SYSTEMATIC INVESTIGATIONS IN PHANEROGAMS, FERNS, AND MOSSES ADVERTISEMENT. The United States National Herbarium, which was founded by the Smithsonian Institution, was transferred in the year 1868 to the Department of Agriculture and continued to be maintained by that department until July 1, 1896, when it was returned to the official ~ custody of the Smithsonian Institution. The Department of Agri-—~. culture, however, continued to publish the series of botanical reports entitled ‘‘Contributions from the United States National Herba- rium,’’ which it had begun in the year 1890, until, on July 1, 1902, the National Museum, in pursuance of an act of Congress, assumed responsibility for the publication. The first seven volumes of the series were issued by the Department of Agriculture. Ricuarp RatTusun, Assistant Secretary, Smithsonian Institution, i charge of the United States National Museum. Il PLATE 1. "NOS TSN “VW HSMSYONY WOINS1dsSy SMITHSONIAN INSTITUTION UNITED STATES NATIONAL MUSEUM CONTRIBUTIONS FROM THE UNITED STATES NATIONAL HERBARIUM VOLUME 16 SYSTEMATIC INVESTIGATIONS IN PHANEROGAMS, FERNS, AND MOSSES WASHINGTON GOVERNMENT PRINTING OFFICE 1912-1916 NOTE. The 14 parts of volume 16 of the Contributions were issued as follows: Part 1, pages 1 to 24, February 13, 1912. Part 2, pages 25 to 62, June 19, 1912. Part 3, pages 63 to 108, July 6, 1912. Part 4, pages 109 to 196, February 12, 1913. Part 5, pages 197 to 224, February 11, 1913. Part 6, pages 225 to 238, February 21, 1913. Part 7, pages 239 to 242, April 10, 1913. Part 8, pages 243 to 254, May 14, 1913. Part 9, pages 255 to 262, June 6, 1913. Part 10, pages 263 to 276, December 13, 1913. Part 11, pages 277 to 286, December 13, 1913. Part 12, pages 287 to 298, August 23, 1913. Part 13, pages 299 to 308, December 29, 1913. Part 14, pages 309 to 371, February 10, 1916. IV PREFACE. Volume 16 of the Contributions consists of 24 papers originally published in 14 parts. The first of four papers constituting part 1 is a discussion, by Mr. William R. Maxon, Associate Curator of the National Herbarium, of tho systematic standing of Asplenium an- drewsii with a reprint of Professor Nelson’s description. The second, much longer one consists of a report, by Dr. J. N. Rose and Mr. Paul C. Standley, Assistant Curator of the National Herbarium, on a collec- tion of plants chiefly from the region between the Pinacate Moun- tains of Sonora, Mexico, and Tucson, Arizona. The plants were obtained by an expedition directed by Dr. D. T. MacDougal, of the Desert Laboratory of the Carnegie Institution. Determinations of the plant species collected are given, together with descriptions of a few which proved to be new. In the third paper Doctor Rose describes a new cucurbitaceous plant from the vicinity of Tucson, Arizona, and bases upon it a new genus, Tumamoca. The last paper, by Mr. R. S. Williams, of the New York Botanical Garden, consists of descriptions of several new mosses collected in Panama by Mr. Maxon, with notes on other interesting species. The second part consists of a third installment of the results of Mr. Maxon’s studies of tropical American ferns. The most impor- tant portion is a monographic treatment of the North American species of Hemitelia, subgenus Cnemidaria, a group which has received little attention from fern students for many years and the members of which have been determined generally under wrong and loosely applied names. With an ample amount of material at hand the problem became chiefly one of taxonomy. That it is possible to offer a solution of its difficulties is due in large measure to the courte- ous assistance extended by the curators of several Kuropean herbaria. Part 3 is devoted to a monograph of the North American species of Nymphaea, the work of Mr. Gerrit S. Miller, jr., Curator of the Division of Mammals, in the National Museum, and Mr. Standley. It contains descriptions and illustrations of all the species of yellow pond lilies known at present from the New World. Heretofore there have been recognized in this area about half a dozen forms; the number in the present paper is 19. Most of the new species come from the Southern States, mainly from the coastal plain of the Gulf of Mexico. The work was initiated by Mr. Miller about ten years ago and was carried far forward by him, but owing to interruption by other duties he has availed himself of the assistance of Mr. Standley Vv VI PREFACE. to complete the undertaking. The history and method of the work are noted in an introduction by Doctor Rose. The next paper (part 4), by Mr. E. O. Wooton, of the United States Department of Agriculture, and Mr. Standley, contains descriptions of many new plants from the State of New Mexico. These have been discovered by the authors while engaged in the preparation of a report upon the flora of that State which is designed for future publication as a volume of the Contributions from the United States National Herbarium.!| That a large number of unde- scribed plants have been detected in the progress of the work has resulted not only from the neglect of this area by most recent students of systematic botany, but also from the fact that there are now assembled for the first time fairly complete collections from all parts of the State. In addition to the diagnoses of many new species and one new genus, the publication of a number of new names is found necessary. There follow seven short papers (composing part 5), all of them systematic studies. The first four of these are by Prof. Charles V. Piper, of the Department of Agriculture, one supplementing his work on the grasses of the genus Festuca, published in volume 10 of the Contributions; a second revising the larkspurs of the group rep- resented by Delphinium simplex and describing a new species; a third settling the application of the name Heuchera cylindrica of Douglas; the fourth presenting miscellancous new or noteworthy plants from the Pacific coast and adjusting questions of nomenclature. The fifth paper, by Mr. William E. Safford, of the Department of Agriculture, describes a new genus of Annonaceae from Colombia, with the species upon which it is based. The sixth paper, by Doctor Rose and Mr. Standley, is a revision of the section N ephromeria of the genus Meibomia, in which nine species, three of them new, are described. In the seventh paper Mr. E. S. Steele, of the National Herbarium, characterizes four new species of goldenrod from the northeastern United States. The next paper (part 6), dealing with certain South American palms, by Messrs. O. F. Cook and C. B. Doyle, of the Bureau of Plant Industry, U. S. Department of Agriculture, is devoted chiefly to descriptions of three new genera of the family Iriarteaceae and the three new species which serve as generic types. All three genera were found in the forests of the Pacific coast of Colombia, near Buenaven- tura, in a region that seems not to have been visited by earlier stu- dents of this group of plants. The specimens, with notes, measure- ments, and photographs, were secured in 1905, when Mr. Doyle accompanied Mr. H. Pittier, also of the Bureau of Plant Industry, 1See volume 19. PREFACE. Vil during a visit of agricultural exploration in Colombia. As much of © the literature of the palms is based on rather fragmentary information, these more complete data, drawn from fresh material, will be appre- ciated. One of the results of the present study is to show the desira- bility of a subdivision of the family Iriarteaccae into three tribes, a synopsis of which is included. Synopses and descriptions are also given of the genera and species of the two tribes in which the new genera are placed. A paper follows (part 7) in which Dr.N. L. Britton and Doctor Rose put on record some results of their joint studies in the cactus family, describing seven new species and revising several old names. As the authors are likely to publish in the Contributions from time to time further results of their work in this field, it has seemed advisable to connect their reports in a series under a uniform title. The succeeding paper (part 8), by Mr. Cook, is in the nature of a continuation of a study on the “Relationships of the Ivory Palms,” published as volume 13, part 5, of these Contributions. The results reached in the former study are applied to the classification of another aberrant genus of palms whose affinities have been misunderstood. In the way of a conclusion the paper gives the outlines of a general treatment of all the American palms. Another paper (part 9), by Doctors Britton and Rose, follows, dealing with Epiphyllum and related genera of the family Cactaceae, the treatment being similar to that of the genus Cereus and its allies in volume 12 of the Contributions, but including in addition South American species. It is parallel also to the preliminary treatment of the tribe Opuntioideae, by the same authors, in the Smithsonian Miscellancous Collections, volume 50, part 4. Two new genera and five new species are described. In part 10 Mr. Safford presents further results of his studies in tropical American Annonaceae, dealing with a subgroup of the genus Annona, which he distinguishes as a new section, Pilannona, with Annona sericea as its type. Ten species are recognized, of which seven are described as new. ‘The older species are redescribed after a critical examination of the type specimens. . In the next number (part 11) Mr. Cook reviews the botanical history of two well-known tropical fruit trees, the sapote and the sapodilla. The following paper (part 12),on the tribes Hauyeae and Gongylo- carpeae of the evening primrose family, by John Donnell Smith and Doctor Rose, is of a monographic character, embodying all that is at present known of these two tribes. The tribe Gongylocarpeae is here first established, as are also two of the genera and one of the species. The authors have availed themselves of the resources of the National Herbarium, and Doctor Rose has had the additional advantage of a field acquaintance with Xylonagra and Burragea in VIII PREFACE, the deserts of Lower California. Of special interest is the unique fruiting habit of Gongylocarpus and Burragea, which bear their cap- sules embedded in the wood of the flowering branches. The two short papers forming part 13 are studies by Mr. Ivar Tide- strom, of the Bureau of Plant Industry, United States Department of Agriculture, of plants in which he has become jnterested., These are considered with reference both to their nomenclatorial history and to their botanical characters. The first paper relates to a well-known fern type, the forms of which are regarded by the author as composing two species. The second treats of two tropical and subtropical weeds which have found their way into the southern United States. In the final paper Mr. Edward A, Goldman puts on record the the botanical results of a biological exploration in Lower California, which was carried out in 1905 and 1906 by Mr. E. W. Nelson and himself, as members of the staff of the Bureau of Biological Survey, United States Department of Agriculture. The paper consists pri- marily of a list of the plants collected, with notes on distribution , to which are added descriptive ecological and economic data and tech- nical descriptions of three new species of oak. A limited use has been made of the collections and records of others. ‘This paper adds mate- rially to our knowledge of an extensive flora, very interesting in its adaptations to diversified conditions of latitude, altitude, and rainfall. Frepericx V. Covi, Curator of the United States National Herbarium. CONTENTS. Tur RELATIONSHIP OF ASPLENIUM ANDREWS. By William R. Maxon...... Report on A COLLECTION OF PLANTS FROM THE PinacaTE REGION oF Sonora. By J. N. Rose and Paul ©, Standley .......