycetes, 1 deo transit onal, in ka? be Psilotum, 151, 158 tracheids of Ru Pearsalii, 440, i der light, study of certain Hy- phomyeetes under, compa idial formation in 661 Pollen fertility in species of Apocynum, 161 Polysaccum tuberosum, 654 Populus taccamahacca, 449 Potato decoetion, i of Blakeslea trispora on, 533, 537 Potato dextrose, Gs) of Blakeslea trispora on, 530, 532, 537, of certain , of Syringo- PL - Ai 471, 472; Lu 471, 472; v icillata, 471, e be testa i e om cynum, Aim of, 160, Anibal verticillata, 143; zey- lanoides, 14 Psilotum, transitional pitting in tra- cheids of, 151, 156, 158; nudum, 152, 156, 158; triquetrum, 152 R Ranunculus jovis, 450 Raulin’s solution n, dr of Syringo- spora inexorabilis o Rhizopogon Vibe au] 646; Wilsonii, I Rhizopogon aurantius, 588 Rhododendron Warrenii, 453 "EL mucilaginosa, 49, var. ru- bro Eu eases Së S ET of gier 20, of Syr spora inexorabilis on, 129, 133 ; 04 aroliniana, Sabouraud's agar: spora, var liniana, 92; 89; sebac ;» var. vulgaris, 93, var A 93 Saccogloea, 88 Salix careta 448; cascadensis, 448; Tweedyi, 448 Mb. 566; Broomeianus, 569; — 570; e compactus, 573 ; byster- EE es, 569; lanatus, 568; lanatus, 571; liospermus, 572; orde 572; pacificus, 568; 570; minor, siculus, 571 Secotium sect. Elasmomyces, 6 krjukowense, 626; Ma ttirolanum, 602; michailowskianum, 627; sessile, 6 Senecio, 456; yc p Fremontti, 456; Harbou rii, 456; spartioides var Fremontii, 456 662 ANNALS OF THE Sepal and petal width in yh of Iris, 488, 489, 490, 491, 492, , 605 Sequoia, 439 ; Banton: 24) ; Semper- 44 Pals pe Bien A new, from Florissant, Colorado, 439 Sequoioxylon, 439; Pearsalii, 440, ana- omieal drawings, 441, 444, 446 rap stump, ra mperžecti Er 31 e angia and ioni pros The effect of Enron en u the produetion of, in Blakeslea jac MR Thaxter, 527, 542 Stephanospora, 597; earotieolor, 597 d in the A poeynace ae, IV. Amer- n genera of Echitoideae, 169 Sugars, effec ous, on growth of Syr ingospora Geier 135; upon eg e production in Blakeslea tri- spora Surface. ae in Hyphomycetes, 10; relation of, to growth of certain Hy- hom cetes, 24 Syringospora inexorabilis (Monilia ma ipit. bm E 139; Braulti, dimorpha, 140; inexorabilis, 139! eytology of, 135, gro various media, 148, 150; Robin, ' 138 E Bee dk Temperature, effect of, on sporangial types in sehe DEM 538 Thallophytes , 553 E 450; P ABRIR 450; ar- , 450; rhombifolia var. annulo- ben mg 450 Thornbera Grayi, 450 To i Merde , 4€ ici ; 49: mueilaginosa, 1, 49, 60; Trachelds: ei Sequoioxylon Pearsalii, 440, 446; of Psilotum, transi- ere Sting in, 151, 156, 158 Tradescantia : canaliculata, 515, dis- n De 518, 519, Suge betw irginiana and, 515 Geen tributis of, 518, ' 519, [ MISSOURI BOTANICAL GARDEN Vor. 23, 1936] Vi emo between, and T. cana- lieulata, 515 Tradescantias, Hybridization in Amer- Trichosporium Pd Meu ptio har 546 547; Pedrosoi, 543, 6, 547 ` Trichosporon etal dl Tuber "moschatum, 647, 651 U Uperhiza, 640, 655; carolinensis, 655 U perrhiza, 640; Boscii , 655 V Vacuoles: in Hj Lampe baggy 8; in Syringospora inexorabilis, 1 37 Syringospora inexorabilis, Volutin in 137 W Williams, Louis, O. D and herba- rium studies, riw Jo t E. EN £ Studies in His . The American Gen Wort agar, Sé of Syringospora in- SE on, 132 X Xerophyllum tenax, 449 Y Yeasts, imperfect: systematic pu of, Yeast-dextrose agar, En of Syringo- spora inexorabilis on, 134 phar tt of, 135; Z pricing deg rettii, 452; latifolia var. Garret Zebrows ki, "de D New genera of Cla- and Car Dodg Elasmomyces, Lech and Manitas 599; Melan ra, ca asin Pe dermatitidis, 129 AC mparative Study « of Conidial Formation dot che ecto an ad: S ome Related 1-68 EN ES wes Helicogloea. a eee te n E dys E. Baker d BY THE er ‘OF TR En S E FEBRUARY, 1936 EE EE M d s Pinkerton $ | S es Physiology and Cus of tis rospo: ‘a EE Sr Lee Ee Eer W. Pipe and Nor Moore 129-150 x i Eg ^2 had > 2 T EUNT (e T gs BLISHED QUARTERLY AT FULTO N, MISSOUR yea OF THE MISSOU RI TCO GARDEN ST. T. LOUIS, MISSOURI. ` tered as s second-class matter ‘at the post at-office at "Fulton, Missouri, ` er the Act of Muse 3, SD Annals of the Missouri Botanical Garden A Quarterly Journal containing Scientific Contributions from the etg Botanieal Garden and the Graduate Laboratory of the Henry Shaw School of Botiny of Washington University in affiliation with the Missouri Botanical Garden. bun The ANNALS OF THE MISSOURI BOTANICAL GARDEN appears four times during the ealendar year: February April, Sebiéber, and November. Four numbers eonstitute a volum Subscription ve dtr $6.00 per volume Single Numbers.......... 1.50 each* Contents of previous issues of the ANNALS OF THE MISSOURI BOTANICAL GARDEN are listed in the Agrieultural Index, published by the D. W. Wilson mpany. * Except No. 3 of Vol. 22, containing 384 pages, price $5.00. STAFF OF THE MISSOURI BOTANICAL GAEDEN Director, GEORGE T. MOORE. Assistant to the Director, Kar NE H. LEIGH. HERMANN VON SCHRENK, EDGAR ANDERSON, athologist. Geneticist. JESSE M. GREENMAN, RoBzRT E. WOODSON, JR., Curator of the Herbarium. Research Assistant. Ernest S. REYNOLDS, Davi C. Y'AIRBURN, Physiologist. Research Assistant. CARROLL W. DODGE, Neti C. HORNER, Mycologist. Librarian and Editor of Publications. BOARD OF TRUSTEES OF THE MISSOURI BOTANICAL, GARDEN President, GEORGE C. HITCHCOCK. Vice-President, DANIEL K. CATLIN oe oe ALBERT T PERKIN L. Ray CARTER. GEORGE 7'. MOORE. SAMUEL ©, DAVIS. EUGENE PETTUS. THOMAS S. MarrrrT. ETHAN 2.. H. 3HEPLEY. FRED G. ZEIBIG. EX-OFFICIO MEMBERS: GEORGE R. THROOP, BERNARD F, DICK SE of Washington Mayor of tle "us ot St. Louis. ROBERT J. TEREY WILLIAM SCARLETT President o! hiss aig of Bik Bishop of the Diocese of Missouri. nee of £t. Louis, JAMES J. FITZGERALD, President of the Board of Education of EL Louis. GERALD E. UrRICI, Secretary.