----.--- 2-2-2 e eee eee eee rece Introduction............--- cece cece ee ce ee ee cee e tence eee e eee eeeeunaes List of plants, with descriptions of new species. .........+----------++-- TUMAMOCA, A NEW GENUS oF CucuRBITACEAE. By J. N. Rose........-.-.-- New or Interestinc Mosses rrom PanaMa. By R. 8. Williams. ......--- Srupies or TropicAL AMERICAN Ferns—No. 38. By William R. Maxon.... Introduction ...........2 22-2 e eee eee ee eee cere eee eee e een n eee ceeeeeeee The North American species of Hemitelia, subgenus Cnemidaria. .......- Further notes on the West Indian species of Polystichum.........-.------ The American species of Pteropsis........-2..---------ee eee e ee eee eee eee Two unusual forms of Dicranopteris......--..---- cence cece cece cece eees The American species of Cibotium............-.---------- 2 ee eee ee ee eee Two new species of Notholaena..........-..2-----+ 2222-2 eee eee ee eee eee Miscellaneous notes and changes of name. .....-.-------- eee e eee eee eee Tae Norra AMERICAN Species or NympHara. By Gerrit 8. Miller and Paul ©. Standley. . .. 2.22... c cece e ee eee ee eee ee eee ete e cece cence ceeeee Introduction...........2-. eee eee eee eee eee een eee eeeeeeees Note by J. N. Rose. ....-2----- eee ee eee eee eee eee cece e eee eees Authors’ introduction. .......---2.--- 2-2-2 e eee ee eee eee cee eee eee Material studied. .....---------- 222 eee eee ee ee eee ee eee eee eee History of names. ........--------------- ween eee eee eee Names applied to genera. . ....--------------+--- eee e eee eee Names applied to species. ............-+---++--+---+---2- eee Systematic treatment. ...............2- 2-22-22 eee eee ee eee eee eee Bibliography. .........---------+--- OY nnn nn nn ccc c cee nn nen teen naaee Descrietion oF New PLants PRELIMINARY TO A REPORT UPON THE FLORA OF or New Mexico. By E. O. Wooton and Paul C. Standley............-.-. Introduction............. 22-2 eee eee eee cece een eee e eee eeeeeeee Descriptions and new names. . . . -.-------+-ee secre eee eee eee ee eee eens SuppLEMENTARY Notes on AMERICAN Species or Festuca. By Charles V. Piper. . 2.2... 2-22 e eee cece cece eee e eee eee e cece nen eneeeceseeesesercsesces DELPHINIUM SIMPLEX AND ITs IMMEDIATE Attres. By Charles V. Piper.... THe IpENTITY oF HrUCHERA CYLINDRICA. By Charles V. Piper............. New or Norewortuy Species or Paciric Coast Puants. By Charles V. Piper... .---.2 2-2 e cece eee cee cee ce eee rece rece rece tence ee eee ence Tar AMERICAN Species or MerpomiA OF THE SecTION NEPHROMERIA. By J. N. Rose and Paul ©. Standley... . 22-22-22 eee eee eee eee eee ee cece RAmMONDIA, A NEW GENUS OF ANNONACEAE FROM CoLomBIA. By W. E. Safford... 2... eee eee eee eee eee eee eee eee c eee eee e eee e eens Four New Species or GOLDENROD FROM THE EasrerRN Unirep States. By E. S. Steele... cee ee eee een eee ee eeeeee Toren New GENERA or Stitt Patms (IRIARTEACEAE) FROM COLOMBIA, WITH A SynopticaL REVIEW orf THE Famity. By O. F. Cook and C. B. Doyle.. Family characters. ...--..-....202 222 eee eee eee ee cee eee eee ee eeee Survey of the genera..........--...---.---++-----+++- +5 ee Tribal division.................. 22-22 eee eee eee eee ence e ee eeee Descriptions of the Colombian genera and species. .....-..-..-.--.------ Catoblasteae. .. . .-- 2-2. eee eee ee eee eee eee eee Srupies in CactacEAE—l. By N. L. Britton and J. N. Rose..............-- New species.........-- 2-0-0222 eee eee eee ee eee ee eee eee eee eens New binomials..........---.--------- eee eee eee eee eee cece eee e eens x CONTENTS. Page. RELATIONSHIPS OF THE FatsE Date Patm or THE FLoripa Keys, WITH A SyNopTicaAL KEY TO THE FAMILIES OF AMERICAN Patms. By O. F. Cook... 243 The genus Pseudophoenix and the family Pseudophoenicaceae.......... 243 Pseudophoenix an isolated type... ..........0.020202-2-0- 2-0 e eee 243 Analogies with fan palms and true date palms. ...................-.- 243 Alleged relationships with other pinnate-leaved palms. ............. 244 Fruit characters...........00- 22220. e cece cece cece ce eee eee eee ee 245 Pseudophoenix the type of a new family...................2.222---- 246 Floral peculiarities of Pseudophoenix.............2..2..2--000e ee eee 247 The Pseudophoenicaceae and allied families distinguished by fruit characters.........0. 2000222 e eee cece ee ee ce ee eee eee eeeeeeeeee 248 Comparison of Pseudophoenix with the wax palms of Colombia...... 249 Comparison of characters with those of cocoid palms................- 250 Germination characters. .......-..-2-. 2222002220 e cece eee e eee 250 Endocarp characters. ...-. 2.2.22. -. 20.2022 cece cece eee eeeeeee 250 The families of American palms...............2.2.20--222eeeceeeeeeeees 251 Characters and synoptical key...............2.0.2022-0-0-eceeeeeeee 251 Alliances with Old World palms. ...............2..2.0.0-00-0-----0-- 253 THe Genus EprenyLium AND ITs ALLIEs. By N. LL. Britton and J. N. Rose.. 255 Introduction.............02.20 0222 e eee eee ce eee eee e ee 255 Systematic treatment. ......-.-. 2.2.2.2. ee eee eee eee 255 ANNONA SERICEA AND ITs Atiuies. By William FE. Safford. ................. 263 Introduction......... 2.22.0... e eee ec eee cece eee ence eeceeees 263 Systematic treatment. ......... 2.20.20. 0.22 e eee cece eeeee 263 NOMENCLATURE OF THE SAPoTE AND THE Sapopitia. By O. F. Cook...-..-- 277 Introduction............-2.. 2.0202 e eee e eee ec ee cee eee eee eeeees 277 Confusion of vernacular names. ......-...-2--- 2.20202 02 cece eee eee eee 278 Essential differences.............-.-.22-2 22222222 e cece eee eee eee ee 278 Plumier’s account of the sapodilla.. ©... 2.2.2... e cece eee 279 Achras substituted for Sapota........--....2..0. 20.2222 cece eee eee eee 280 Confusion of species by Linneus......-...-..... 2.222022 2 222 eee eee ee eee 280 Adjustment of Linnwan names...............-..-2 2-22. 2ec eee ee ence eee 281 Two varieties of sapodilla named by Jacquin. ..-................2.-22--- 281 The name Sapota not to be revived. ..........22.22.-.0-002002220 2c eee eee 282 Lucuma and Vitellaria not applicable to the sapote...........2..2...-.. 283 Calospermum and Calocarpum as homonyms.................2.2.222---- 283 A new generic name for the sapote................222 2-222. eee eee eee eee 284 Summary of principal synonyms...................22222 cece eee cece eee 285 A MonoGRaPH OF THE HAUYEAE AND GONGYLOCARPEAE, TRIBES OF THE ONA- GRACEAE. By John Donnell Smith and J. N. Rose.............2.......20. 287 Introduction............. 2.22.22 2222 ence eee ee ee eeeeee 287 Systematic treatment. .........-2 2-22. 2 eee eee ce eee eee 287 BoTRYCHIUM VIRGINIANUM AND ITs rors, By Ivar Tidestrom............. 299 Botrychium virginianum.............2..2.22 22-0220 cece eee eee ee eeee 299 Botrychium dichronum......... we eee eee eee eee eee e eee 301 Botrychium brachystachys.........----. 2.22222. 0eee cece cece cee cececeee 302 Conclusion... .. a TTT nt ee eee ee ee eee eee eee eee eens 303 SPHENOCLEA ZEYLANICA AND CAPERONIA PALUSTRIS IN THE SOUTHERN UNITED Stares, By Ivar Tidestrom............-..2. 2.0.0.2 c cece eee e ce eee ee eee 305 Sphenoclea zeylanica. ........2 2.22222. 2 222 cece eee cece eee eens 305 Caperonia palustris. ..........2...2 2.22222 e eee ee eee e eee e eee 307 Piant Recorps or an Expepirion ro Lower Cauirornia. By Edward A, Goldman..............2. 2202222 e cece ee eeeeee 309 Introduction... ..... 2.2.0. 20 2.22 e eee cee cece cee cece eeeene 309 PLATE ILLUSTRATIONS. PLATES. Facing page. 1. Asplenium andrewsii A, Nelson. ...--------+++++2+eee2er eee Frontispiece 2. Asplenium andrewsit A, Nelson. .-..------+-+++-++2e-see streets 2 3. Olneya tesota A. Gray. .--.- 2-22-2222 eee eee eee e cere e sees 11 4. Abutilon macdougalit Rose & Standley. .....---------+++++++++++: 13 5. Sphaeralcea macdougalit Rose & Standley. ...-.-------+-++-+++++++ 13 6. Elaphrium microphyllum (A. Gray) Rose. ....---+++-+++-+2225+00+ 14 7. Carnegiea gigantea (Engelm.) Britton & Rose. ..-.--------------+-- 15 8. Echinocereus engelmanni (Parry) Riimpl.......----------++++++-++- 15 9, Echinocereus engelmanni (Parry) Riimpl.-.......---------++++++++- 15 10. Opuntia bigelovii Engelm....-.------------ +--+ 22eerce rere 16 11. Euploca aurea Rose & Standley.....--.-------+-+++++++2+222eereee 16 12, Dicoria calliptera Rose & Standley. ...-------------+++++++22e500+ 18 13. Isocoma fruticosa Rose & Standley. ..-.-----------+++++e2re rrr ere 18 14. Isocoma limitanea Rose & Standley... --.----------------+-+e0+++ 19 15. Sideranthus viridis Rose & Standley. .....-----------------+-++e-0+ 19 16. Viguiera sonorae Rose & Standley........---------++2++eee errr eee 20 17. Tumamoca macdougalit Rose.....---..---+-++----22 22 ere ttt rerecee 21 18. Hemitelia contigua (Underw.) Maxon........----------++++++++- 32 19, Hemitelia pittieri Maxon and H, subglabra (Underw.) Maxon.....--- 32 90. Hemitelia chiricana Maxon.........2+-------2- eee e eee e eerie 33 21. Hemitelia arachnoidea (Underw.) Maxon and H, mutica Christ--....- 34 92. Hemitelia apiculata Hook....-.-.-------------+ 22 eer errr e cree 35 93. Hemitelia grandis Maxon......-------+--------- 222 e crete terres 37 24. Four species of Hemitelia.....-.-.-------------+-+-+2+-+eerrrrcee 39 25. Hemitelia grandifolia Willd.......---+--------- +2222 re rere rere 41 26. Hemitelia kohautiana (Presl) Kunze.....-...-----------+-+--++++++- 45 27. Polystichum ambiguum Maxon. ...-.---------++-+2+2+22022e0erctee 50 98. Pteropsis underwoodiana Maxon.....-.--------+-+-++++-+55+0erttte 51 29, Dicranopteris bifida (Willd.) Maxon. ....----------+-++++2++++777+ 52 80. Cibotium sehiedei Schlecht. & Cham.........----------++++++++-+-- 55 31. Cibotium regale Versch. & Lem. ..-.--------------+++2+222+ereree 56 39. Cibotium wendlandi Mett. and C, guatemalense Reichenb...-..----- 57 33. Goniophlebium eatoni (Baker) Maxon....----------+------+-++++0+- 60 34. Goniophlebium rachipterygium (Liebm.) Moore.....-.------ sueueues 61 35 A. Fruit of Nymphaea microphylla Pers. B. Fruit of Nymphaea fraterna Miller & Standley. C. Fruit of Nymphaea advena Ait. -. 72 36. Seeds of several species of Nymphaea..-...-...----------+2eee-+: 73 37. Nymphaea americana (Provancher) Miller & Standley, in Flathead Lake, Montana. ...----.--2--- cece eee eee eee eee cece tesesessees 78 38. Fruit of Nymphaea americana (Provancher) Miller & Standley, from Springfield, Massachusetts.......---------+--+eeece ere rc eseeeee 79 39. Nymphaea advena Ait., in Monteers Pond, Knox County, Indiana. . 84 40. A. Earlier vernal stage of Nymphaea advena Ait., Four Mile Run, Vir- ginia. B. Later vernal stage of Nymphaea advena Ait., Four Mile Run, Virginia. .....-...-- 2... ee eee eee eee cece een e eee e eee 85 xi ILLUSTRATIONS. Facing page. Puate 41. A. Fruit of Nymphaea microcarpa Miller & Standley. B. Fruit and unopened flower of Nymphaea ludoviciana Miller & Standley.... 92 42. Seeds of several species of Nymphaea.............0.2.2.00eeeeeee 96 43. A. Fruit and unopened flower of Nymphaea ulvacea Miller & Stand- ley. B. Fruits of Nymphaea ovata Miller & Standley........... 97 44, A. Flower and fruit of Nymphaea sagittifolia Walt. B. Fruit of Nymphaea puberula Miller & Standley..............0.0ceceeeee 99 45. A. Flower and fruit of Nymphaea orbiculata Small. B. Flower and fruit of Nymphaea bombycina Miller & Standley................ 101 46. Nymphaea polysepala(Engelm.) Greene at Crater Butte, Colorado.. 108 47. Fruit of Nymphaea polysepala (Engelm.) Greene..............---- 104 48. Agave neomexicana Wooton & Standley in the Organ Mountains... 116 49. Eriogonum gypsophilum Wooton & Standley..............-.------ 118 50. Herrickia horrida Wooton & Standley..................ceceeeeees 187 51. Fruits of eight species of Meibomia.................-.0.-0eeeeeeee 216 52. Raimondia monoica Safford............2222. cece cece e cece aeons 218 53. Raimondia monoica Safford ...... 2.20.2. eee e cece eee cee | ceeee 219 54, A. Acrostigma equale Cook & Doyle. B. Wettinella quinaria Cook & Doyle... 22.2... eee eee cece cece cee cee eeeeeeeees= 230 55. Acrostigma equale Cook & Doyle. ..-...........eeceeeeeeeceeeces 230 56. A. Acrostigma equale Cook & Doyle. B. Catostigma radiatum Cook & Doyle. OC. Wettinella quinaria Cook & Doyle..............-- 230 57. Acrostigma equale Cook & Doyle. .........2.2.-.ceeeeceeeeees ~« 230 58. Acrostigma equale Cook & Doyle. .........-.2..0-eceeeeececceres 230 59. A. Fruits of Catostigma radiatum Cook & Doyle. B. Fruits of Wet- tinella quinaria Cook & Doyle...........2.0..02ceee eee cece --+. 231 60. Catostigma radiatum Cook & Doyle..........22222222ceeee eee eee 231 61. Catoblastus pubescens (Karst.) Wendl. and C. praemorsus (Willd.) Wendl... 2.2.22 ccc eee cece cece cece ececeerens 232 62. Flowersand fruit of Catoblastus pubescens (Karst.) Wendl......... 233 63. A. Wettinia augusta Poepp. & Endl. B. Wettinella quinaria Cook & Doyle. ....... 20. lee ec ee ee eee ee ee eeee 235 64. Wettinella quinaria Cook & Doyle...........22.2222020202 22 eee eee 236 65. Wettinella quinaria Cook & Doyle. ............2....2.02022222--- 237 66. Echinocactus alamosanus Britt. & Rose............20.02.0ceee ee ee 239 67. Echinocereus luteus Britt. & Rose...........202022- 2.2 e eee eee cess 239 68. Epiphyllum gaillardae Britt. & Rose...........-.--.------2---2----- 240 69. Hylocereus minutiflorus Britt. & Rose..............--2------+----- 240 70. Nyctocereus guatemalensis Britt. & Rose............22--22--2-20-- 240 71. Nyctocereus guatemalensis Britt. & Rose...............-..-------- 240 72. Opuntia chaffeyi Britt. & Rose........ 0.22 eee eee ee ee ee eee 241 73. Wittia panamensis Britt. & Rose........ 2222. c eee eee ccc cc eeee 241 74. Pseudophoenix sargentii Wend]. ..........220.0-0- eee ee eeeeeeeee 246 75. Pseudophoeniz sargentii Wend1.........2....22- 2-2 e eee cece eee eee 246 76. Pseudophoenizx sargentit Wend1.........2..22 222 e cece eee ceececes 247 77. Pseudophoenix sargentii Wendl. ...........2.20- 00 20c cece ececeee 247 78. Epiphyllum guatemalense Britt. & Rose............---.-0eeeeeeeee 257 79. Disocactus eichlamii (Weing.) Britt. & Rose..........2---.-e.eeeee 259 80. Zygocactus truncatus (Haw.) Schum. .........2....22-2-22-2--e0ee 260 81. Schlumbergera russelliana (Hook.) Britt. & Rose.........-.--.+---- 261 82. Wittia costaricensis Britt. & Rose...........2-.2.2.eceeeeeeceeees 261 83. Eccremocactus bradei Britt. & Rose............2 20002 e cece ee ee eee 262 84. Strophocactus wittit (Schum.) Britt. & Rose........-..----------- 262 PLATE ILLUSTRATIONS. XII Facing page. . Annona sericea Dunal. .....-.--------- 222s e eee ee eee eee 266 . Annona sericea Dunal. ...--.-.------ +--+ 0-22 eee ee eee eres 267 . Annona jenmanti Safford .....--.--- +--+ --+-- 2-2 e rete eee tees 268 . Annona trinitensis Safford..........------------2ee2 eee eee eee 268 . Annona longipes Safford........---------+--2 e222 eee eter eee 269 . Annona holosericea Safford......---------++++ 2222-2 eee e reese 270 _ A. Annona sericea Dunal. B. Annona holosericea Safford.......- 270 . Annona spraguet Safford.....-.--------- +--+ eee e eee cece cere 271 . Annona spraguet Safford.....-.-------+--+++-+++++ we ee eee ceeeee 271 . Annona cercocarpa Safford....---------+-++++22 22ers etter reese 272 . Annona echinata Safford.......-.-.----- 202s eee eee ee eee cece eeee 273 . Annona echinata Safford....-....----------- ee eee eee eee eee eeee 273 . Annona acuminata Safford......--..------+----2- +e eee reece eee 274 . Annona jamaicensis Sprague...--.-------------22ce terres cree 275 . Annona jamaicensis Sprague.....------------+2+2+ereeeere rete 275 . Plumier’s description and figures of Sapota. ....--------------- 279 . Three forms of sapodilla fruits.......------------+----++++eeeee- 280 _ Abnormal state of Botrychium cicutarium (Savigny) Swartz...... 302 . Caperonia palustris St. Hil...-.-.--2+-++-++eeece sere r eee tee 307 A. Libocedrus decurrens Torr., San Pedro Martir Mountains. B. Pinus contorta Dougl., San Pedro Mértir Mountains.....--....- 313 _ A. Pinus jeffreyi Oreg. Com., San Pedro Martir Mountains. B. Pinus lambertiana Dougl. and P. jeffreyi, San Pedro Martir Mountains.....----..--..--- ee eee eee eee eet 313 . Glaucothea armata (S. Wats.) Cook (?), Jaraguay-.......-.-------- 316 _ A. Glaucothea armata (S. Wats.) Cook, Catavifia Canyon. B. Phoenix dactylifera L., San Angel.......-----.-----++++2+20-- 316 . Nolina beldingi T. S. Brandeg., Sierra de la Laguna...--..------ 317 . Nolina bigelovii 8. Wats., Yubay........-------++-+-++222eeeeee- 317 _ Yucca valida T. 8. Brandeg., Santa Rosalia Bay...------------- 317 . Four species of Agave: A. A. goldmaniana Trel.,. San Andrés; B. A. promontorii Trel., Sierra de la Victoria; ©. A. vexans Trel., El Potrero; D. A. nelsoni Trel., San Fernando. .-..---------- 318 . Dudleya anthonyt Rose, San Martin Island......-----------+-+--- 329 . Lysiloma candida T. 8. Brandeg., 40 miles west of Santa Rosalia... 333 . Cercidium peninsulare Rose, Santa Anita......--------+++------- 336 . Elaphrium odoratum (T. 8. Brandeg.) Rose, El Potrero.......--. 340 . A. Jatropha canescens Muell. Arg., Cape San Lucas. B. Jatrop spathulata Muell. Arg., in dry season, Santa Rosalfa......---.- 342 _ A. Pedilanthus macrocarpus Benth., Matancita. B, Sebastiania bilocularis S. Wats., El Pescadero......-.----------+-++++---- 343 _ A. Pachycormus discolor (Benth.) Coville, in dry season, San An- drés, B. Same species, showing larger growth, Santa Clara Mountains.........-.---------- eee eee eee 344 . Tapirira edulis T, 8. Brandeg., Cape San Lucas.....------------- 345 . Fouquieria peninsularis Nash, San Andrés......------+-----+-+++ 349 . Idria columnaris Kellogg, San Fernando.....--..-------------++- 350 _ A. Idria columnaris Kellogg (young plant, in flower), Agua Dulce B. The same species (full grown), Calmalli.....-.----------- 350 _ A. Echinocactus falconeri Orcutt, San José del Cabo. B. Echino- cactus digueti Web., Cerralvo Island....-.-..-------++++-++--- 351 . Echinocereus brandegei (Coult.) Schum., Matancita....-.-------- 302 XIV PLATE FIGURE 125. 126 127, 128, 129. 130. 131. 132. 133. 1. a ILLUSTRATIONS. Facing page. A, Lemaireocereus thurberi (Engelm.) Britt. & Rose, Tinaja de San- tana. B. Lophocereus schottii (Engelm.) Britt. & Rose, San Francisquito. .....2...22. 2.202.202 c cee cee cece eee eeeeeeeeee A. Lemaireocereus gummosus (Engelm.) Britt. & Rose, Espfritu Santo Island. B. Lophocereus australis (K. Brandeg.) Britt. & Rose, Cape San Lucas....--.....2.2.0200002c cece eee ce cee ecee Lemaireocereus eruca (T. 8. Brandeg.) Britt. & Rose, Santo Do- A. Opuntia cholla Engelm., Espiritu Santo Island, B. Opuntia bigelovit Engelm., San Felipe Bay.............2-.2--2-0--0-0- A, Opuntia clavellina Engelm., San Andrés, B. Opuntia sp., Agua Dulce... 22... ccc eee cece cece ee eecee A. Pachycereus calvus (Engelm.) Britt. & Rose (young plant), San José del Cabo. B. Pachycereus titan (Engelin.) Britt. & Rose (young plant), San José del Cabo (Plants of P. calvus in back- ground). 2.2... eee ec c cece cece cececeee A. Pachycereus calvus (Engelm.) Britt, & Rose, San José del Cabo, B. Pachycereus pecten-aboriginum (Engelm.) Britt. & Rose, Cape San Lucas. ... 22.0.0... 2... cece cece ee ceee TEXT FIGURES. Map of the Sonoran Desert Region between Tucson, Arizona, and the Gulf of California............2....2....2.---.-..- see eeeeee . Leaf outline of Nymphaea microphylla...............0.0-0-000-- . Stigmatic pattern of Nymphaca microphylla...............-...-- . Map showing distribution of (a) Nymphaea microcarpa; (b) Nym- phaea orbiculata; (c) Nymphaea microphylla.............2-000-- . Leaf outline of Nymphaea rubrodisca.............2.2..-.---20-- . Stigmatic pattern of Nymphaea rubrodisca............2-..22224- - Map showing distribution of (4) Nymphaea ovata; (b) Nymphaea rubrodisca; (c) Nymphaea ludoviciana; (d) Nymphaea ulvacea; (e) Nymphaea bombycina; (f) Nymphaea sagittifolia.......-.22... . Leaf outline of Nymphaca americana....................2.20---- . Stigmatic pattern of Nymphaea americana...................-.--- . Map showing distribution of (2) Nymphaea puberula; (b) Nymphaea Sluviatilis; (c) Nymphaea americana... 2.2.2.2... eee eee ee . Leaf outline of Nymphaea fraterna.......... 2.20022. 2 eee eee . Stigmatic pattern of Nymphaca fraterna...............-.20.0005- - Map showing distribution of (a) Nymphaea ozarkana; (b) Nymphaea Jraterna; (c) Nymphaea chartacea.............--..- weet eee eee . Leaf outline of Nymphaea -advena.............2.2-22222 eee eee . Stigmatic pattern of Nymphaea advena..........-.--2-. 2.0 eee . Map showing distribution of (a) Nymphaca advena macro phylla; (b) Nymphaea advena; (c) Nymphaea polysepala; (d) Nymphaea advena CTYthTQed... 2.0.22 e ee eee ee eee wee eee eee eee ee eee eee eee . Leaf outline of Nymphaea advena macrophylla.............2..+-. . Leaf outline of Nymphaea ozarkana ......... 2-2-0202 eee eee eee ee . Leaf outline of Nymphaea ludoviciana................22-22222-- . Stigmatic pattern of Nymphaea ludoviciana................-..--- . Leaf outline of Nymphaea fluviatilis... .. 222.22... ee 300 ILLUSTRATIONS. xV Facing page. Figure 22. Stigmatic pattern of Nymphaea fluviatilis.....--.------+-++++-++--- 94 23. Leaf outline of Nymphaea chartacea......-------------++-++++++-- 95 24. Leaf outline of Nymphaea sagittifolia......-.----------------++++++- 96 25. Stigmatic pattern of Nymphaea sagittifolia...-..--------+-+-++++-- 96 26. Leaf outline of Nymphaca ulvacea.......-.--------- +++ +202 eeee ee 97 27. Stigmatic pattern of Nymphaea ulvacea......--------++-++2+++0+- 97 28. Leaf outline of Nymphaea ovata........---- +--+ 2-22-22 reece 98 29. Stigmatic pattern of Nymphaea ovata.....---.----+-+-++++22+005+>- 98 30. Leaf outline of Nymphaea puberula......-------- +--+ +++ 2-20 +2000 99 31. Stigmatic pattern of Nymphaea puberula........--+-----++++-++++- 99 32. Leaf outline of Nymphaea microcarpa...--------+------- Lecce ee eeee 100 33. Stigmatic pattern of Nymphaea microcarpa......--------+--+---+-- 100 34. Leaf outline of Nymphaea orbiculata.....-----------+++++++++++-+- 101 35. Stigmatic pattern of Nymphaea orbiculata......------------------ 101 36. Leaf outline of Nymphaea bombycina......--------++----+--------- 102 37. Stigmatic pattern of Nymphaca bombycina...--------++-+++-+------ 102 38. Leaf outling of Nymphaea polysepala, typical form...........------ 103 39. Stigmatic pattern of Nymphaea polysepala......----------++-+-++-- 103 40. Leaf outline of Nymphaea polysepala, southern form.....--------- 105 41. Ovary of Catoblastus drudet..-.....-------++20ee cer crt eter tttte 233 42. Fruit of Annona sericed.....---------+--e2 eee errr 266 43. Leaf and fruit of Annona spraguct.....------+-++++-2-- errr 271 44, Leafy twig and fruit of Annona cercocarpa.....--------++++++++777- 272 45. Leaf of Hauya heydeana........----+----- +2002 e crete eset rtttee 290 46. Leaf of Hauya barcenac.......---- +--+ +--+ 22 ee rere re cette 290 47. Leaf of Hauya lucida.......-..+---------022e eter ccc r crs 290 48. Leaf of Hauya rusbyi..-.-.----------- +202 rece e ccc ctrtrtrtete 291 49. Leaf of Hauya cornuta........----------+-2 erect errr crt 292 50. Leaf of Hawya microcerata.....-------+--+-+-+ eee er ececcttte 292 51. Leaf of Hauya rodriguezti.....-.-------+---+2r reese cre reteeteee 293 52. Leaf of Hauya quercetorwm....-----------+-02 eer errr crt rctettee 293 53. Leaf of Hauya ruacophila .....----------- 2222s eter e recess 294 54. Leaf of Hauya lemnophila....--..-----------02e errr crete crete 295 MAP. 309 Map of Lower California........---+-+--+++++22550ecccccrrrrtsrs rss rt rte THE RELATIONSHIP OF ASPLENIUM ANDREWSIL. By Wittram R. Maxon. In the final brochure ! of the Proceedings of the Biological Society of Washington for 1904, volume 17, Prof. Aven Nelson published, as one of several undescribed species collected in Colorado by Mr. D. M. Andrews, a supposed new Asplenium, which he dedicated to its discoverer as Aspleniuwm andrewsii.?, The description and notes are as follows: “Rootstock short, wholly enveloped in matted roots; stipes naked, ebeneous below, becoming green above, from 2-10 cm. long, somewhat angled or striate; lamina thinly herbaceous, deltoid-ovate or narrower, 3-10 cm, Jong, somewhat narrower at its widest part, bipinnatifid, diminishing nearly uniformly from base to tip; pinnz lanceolate, the lower nearly at right angles to the rachis, the upper ascending, gradually diminish- ing and passing into the pinnatifid tip, all rather closely approximate and subopposite or the lower more distant (1 cm. or more) and alternate; pinnules 3-12 mm. long, ovate, more or less cuneate at base, sharply incised but cut not quite to the costa, sharply and somewhat incisely serrate; the veins rather inconspicuous and but slig::tly divergent; sori short but nearly connecting to those in the successive lobes, so forming almost a continuous sorus from base to apex of pinnule; indusium straight, forced back and finally concealed by the sporangia. ‘Perhaps most nearly allied to A. bradleyi D.C. Eaton, but probably not very closely even to this. Mr. Andrews writes of it as follows: ‘The most interesting item on the list to me. I am sending a better specimen. It is certainly indigenous and grows on the south face of a white sandstone (alkaline) cliff extending along Boulder Creek for a mile or more, the ferns growing in crevices abundantly for nearly the whole dis- tance. Itis growing with Cheilanthes feci, a specimen of which I send you. The sand- stone is porous and is not entirely dry.’” Not long after this Professor Underwood called my attention to the obviously close relationship of this form to the Old World Asplenvum adiantum-nigrum, and, if I am not mistaken, to the great difficulty or impossibility of distinguishing it specifically from that species, as usually accepted by European botanists. In 1906, however, the species was recognized by him as valid in Rydberg’s Flora of Colo- rado, and again in his article ‘‘American Ferns, VI—Species added to the flora of the United States from 1900 to 1905,”’* here with the comment, ‘This new discovery from Colorado is a member of the adiantum-nigrum group of Asplenium, and is closely related to Asple- 1Tssued December 27, 1904. 2 Page 174. 1 2 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. species it will stand nearest to A. montanum.” From which it may be noted that Doctor Underwood was not only unwilling, apparently, to ascribe to A. adiantum-nigrum itself an extra-Kuropean range (though the species is usually accredited also to various parts of Africa, and to the Himalayas, Asia Minor, and other parts of Asia), but that also, in accordance with his often-expressed views as to the relatively restricted range of fern species, he was inclined upon @ priori grounds to look upon the American plant as specifically distinct from that of the Old World. That A. adiantum-nigrum as ordina- rily accepted does occur in Africa and Asia is apparent from speci- mens at hand; and if we admit the various forms distinguished by Milde,' Luerrsen,? Christ,? and others as constituting but a single highly variable species, there seems to be no logical ground for re- garding A.andrewsii as other than a geographical phase of A. adian- tum-mgrum. Excepting only Athyrium filizx-foemina, there is, prob- ably, no fern occurring in the United States which closely approaches it in extent of variation. The two illustrations (Pls. 1 and 2) herewith represent at natural size the type specimens of Asplenium andrewsti, which have courteously been lent from the Rocky Mountain Herbarium, University of Wyo- ming, by ProfessorAven Nelson. Ingeneral shape the laminais appar- ently unusual for A. adiantum-nigrum in its relatively great width. Most of the foreign material at hand most closely resembling this shows blades elongate-deltoid in form, the upper portion often atten- uate,—a leaf shape more nearly approached by some of the smaller fronds here shown. In fact not one frond of the foreign material available for comparison has precisely the same leaf shape as that of A. andrewsii, the nearest approach being in specimens from Doul- lens, France, Copineau, July 12, 1887, and from Devonshire, England, Ware, July 15, 1904, These appear to represent the variety argutum, as described by Luerrsen. Lacking a first-hand knowledge of A. adiantum-mgrum as it occurs in Europe, I hesitate to refer to it without reservation this American form, which is known only from such meager material; but I believe that the highly complex ‘“‘spe- cies’? A. adiantum-nigrum, as generally understood at present, embraces among its various and varying forms several elements which, in their extreme states, differ more widely from each other than from A. andrewsii. In degree of dissection, leaf texture, color of leaf tissue and of vascular parts, shape of pinne, extent and char- acter of soriation, and in more minute characters, such as the pecul- iar form and structure of the long, slender, hair-pointed scales of the rhizome, the American plant certainly agrees very closely with some 1 Milde, Fil. Eur. Atlant. 85. 1867. ? Luerrsen, Die Farnpflanzen 260. 1889. 3 Christ, Die Farnkriiuter der Schweiz 68. 1900. Contr. Nat. Herb., Vor. 16. PLATE 2. ASPLENIUM ANDREWSII A. NELSON. MAXON—RELATIONSHIP OF ASPLENIUM ANDREWSII. 3 of the European material, especially with the Devonshire specimen cited. The differences consist principally in its more broadly deltoid fronds and perhaps in its shorter stipe; the stipes of the two large fronds of the type, however, are incomplete, and one of the fronds -isseen to be malformed. Through the generosity of Professor Nelson one of the smaller fronds of the type has been presented to the National Herbarium. The collecting of more adequate material of A. andrewsi at the type locality, which is, I believe, not very readily accessible, would be of the greatest interest as throwing light upon the extent of variation in the American plant and its relationship to the Old World forms; and it is hoped that the publication of illustrations of the type specimens (Pls. 1 and 2) may assist in calling to the atten- tion of collectors in the Rocky Mountain region the main characters of a plant which has remained so little known since the time of its description. The original locality has, indeed, been revisited by Prof. E. Bethel, who collected further specimens. Of these a small plant has recently been figured in the Fern Bulletin, with notes by Mr. Clute.' A somewhat similar instance of distribution among the ferns 1s that of Asplenium septentrionale, a species which iscommon in Europe, occurs in the Caucasus, the Himalayas, and Tibet, and in North Amer- ica ranges from the Black Hills of South Dakota (Rydberg 1194) to New Mexico (several collectors), Arizona (MacDougal 68), Colorado (several collectors), and Wyoming (A. Nelson 8900), and is known even from the San Pedro Martir Range of Lower Califorina (Brandegee, May 18, 1893). Specimens from all these localities are in the Na- tional Herbarium. 1Fern Bull. 19: 3. Frontis. 1911. REPORT ON A COLLECTION OF PLANTS FROM THE PINACATE REGION OF SONORA. By J. N. Rose anp Paut C. STANDLEY. INTRODUCTION. An expedition was organized at the Desert Laboratory of the Carnegie Institution by Dr. D. T. MacDougal in the autumn of 1907 for the purpose of making a general bio-geographical reconnaissance of the region between Tucson and the Gulf of California. Attention was to be directed chiefly to a comparison of the physical features of the coastal desert with those of the elevated arid area in which the Desert Laboratory is located and to obtaining data regarding the general features of distribution and environment of the higher plants and animals. The expedition left Tucson, Arizona, November 2, 1907, going westward 125 miles to the northern end of the Ajo Mountains, thence southward across the Mexican boundary to the village of Sonoyta, and westward through Santo Domingo and Quitovaquito. The course of the Sonoyta River was now followed southward to Agua Dulce where its waters are lost in the sands and then the route was laid across the desert to Monument 180 on the boundary, from which a departure was made that took the party southward along the western side of the Pinacate Mountains, the principal stations being Papago Tanks, Tule Tanks, and Pinacate Peak. In addition to this Mr. G. Sykes made a forced march to the shore of Adair Bay on the Gulf of California. The Pinacate Mountains are the highest in northwestern Sonora. They run north and south just east of parallel 113° 30’ longitude, between 31° 40’ and 31° 50’ north latitude. The highest peak, Pinacate, is about 1,218 meters in height, its slopes extending with but slight interruption to the shore of the Gulf. The entire range is of recent volcanic origin, with many sunken or elevated craters, the formations including great areas of volcanic sand, ashes, tufa, and hard lava, and the range lies in a vast field of broken lava which extends northward into Arizona. A careful survey of the region traversed was made by Mr. G. Sykes and his most excellent map is reproduced in connection with this article. 5 6 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. A small but very interesting collection of herbarium specimens made by Doctor MacDougal forms the basis of this paper. No attempt was made to obtain a full representation of the flora of the a a> ua) Or 15 miles 13" an Francisco Pk 5 i ee .4—----S = 3°30" Crater 330° % 2 &s 3 3 tee >? & . Ss Neen : ORS. a Ny > § [ri ye) os 3 é ry) a =e o WS Bs O y xy *uey p AS = os ns) & , ‘t 5 a oO q 2 2 . s ¢ 1) 3 ec. ” c 3 DE | y ple Ly Soe ° hs a 3 4) é G ‘ty ‘a8 Fic. 1.—Map of the Sonoran Desert Region between Tucson, Arizona, and the Gulf of California. region but only those plants were taken which were especially inter- esting or were in suitable condition. In addition a large series of photographs showing the botanical and geographical features was ROSE AND STANDLEY—-PLANTS FROM THE PINACATE REGION. 7 obtained. A general discussion of the principal facts as to the dis- tribution of the plants has been published by Doctor MacDougal in the Year Book of the Carnegie Institution of Washington for 1908 and a fuller treatment in the Bulletin of the American Geographical Society for December, 1908, and in The Plant World for May and June, 1908. The data respecting the animal life of the region obtained by Dr. W. T. Hornaday and Mr, J. 8. Phillips have been published under the title of ‘‘Camp-Fires on Desert and Lava” by Doctor Hornaday, a most delightful book dealing with the experiences of the naturalists on the expedition, illustrated with many plates showing characteristic desert shrubs and trees. The botanical collections, although small, have proved to be most interesting, as was to be expected from an unexplored region, Several of the plants appear to be new and these are here described. Some of those already known are noteworthy, representing forms seldom collected. No botanical collector had ever visited the Pinacate region. At Sonoyta asmall collection was made by Dr. KE. A. Mearns in 1894. A list of his plants will be found in Bulletin 56 of the U.S. National Museum. The photographs here reproduced were taken by Dr. MacDougal, and these with the line drawings are the gift of the Carnegie Institu- tion of Washington. LIST OF PLANTS, WITH DESCRIPTIONS OF NEW SPECIES. Aristida bromoides H. LB. K. Nov. Gen. & Sp, 1: 122. 1815. Type locality, ‘In montibus regni Quitensis, juxta Tambo de Guamote et Llanos de Tiocaxas, alt. 1600 hexap.”’ Quitovaquito, November 11, 1907, MacDougal. Aristida californica major Vasey, Proc. Calif. Acad. IT, 2: 212. 1889. Type locality, Magdalena Island, Lower California. MacDougal Pass, Pinacate Mountains, November 14, 1907, MacDougal 32. Bouteloua polystachya Torr, U. S. Rep. Expl. Miss. Pacif, 5: 366. 1853. Chondrosium polystachyum Benth. Bot. Voy. Sulph. 56. 1844. Type locality, ‘Bay of Magdalena,” Lower Cali fornia. Papago Tanks, November 17, 1907, MacDougal. Cenchrus palmeri Vasey, Proc. Calif. Acad. IT. 2: 211, 1889. Type locality, ‘‘Guaymas, Mexico.”’ Sand hills near the Adair Box, November 20, 1907, Sykes 58. Heteropogon contortus (L.) Beauv.; Roem. & Schult. Syst. 2: 836. 1817. Andropogon contortus L. Sp. Pl. 1045, 1753, Type locality, ‘In India.”’ Papago Tanks, November 17, 1907, MacDougal 52. Leptochloa mucronata pulchella Scribn. Bull. Torrey Club 9: 147, 1882. Type locality, “Santa Cruz Valley, near Tucson.”’ Papago Tanks, Pinacate Mountains, November 17, 1907, MacDougal 40. 8 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Triodia pulchella H. B. K. Nov. Gen. & Sp. 1: 155. pl. 47. 1815. Type locality, “In subfrigidis, siccis, apricis regni Mexicani inter Guanaxuato, Mina de Bel grado et Cubilete, alt. 1050 hexap.” Agua Dulce, November 11, 1907, MacDougal. Hesperocallis undulata A. Gray, Proc. Amer. Acad. 7: 390. 1868. Type locality, “Desert plains at Jessup Rapids, Arizona.” This was seen at Walls Well, in southern Arizona, but no specimens were collected. The plant is commonly known as “‘ajo,”’ and it is from this plant that the Ajo Range receives itsname. The word signifies ‘‘garlic”’ and the roots are said to have a strong alliaceous flavor. Momisia pallida (Torr.) Planch. in DC. Prodr, 17: 191. 1873. Celtis pallida Torr. Bot. Mex. Bound. 203. 1859. Type locality, ‘In western Texas and along the Rio Grande from Fort Duncan to the Gulf and west to Magdalena, in Sonora.”’ Walls Well, Ajo Mountains, Arizona, November 7, 1907, MacDougal 8. There are three other species of the genus Momisia in Mexico for which the proper names seem not to have been formed. They may, for convenience, be entered here. MomIsIA ANFRAcCTUOSA (Liebm.) Rose & Standley. Celtis anfractuosa Liebm. Vidensk. Selsk. Skr. V. 2: 338. 1851. Type locality, ‘“Xalcomulco, Vera Cruz.”’ MoMISIA PLATYCAULIS (Greenm.) Rose & Standley. Celtis platycaulis Greenm. Proc. Amer. Acad. 39: 78. 1903. Type locality, “State of Morelos; volcanic hills near Yautepec.”’ MomIsIA IGUANAEA (Jacq.) Rose & Standley. Rhamnus iguanaea Jacq. Enum. Pl. Carib. 16. 1762. Celtis aculeata Swartz, Prodr. Veg. Ind. Occ. 53. 1783. Mertensia laevigata H. B. K. Nov. Gen. & Sp. 2: 31. pl. 103. 1817. Momisia aculeata Klotzsch, Linnaea 20: 539. 1847. Celtis iguanaca Sarg. Silva N. Amer. 7: 64. 1895. Type locality, West Indies. Phoradendron californicum Nutt. Journ. Acad. Phila. n. ser. 1: 185. 1847, Type locality, “‘In the mountains of upper California. Parasitic on the trunks and branches of a Strombocarpus.”’ Pinacate Region, November, 1907, MacDougal. In his account of the trip Doctor Hornaday writes as follows concerning this plant:! “Throughout our trip we found the large mesquite trees of the valleys and flood plains grievously afflicted with mistletoe. It usually appears as a great, dark-colored bunch 2 feet in diameter, and sometimes we found a dozen clumps in one tree. This parasite, like most others, is destructive when overdone. We saw many hapless trees that had literally been murdered by it and were only lifeless stubs. It was in the valley of the Sonoyta River, near Agua Dulce, that Doctor MacDougal photo- graphed a wide-spreading mesquite whose top was so overloaded with mistletoe that it looked as if a small load of clover hay had been pitched into it.”’ 2 Eriogonum fasciculatum Benth. Trans. Linn. Soc. 17: 411. 1837. Type locality, ‘‘Upper California.” Walls Well, Ajo Mountains, November 5, 1907, MacDougal 9. 'Camp-Fires on Desert and Lava, page 48, ? For illustration of this tree, see plate opposite page 48 of the ‘‘ Camp-Fires.”” ROSE AND STANDLEY—PLANTS FROM THE PINACATE REGION. 9 Eriogonum pinetorum Greene, Muhlenbergia 6: 3. 1910. Type locality, ‘Black Range, Sierra County, New Mexico.” Paso Blanco, November 6, 1907, MacDougal 2. This species has long been confused with Eriogonum abertianum, but is readily distinguished from it. Its habit is strikingly different and the calyx is a light pink instead of dark red. Eriogonum vimineum Dougl.; Benth. Trans. Linn. Soc. 17: 416. 1837. Type locality, ‘‘Columbia river. ” Pinacate Mountains, at 1,200 meters, November 21, 1907, MacDougal 71. The spec- imens appear to belong to this species, although they are not in the best condition for determination. Rumex hymenosepalus Torr. Bot. Mex. Bound. 177. 1859. Type locality, ‘‘Sandy soils from El Paso to the cafions of the Rio Grande.” No specimens were collected but it was seen at Walls Well in the Ajo Mountains. It is a common southwestern plant whose roots are much used for tanning. Atriplex canescens (Pursh) James, Trans. Amer. Phil. Soc. 2: 178. 1825. Calligonum canescens Pursh, Fl. Amer. Sept. 370. 1814. Type locality, ‘‘In the plains of the Missouri, near the Big-bend.” Walls Well, Ajo Mountains, November 8, 1907, MacDougal 4. Amaranthus palmeri S. Wats. Proc. Amer. Acad. 12: 274. 1877. Type locality, ‘‘At Larkin’s Station, San Diego County, California.” MacDougal Crater, Pinacate Mountains, November 14, 1907, Sykes. This is per- haps the commonest species of Amaranthus in the Southwest. Very frequently the plants occur in such abundance that they are cut and cured for hay. Cladothrix lanuginosa Nutt.; Moq. in DC. Prodr. 18%: 360. 1849. Alternanthera lanuginosa Moq. op. cit. 359, 1849. "Type locality, ‘“Secus Salt-river et Red-river.” MacDougal Crater, Pinacate Mountains, November 14, 1907, Sykes 29 and 31. Boerhaavia wrightii A. Gray, Amer. Journ. Sci. IT. 15: 322. 1853. Type locality, ‘‘Pebbly hills near El Paso.” Papago Tanks, Pinacate Mountains, November 16, 1907, MacDougal 43. Wedeliella incarnata (L.) Cockerell, Torreya 9: 167. 1909. Allionia incarnata L. Syst. Nat. ed. 10. 890. 1759. Wedelia incarnata Kuntze, Rev. Gen. Pl. 533. 1891. Type locality, Peru. MacDougal Crater, Pinacate Mountains, November 14, 1907, MacDougal 30. Isomeris arborea Nutt.; Torr. & Gr. Fl. N. Amer, 1: 124, 1838. Type locality, ‘‘St. Diego, California. ” Pinacate Mountains, at 600 to 900 meters, November 21, 1907, MacDougal. Wislizenia costellata Rose, Proc. Biol. Soc. Washington 19: 132. 1906. Type locality, ‘‘Sonora, Mexico. Between Nogales and Guaymas. ” Sonoyta, November 8, 1907, MacDougal 12. Krameria glandulosa Rose & Painter, Contr. Nat. Herb. 10: 108. 1906. Type locality, ‘‘Near El Paso, Texas.” Hornaday Range, Pinacate Mountains, November 14, 1907, MacDougal 23. Acacia greggii A. Gray, Smiths. Contr. Knowl. 3: 65. 1852. Type locality, ‘“Dry valley west of Patos, Northern Mexico.” No specimens were collected but the plant was observed throughout the region visited. 10 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Parkinsonia microphylla Torr. U.S. Rep. Expl. Miss. Pacif. 4: 82. 1856. Type locality, ‘‘ Banks of the Colorado, and on Williams’ river,” Arizona. This, the common palo verde, is discussed and illustrated by Doctor Hornaday in the ‘“‘Camp-Fires.” Doctor Hornaday speaks of it as follows: } “Of all the tree products of the desert the palo verde is one of the most beautiful and interesting. Itsname is Spanish and means ‘green tree.’ According to its soil and water supply, it may be as large as an adult apple tree—fifteen feet high, with a trunk nine inches in diameter—or as small as a mountain laurel bush three feet high. Almost as far as it can be seen, you recognize it at once as something different and remark- able. Instead of a top that is made up of leaf masses, one laid upon another, you see that its foliage—or rather the masses where its foliage ought to be—is composed of straight lines, and angles. The palo verde bears a few tiny leaflets, so small that it would take about twelve of them to cover a postage stamp; but in November they exert no influence whatever upon the general aspect of the tree. ‘Regardless of leaves, however, from root to top the palo verde is of the most beau- tiful green that could be imagined. It is not the bold, waxy, aggressive green of the creosote bush, but the soft, smooth, and delicate green of the asparagus. “The bark is as smooth as the surface of polished oak, and trunk, branch, and twig are alike persistent green. Even the bark of the trunk has a surface like a robin’s egg. “The terminal twigs are long, straight, and slender, like masses of green darning needles set where the leaves ought to be. The density of their color, added to their unique form, gives the tree as a whole a peculiarly lineated top. This is one of the very few desert trees that is free from thorns. “This tree is not particularly useful. Its chief purpose is to ornament the arroyos and flood basins of the desert regions, and to furnish brake blocks for desert freight- wagons. It strings along the arroyos, wherever the water supply is a little above the average, but on the open, level plains it is rare. Often from many a square mile it is quite absent. In density and grain, its wood is much like that of the white birch. The trunk consists of a single stem, upon which the branches are set in very abrupt and angular fashion, all of which merely adds to the odd appearance of the tree. ” Prosopis velutina Wooton, Bull. Torrey Club 25: 456. 1898. Type locality, ‘Probably first collected with young fruit in the valley of the Nazas in northern (?) Mexico by Gregg.” No specimens of this were collected, but it is illustrated in several of the illustrations of the ‘‘Camp-Fires.’’ Doctor Hornaday writes of it as follows: 2 “The honey-pod mesquite is the most persistent bush tree of the deserts. Both in form and size it is much like the palo verde, and in southern Arizona and Mexico the two species are almost inseparable companions. On the desert plains, where water is scarce and dear, the mesquite is a modest little bush three feet high; but along the arroyos, the valleys, and in the business centers of the flood basins, where the water wagon is more in evidence, it develops into a real tree. Often it grows to a height of twenty-five feet, with a writhing trunk twelve or more ‘uches in diameter. In growth habit it is very much likean apple tree—a low, heavy, wide-spreading top with crooked bra ches that frequently are horizontal, on a short, stout trunk of irregular shape. The bark is gray and the foliage is of a pale gray green tint—not so pleasing as the asparagus green of the palo verde. * * * Both foliage and ‘“‘beans” are eaten by horses and cattle when grass is not obtainable and hunger is great. Its seeds are greedily eaten by all the small rodents of the deserts, and by many birds also. Although its leaves are very small, the shade of the mesquite is very grateful and comforting. ‘The mesquite is well provided with thorns, but, fortunately for the proletariat, they point forward instead of back. Its wood is hard, fine-grained, durable, and the general stand-by for fuel throughout the whole Southwest. Blessed is the desert wayfarer ' Page 45; see also plate facing p. 70, same work. 2 Page 47. PLaTe 3. 16, Nat. Herb., Vol. tr. Cor "AVHD “YW VLOSSL VASNIO ROSE AND STANDLEY—PLANTS -FROM THE PINACATE REGION. 1] who has dry mesquite for his camp-fire; for without it fire making is a serious problem. It burns freely, makes a hot fire, and quickly produces a good bed of coals for the baking of bread and the frying of meat. ‘“‘In the simple house building of the deserts, mesquite constitutes well-nigh the only wood that is available. The stems are used to support the earth roofs of houses, to build into fences for corrals and cultivated fields, and to repair broken wagons. It is said that the Mexicans also use it in the making of furniture.”’ Olneya tesota A. Gray, Mem. Amer, Acad. IT. 5: 328. 1855. PLATE 3. Type locality, ‘‘On the table-lands of the Gila.” No specimens of this were collected, either, but it is common about the Pinacates and in southern Arizona, where it is known asironwood. Ofit Doctor Hornaday says:? ‘The ironwood tree is not of sufficient importance to justify prolonged attention. It looks very much like the mesquite, but its wood is as hard as its name implies, and so heavy that it will not float in water. The largest specimen I noted particularly was @ conspicuous part of our aforesaid bivouac on Pinacate. A trunk fully a foot in diameter and twenty feet long was twisted almost into a figure-8 knot, but it was what cattlemen call a ‘lazy 8,’ for it lay upon the ground.” EXPLANATION OF PLATE 3.—From a photograph by Dr. D. T. MacDougal. Parosela emoryi (A. Gray) Heller, Cat. N. Amer. Pl. ed. 2. 6. 1900. Dalea emoryi A. Gray, Mem. Amer. Acad. II. 5: 315, 1855. Type locality, ‘‘On the desert table-lands of the Gila,” Adair Box, November 20, 1907, Sykes 65. Parosela spinosa (A. Gray) Heller, Cat. N. Amer, Pl. ed. 2. 7. 1900. Dalea spinosa A. Gray, Mem. Amer. Acad. II. 5: 315. 1855. Type locality, ‘‘Arroyos on the Gila; and on the Californian desert west of the Col- orado.”’ Sandhills, Adair Box, November 20, 1907, Sykes 64. This is spoken of as the “‘spiny smoke tree.”’ It is well illustrated in the ‘‘Camp- Fires.’’ ? Phaseolus wrightii A. Gray, Smiths. Contr. Knowl. 3: 43. 1852. Type locality, ‘‘Declivity of a mountain, near El Paso.” Papago Tanks, Sonora, November 17, 1907, MacDougal 48. Covillea glutinosa (Engelm.) Rydb. N. Amer. Fl. 25%: 108. 1910. Larrea glutinosa Engelm. in Wisliz. Mem. North. Mex. 93, 1848. Type locality, ‘‘Olla and Fray Cristobal,’’? New Mexico, The common creosote bush, occurring nearly throughout the arid Southwest. Again we quote Doctor Hornaday from the ‘‘Camp-Fires:” ® ‘“‘Last of the important bushes and trees of the desert—but often it is the first—is the creosote bush. It is by far the most omnipresent representative of the plant world throughout the region we traversed. I think we saw hundreds of square miles of it, and most of all was on the trail from the Ajo mines up to Gila Bend. “The specimen shown with Mr. Sykes and the grave of the murdered Mexican is an excellent picture of a creosote bush, which may be regarded as the type of ten million others. The creosote bush is a big cluster of small and brittle woody stems, covered with smooth brown bark. The stems do not branch until near their tops, and there they send off a few fine twigs to support the irregular clusters of tiny leaves that form the outer surface of the bush. The leavesare of a rich, bright green color, and so shiny that they look as if recently varnished. They taste unpleasantly like creosote (oil of smoke), and no animal can eat them. 1 Page 52. ? Plate opposite page 182. 3 Page 53. 13540°—12——2 12 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. “The leaves of the creosote bush are so wholly on its outer surface that it would be quite easy to shear them all off, as one shears a sheep, and leave the bush nearly full size but perfectly bare. The usual height of this bush is from two to three feet. The clumps stand about ten feet apart, and usually there are from 100 to 150 peracre. In a few localities we saw some very large specimens, which grew fully ten feet in height.”” Kallstroemia grandiflora Torr.; A. Gray, Smiths. Contr. Knowl. 3: 28. 1852. Type locality, ‘‘ Borders of the Gila,” New Mexico or Arizona. MacDougal Crater, Pinacate Mountains, November 14, 1907, MacDougal 25. Chamaesyce pediculifera (Engelm.) Rose & Standley. Euphorbia pediculifera Engelm. in Torr. Bot. Mex. Bound, 186. 1859. Type locality, ‘‘Sonora.” Quitovaquito, Sonora, November 11, 1907, MacDougal 17; MacDougal Crater, Pin- acate Mountains, November 14, 1907, Sykes 28. Croton arenicola Rose & Standley, sp. nov. Low shrub, less than a meter high, much branched, the stems strictly erect, whitish, slender, densely lepidote throughout; leaves linear to linear-oblong or lanceolate, 20 to 35 mm. long, 2 to 6 mm. wide, rounded at the apex, attenuate to the base, densely lepidote-stellate on both surfaces, whitish, on slender petioles 5 to 14 mm. long; flowers dicecious, both kinds apetalous; staminate flowers in few-flowered racemes 15 to 30 mm. long, naked below, the calyx lobes ovate, densely stellate and lepidote, obtuse, the flowers 4 mm. broad, on pedicels 5 to 8 mm. long; stamens slightly exceeding the sepals; pistillate raceme about 3 cm. long, sometimes less, the flowers on stout pedicels 4 to 6 mm. long, the calyx lobes ovate, obtuse; capsule 10 or 11 mm. high, densely and finely stellate and somewhat lepidote; seeds oval or oblong, 7 or 8 mm. long, variegated with brown and gray, the caruncle stipitate, small. Type in the U. 8. National Herbarium, no. 574267, collected on sand hills about Adair Bay, Gulf of California, in northwestern Sonora, November 20, 1908, by Mr. G. Sykes (no. 62). , This is near Croton tenuis but has more abundant pubescence so that the plant appears silvery throughout; the leaves are also narrower, and the seeds are much larger with a different caruncle. Ditaxis odontophylla Rose & Standley, sp. nov. Low, 20 cm. high or less, erect or ascending; stems stout, pilose; leaves oblanceolate, attenuate at the base into a short petiole, rather thin, bright green, more or less pilose on both surfaces, broadly obtuse and coarsely dentate near the apex; staminate flowers with linear-oblong sepals and oval, clawed petals, the latter white tinged with purple near the base, the sepals pilose; pistillate flowers with linear-lanceolate, hirsute sepals, the style tips not enlarged; capsule strongly hirsute; seeds subspherical, smooth, brown. Type in the U. 8. National Herbarium, no. 574248, collected at the Papago Tanks, Sonora, November 14, 1908, by Dr. D. T. MacDougal (no. 36). The plant is similar to Ditaris neomexicana but has very different leaves and much more abundant pubescence. Here may be inserted a description of another apparently new species of Ditaxis, detected while attempting to determine Ditaxis odontophylla. Diraxis Gractuis Rose & Standley, sp. nov. Low, sparingly branched, slender annual, 30 to 40 cm. high; stems sparingly pilose, pale green; leaves lanceolate or elliptic-lanceolate, 5 to 6 cm. long and 20 to 25 mm. wide, acute, somewhat attenuate at the base, thin, bright green, all except the youngest glabrous, all on slender petioles 10 to 18 mm. long; racemes few-flowered and PLATE 4. 16, Contr. Nat. Herb., Vol, ABUTILON MACDOUGALII ROSE & STANDLEY. Contr. Nat. Herb., Vol. 16. PLATE 5. S i SPHAERALCEA MACDOUGALII ROSE & STANDLEY. ROSE AND STANDLEY—-PLANTS FROM THE PINACATE REGION. 18 axillary; staminate flowers with linear, acute, sepals, their pale green petals broadly lanceolate and one-half longer; pistillate flowers with linear-lanceolate, attenuate sepals 6 to 8 mm. long, the petals spatulate, very short and inconspicuous; style tips not enlarged; capsule hirsute, the mature ones and seeds not seen. Type in the U. 8. National Herbarium, no. 45193, collected at Guaymas, Sonora, in 1887 by Dr. Edward Palmer (no. 624). A very distinct species, readily separated by its large, thin, finally glabrous leaves with slender petioles. Another species of Ditaxis from Lower California seems never to have been referred to the correct generic name: DitaXIS BRANDEGEI (Millspaugh) Rose & Standley. Argythamnia brandeget Millspaugh, Proc. Calif. Acad. II. 2: 220. 1889. Type locality, ‘‘San Gregorio,’ Lower California. Mozinna spathulata Orteg. Hort. Matr. Dec. 8: 105. pl. 13. 1799. Jatropha spathulata Muell. Arg. in DC. Prodr. 15 ?: 1081. 1866. Type locality, ‘‘Habitat in Nova Hispania.” Hornaday Range, Pinacate Mountains, November 14, 1907, MacDougal 21. Poinsettia eriantha (Benth.) Rose & Standley. Euphorbia eriantha Benth. Bot. Voy. Sulph. 51. 1844. Type locality, ‘‘Bay of Magdalena,” Lower California. Pinacate Mountains, at 600 to 900 meters, November 21, 1907, MacDougal 69. Stillingia linearifolia 8. Wats. Proc. Amer. Acad. 14: 297. 1879. Type locality, ‘‘S. California; near Boundary Monument, San Diego.” MacDougal Pass, Pinacate Mountains, November 14, 1907, MacDougal 59. Abutilon macdougalii Rose & Standley, sp. nov. PLATE 4, Herbaceous throughout, tall, probably about a meter high; stems stout, much branched, densely covered with soft, short, spreading hairs; leaves broadly ovate- cordate, 9 cm. long or usually smaller, the sinus closed, irregularly serrate, thick, velvety-tomentose on both sides, canescent beneath, all on petioles longer than the blades; inflorescence a terminal, sparingly branched panicle, the flowers on pedicels 10 to 15 mm. long; lobes of the calyx triangular-ovate, attenuate, divided two-thirds of the way to the base, densely villous; petals orange yellow, 20 mm. long, more than twice as long as the calyx; carpels slightly surpassing the calyx, 10 in number, villous, with conspicuous, divergent, rather long beaks; seeds brown, glabrous, papillose. Type in the U. 8. National Herbarium, no, 574255, collected in the Pinacate Moun- tains, November 22, 1907, by Dr. D. T. MacDougal (no. 47). Near Abutilon aurantiacum but with different inflorescence and seeds, and with shorter calyces with narrower lobes. EXPLANATION OF PLATE 4.—Branch of the type specimen. Natural size. Hibiscus denudatus Benth. Bot. Voy. Sulph. 7. pl. 3. 1844. Type locality, ‘‘Bay of Magdalena,’’ Lower California. Papago Tanks, Pinacate Mountains, November 20, 1907, MacDougal. Sphaeralcea macdougalii Rose & Standley, sp. nov. Puiate 5. Stems stout, erect, branched, densely velvety-stellate; petioles 15 to 20 mm. long; leaf blades ovate, obscurely 3-lobed, obtuse, cordate at the base, densely velvety- stellate on both surfaces, prominently veined, the margins somewhat undulate; flowers few, in short terminal racemes; pedicels 1 cm. long or less; bracts subulate, incon- spicuous; calyx 10 to12 mm. high, cleft nearly to the base, the lobes oblong-lanceolate, acute, densely stellate; petals 2 cm. long, purplish red; immature carpels densely stellate on the back, 2-seeded, blunt, nearly smooth on the inner faces. 14 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Type in the U. 8. National Herbarium, no. 574253, collected at the Papago Tanks in the Pinacate Mountains, Sonora, November 16, 1907, by Dr. D. T. MacDougal (no. 45). EXPLANATION OF PLATE 5.—Branch of type specimen. Natural size. Elaphrium microphyllum (A. Gray) Rose, N. Amer. Fl. 25%: 250.1911. Puare 6. Bursera microphylla A. Gray, Proc. Amer. Acad. 5: 155. 1861. Terebinthus microphylla Rose, Contr. Nat. Herb. 10: 120. 1906. Type locality, Lower California. Hornaday Range, Pinacate Mountains, November 14, 1907, MacDougal 22; slope of the Pinacate Mountains, November 20, 1907, MacDougal 55. EXPLANATION OF PLATE 6.—From a photograph by Dr. D. T. MacDougal. Fouquieria splendens Engelm. in Wisliz. Mem. North. Mex. 98. 1848. Type locality, ‘Jornada del Muerto,’’ New Mexico. This is illustrated in several plates of the ‘‘Camp-Fires.’’!_ Concerning this charac- teristic desert plant Doctor Hornaday writes: ? ‘There is one other arboreal feature of the deserts which, because of its picturesque oddity, I have reserved to the last. It is a product of the plant world unique in character, and standing as much apart from related genera and species as does the prong-horned antelope among hoofed animals, It is the Ocotillo, the Spanish name of which is pronounced o-co-tee’-yo. Next to the giant cactus, it was the most monu- mental and picturesque thing of plant growth found by us in two hundred miles of fertile deserts. “The ocotillo is a multiform tree, and there is nothing else that is at all like it. Instead of having a tall main stem and many branches, large and small, it has an exceedingly short stem and many very long, wandlike branches. The leaves grow all along each branch, from bottom to tip. The stem is a big, thick mass of solid wood, all underneath the earth (where the earth has not been blown away), and the top of it is large enough to afford holding ground for each branch. From the very limited upper surface of the main stem, starting usually at the level of the ground, there rise a score or more of long, slender rods of light wood, their bases firmly packed together, but otherwise free. They are like slender and very symmetrical fishing rods. As they rise they droop outward and spread apart, until they form a group shaped like a morning-glory vase. When it is in full leaf, the ocotillo is like a bouquet of green wands held at the bottom by an invisible hand. ‘The stems vary in number from three to seventy-three or even more. I can vouch for the last-named number by count. The largest ocotillo that I particularly noted had some stems that were, by measurement, eighteen feet long. ‘‘One of the strangest features of this odd multiple-tree is its leaves and thorns. The leaves grow thickly all along the stem, each blade an inch and a half in length. The blade springs full-fledged from the upright woody stem, with no free petiole, and its color is dark pea green. This profusion of leaves gives each stem of the ocotillo a highly pleasing appearance, and denotes water in the not-far-distant yesterday. A large ocotillo in full leaf is a beautiful object, and every line of its ensemble bespeaks development in a land of queer things. ‘But mark the transformation. ‘‘When the last rain has become only a distant memory, when the hungry roots have sucked the last drop of moisture from the sandy soil, the hour for the change has struck. Fleshy leaves an inch and a half long are far too luxuriant to last long in a 1 On the plate opposite page 52 several fine specimens are shown, photographed in the Ajo Valley 10 miles south of Montezumas Head, Arizona. A young plant in full leaf is shown opposite page 80; a plant in full leaf, in color, opposite page 100; an- other opposite page 230. 2 Page 49. Contr. Nat. Herb., Vol. 16 PLATE 6. (A. GRAY) ROSE. ELAPHRIUM MICROPHYLLUM PLATE 7. "3SOY ® NOLLIG WIZ9NQ) VSLNVSID VSIDSNUVO PLATE @. Add PLATE 9. “IGWNY (Addvd INNVWISUNG SNSYHSOONIHOF ROSE AND STANDLEY—PLANTS FROM THE PINACATE REGION, 15 desert. They dry up, and they drop off—all but the midrib, which takes form as a big, woody thorn an inch or more in length. Then and thereafter each stem presents the most frightful array of thorns to be found on anything outside the cactus family. So far as cattle, burros, and wild animals are concerned, an ocotillo in a state of defence is practically impregnable. We saw only two stems that had been barked by food- seeking animals, and that work had been done by wild burros, at great trouble and expense, “Except on the plains dedicated to the creosote bush and mesquite, the ocotillo stayed with us from Tucson to the very foot of Pinacate Peak. It is the inseparable companion of the giant cactus, but, unlike the latter, it grows larger along the inter- national boundary than fifty miles farther north. On the night that three of us ‘laid out” on the slope of Pinacate, we found near our bivouac a large dead ocotillo whose rods of clean white wood burned with a brilliant light—too bright to last. These naked rods are used by the Papago Indians in building fences, and screens around the verandas of their adobe houses.”’ Petalonyx thurberi A. Gray, Mem. Amer. Acad. IT. 5: 319. 1855. Type locality, ‘‘ Valley of the Rio Gila.”’ Sandhills, Adair Bay, November 20, 1907, Sykes. Sympetaleia rupestris (Baill.) A. Gray; 8. Wats. Proc. Amer. Acad. 24: 50. 1889. Loasella rupestris Baill, Bull. Mens. Soc. Linn. Paris. 1: 650. 1886. Type locality not ascertained. Pinacate Mountains, November 21, 1907, MacDougal 74 Carnegiea gigantea (Engelm.) Britton & Rose, Journ. N. Y. Bot. Gard. 9: 188. 1908. PLATE 7. Cereus giganteus Engelm. in Emory, Mil. Reconn. 158. 1848. Type locality, along the Gila River, Arizona. No specimens were collected but many fine photographs were taken. The finest specimen seen is illustrated by Doctor Hornaday in a beautiful colored plate.' It was about 60 feet high and had 9 branches, an- unusually large number. The species was found to range from Tucson to the Pinacates, and from near sea level to an altitude of 1,200 meters. EXPLANATION OF PLATE 7.—From a photograph by Dr. D. T. MacDougal. Echinocactus emoryi Engelm. in Emory, Mil. Reconn. 156. 1848. Type locality not specifically given; in southeastern Arizona, near the New Mexican line. A living specimen was collected in the Pinacate Mountains and sent to Washington. Echinocactus wislizeni Engelm. in Wisliz. Mem. North. Mex. 96. 1848. Type locality, ‘‘Dofiana,’’ New Mexico. No material of this species was taken but Doctor Hornaday gives an illustration ? of a very large plant from which water is being extracted. Echinocereus engelmanni (Parry) Riimpl.; Férst. Handb. Cact. ed. 2. 805. 1886. PLaATEs 8, 9. Cereus engelmanni Parry, Amer. Journ. Sci. II. 14: 338. 1852. Type locality, ‘Mountains about San Felipe,’’ California. A cluster of living specimens was collected on the Pinacate Mountains and sent to Washington. Doctor Hornaday illustrates a group of these.’ EXPLANATION OF PLATES 8, 9.—From photographs by Dr. D. T. MacDougal. 1 << Camp-Fires,’’ facing page 72. 2 Thid., facing page 216. 3 Thid., facing page 236. 16 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Lemaireocereus thurberi (Engelm.) Britton & Rose, Contr. Nat. Herb. 10: 426. 1909. Cereus thurberi Engelm. Amer. Journ. Sci. II. 17: 234. 1854. Type locality, ‘‘Canyon near the mountain pass Bachuachi.” A living specimen and fruit were collected at Sonoyta, Sonora, and sent to Wash- ington. A fine colored illustration of this species is given by Doctor Hornaday.' The plant illustrated had 22 stems, the tallest being 20 feet high. The species was first seen at Sierra Blanca and last at the Ajo mines, Arizona, at about the same latitude as Tucson, and these may be considered as the northern limits for the species. Lophocereus schottii (Engelm.) Britton & Rose, Contr. Nat. Herb, 12: 427. 1909. Cereus schottit Engelm. Proc. Amer. Acad. 8: 288. 1856. Pilocereus schottti Lem. Rev. Hort. 1862: 428. 1862. Type locality, “Toward Santa Magdalena,’’ Sonora, Mexico. A living specimen was collected at Sonoyta, Sonora, and sent to Washington. Mamillaria grahami Engelm. Proc. Amer. Acad. 3: 262. 1856. Type locality, ‘Mountains from El Paso southward and westward to the Gila and Colorado, and up the latter river.’’ Living specimens were collected on the Pinacate Mountains and sent to Washington. Opuntia bigelovii Engelm. Proc. Amer. Acad. 3: 307. 1856. Prare 10. Type locality, “On Williams River’’ (Bill Williams River), Arizona. Common on Hornaday Mountain, Sonora. EXPLANATION OF PLATE 10,—From photograph by Dr. D. 'T. MacDougal. Opuntia chlorotica Engelm. & Bigel. Proc. Amer. Acad. 3: 291. 1856. Type locality, “From San Francisco Mountains to Mojave Creek,” Arizona. | Near summit of Pinacate Mountains, MacDougal. Opuntia fulgida Engelm. Proc, Amer. Acad. 8: 306. 1856. Type locality, “ Mountains of western Sonora,’’ Mexico. Sonoyta, Sonora, MacDougal. Philibertella hartwegii heterophylla (Engelm.) Vail, Bull. Torrey Club 24: 308. 1897. Sarcostemma heterophylla Engelm. in Torr. U. 8. Rep. Expl. Miss. Pacif. 5: 362. 1856. Type locality, “Near Fort Yuma,’’ Arizona. Walls Well, Ajo Mountains, Arizona, November 7, 1907, MacDougal 7. Cuscuta californica Choisy, Mem. Soc. Phys. Hist. Nat. Gendve 9: 279. 1841. Type locality, “Nov. [am] Californiam.”’ — MacDougal Crater, Pinacate Mountains, November 14, 1907, MacDougal 26. The immature plants are growing on Kallstroemia grandiflora. Euploca aurea Rose & Standley, sp. nov. PLATE 11, Low, much branched annual, 30 cm. high or less; branches spreading, stout, hirsute; leaves oblong to elliptic or oval, thick, yellowish green, hirsute, small, mostly about 1 cm. Jong, acutish, rounded at the base, all on short, stout petioles one-third as long as the blades; flowers axillary, scattered; lobes of the calyx linear- subulate, strigose; corolla bright yellow, the limb about 6 mm. wide, the throat some- what inflated; style long and slender; stigma penicillate; achenes 2, hemispherical, each finally splitting into 2, strigillose, smooth. Type in the U. S. National Herbarium, no. 574265, collected on sand hills near Adair Bay, Gulf of California, November 20, 1907, by Mr. G. Sykes (no. 61). This is not closely related to any other species of the genus, being distinguished chiefly by its yellow corollas and bristling indumentum. EXPLANATION OF PLATE 11.—a, Branch of type; b, calyx lobe. a, Natural size; b, scale 4. ' “Camp-Fires,’’ facing page 136. PLATE 10. Nat. Contr. ‘WIS9N9 NAOTSDIA VILNAdO , obtusely and abruptly constricted below and long-decurrent, a narrow decreasing foliaceous wing (1 to 4 mm. broad) extending downward upon each side of the rachis (above) to the pinna next below; characteristic middle pinne subsessile, spreading, unequally and obtusely cuneate, the short petiolule distinctly foliaceo-marginate, tt.e decurrent wing about 2 mm, broad at origin, narrowed downward; coste elevated, yellowish, sparingly clothed toward the base below with whitish membranous, rounded or ovate, erose scales; margins remotely and lightly crenate-serrate, the teeth at the middle of the pinne about 6 to 8 mm. broad and | mm, long; main veins about 25 pairs or fewer, oblique, 5 to 9 mm. apart, clevated below, slender, with about 4 pairs of similar veinlets, the basal ones of adjacent groups joined by an obliquely transverse veinlet (forming a narrowly pentagonal elongate costal areole, cuneate proximally, much broader distally), then excurrent toward the margin, one or both extending to the minute sinus; other veinlets very oblique, nearly parallel, excurrent to the margin; sori few, occupying a slightly inframedial zone between the costa and the margin, in the partially infertile specimen at hand confined to the basal veinlets (above the areole and distant about 2 mm. from it) and the second pair of veinlets; indusium proximal, whitish, lobate, erose; leaf tissue rigidly membrano-herbaceous, lustrous above, pale below. Tyre Locauity: Mountain forests near Lobani, District of Chinantla, Oaxaca, Mexico, altitude 900 to 1,050 meters, Liehbmann. DisTRIBUTION: Known only from Oaxaca, The material at hand, which was received from Copenhagen, comprises a short sec- tion of the rachis, to which are attached the fourth pair of pinnae, With this was sent by Mr. Christensen a sketch of the upper pinne showing the decurrent wings which extend along the rachis. It is sufficient to establish the validity of the species, which must be reckoned one of the most peculiar in the genus, 1 Hook. & Baker, Syn. Fil. 28. 1868. Contr, Nat. Herp, You. 16, PLaTe 24. Four SPECIES OF HEMITELIA. MAXON—STUDIES OF TROPICAL AMERICAN FERNS, 39 14. Hemitelia mexicana Liebm. Vid. Selsk. Skr. V. 1: 287. 1849. Pate 24, b. Hemistegia mexicana Fourn, Mex. Pl, 1: 135. 1872. Caudex about 30 cm. high; fronds 1.8 to 2.25 meters long; stipe 30 to 45 cm, _ long, squamulose, at length glabrescent, toward the base very minutely aculeolate; lamina broadly lanceolate, 1.5 to 1.8 meters long, 45 cm. or more broad, simply pinnate, acuminate; pinne numerous, alternate, oblique, linear, about 30 cm. long, 2.5 to 3 em. broad, straight or slightly falcate, sessile, not narrowed at the inequilateral base, obliquely truncate at the upper side, rounded below, deeply crenate about one-third the distance to the costa at the base, about one-fourth the distance to the costa (or less) at the middle, only the outer fourth obliquely and broadly crenate-serrate, the apex appressed-serrulate; major crenations 3 to 5 mm. long, 5 to 7 mm. broad at the base, rounded-subtruncate, short-apiculate distally, the margins minutely revolute, entire; cost elevated, greenish brown, glabrescent, a very few minute flattish yel- lowish scales subpersistent at the sides; main veins 50 pairs, nearly at right angles, 4 to 7 mm. apart, elevated, very slender, with 5 to 8 pairs of simple oblique arcuate veinlets, the basal of these usually joined by a transverse veinlet, then (one or both) excurrent to the sharply acute sinus, the other veinlets also extending to the margin, but mostly beyond the sinus; veins and veinlets minutely squamulose-setiferous; areoles relatively broad, ample, subhexagonal, broadest distally; sori 2 or 3 pairs to each group of veinlets, occupying a slightly inframedial zone between the costa and margin of the pinne; indusium light brown, semicircular or commonly suborbicular, crenately 3 to 6 lobed, the lobes repand-pateriform, reflexed; receptacle capitate, squamulose; leaf tissue membrano-herbaceous, dark shining green above, lighter below. Type LocaLity: Mountain forests near Cacolé (printed Cacoba, in error), District of Chinantla, Oaxaca, Mexico, altitude 750 to 900 meters, Iiebmann. DistRIBUTION: Known definitely only from the original collection. The above description is drawn partly from two middle pinne (with a portion of the rachis) of the type specimens, which have been received from the Botanisk Museum, Copenhagen, through the kindness of Mr. Carl Christensen; and partly from Liebmann’s original diagnosis. The species is well marked and will probably be found not uncommon in the mountainous districts of Oaxaca at mid-elevations. EXPLANATION OF PLATE 24.—Portions of characteristic pinne (a) of type specimen of Hemitelia guate- malensis; (b) of type specimen of H. mezicana; (c) from type collection of H. (wcida; (d) of type specimen of H. choricarpa. All natural size. 15. Hemitelia lucida (I*ée) Maxon. PLATE 24, ¢. Hemistegia lucida Fée, Gen, Fil. 351, 1850-52. Caudex undescribed; stipe very stout, somewhat spiny, paleaceo-furfuraceous; lamina (estimated) about 2 meters long, apparently ovate-oblong, about 80 cm. broad near the middle, 40 em. broad at the base; pinnsze numerous, divergent, the larger ones about 8 cm, apart on each side, ligulate-lanceolate, straight or upwardly faleate, up to 45 cm. long, 3.5 to 4.2 em, broad, serrate to serrulate at the gradually tapering, long-acuminate apex, elsewhere crenate to crenately lobed; principal cre- nations or lobes 28 to 34 pairs, rounded, 9 to 12 mm. broad at the sinuses, less than one-half as long, rounded, directed forw ard and subrectangular at the distal border, the sinuses obliquely triangular, open and acute, the margins subentire, minutely revolute; coste elevated, glabrous above, nearly glabrous below, there bearing a few minute deciduous dirty white scales; midveins (or costules) divergent from the costa, subopposite to alternate, 9 to 10 mm, apart on each side, elevated, glabrous upon both surfaces, discontinuous (7. e., forked) at the apex of the lobes, below this nearly straight; veinlets 5 to 9 pairs, oblique, simple, the lowermost ones of adjacent crenations joined by a transverse veinlet (this distant 2 to 2.5 mm. from the costa), then immediately soriferous and produced, commonly joining the second oblique vein of the same group at a point about one-half the distance to the sinus (or beyond) 40 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. at an acute angle; third pair of veins usually free and excurrent to or above the sinus; succeeding veins free, closer, extending to the margin above the sinus; sori 4 to 9 pairs, large, the basal ones 3 to 4 mm. distant from the midvein and 2 to 3 mm, from the costa, those above gradually approaching the midvein, the uppermost. basal upon the veinlets, the sori of each lobe or crenation thus borne in an elongate A-shaped line; indusium ample, irregularly repand-lobate, the lobes shallow and reflexed: recep- tacle relatively large, globose, squamulose-setiferous; leaf tissue membrano-herba- ceous, dark green and shining above, much paler below. Typx Locauity: District of Chinantla, Oaxaca, Mexico, altitude about 2,000 meters, DisrriBuTION: Known only from the original collection, Galeotti 6537. Fée’s original diagnosis is very incomplete and has afforded scant data for the above redescription, which is based almost wholly upon material of the type col- lection forwarded to the U, 8, National Museum from Brussels by Professor de Willde- man. This species is obviously related to H. guatemalensis, but appears to differ constantly in its deeper crenations and the very frequent junction of the first and second veinlets of each group at an acute angle below the sinus. EXPLANATION OF PLATE 24.—See p. 39. 16. Hemitelia guatemalensis Maxon, sp. nov. ; Pate 24, a, Caudex and stipe wanting; lamina ample, presumably about 2 meters long, at least 60 cm. broad, the apex very abruptly short-acuminate, the rachis stout, 6 to 8 mm, broad, firm, compressed, narrowly canaliculate ventrally, brownish, closely grayish squamulose-pubescent; pinnz alternate, spaced, divergent, faleate and decurved, ligulate-lanceolate, the larger ones about 32 cm. long, 3 to 3.5 em. broad, sessile or stoutly short-petiolate, slightly narrowed at the unequally rounded subcordate base, or obtusely cuneate at the upper side, irregularly and obliquely crenate nearly throughout, only the apical fourth serrate, gradually serrulate at the long-acuminate apex; crenations 2 mm. or rarely 3 mm. long, 6.to 8 (casually 10) mm. broad at the base, rounded, acutish distally, the margins entire, closely revolute, the sinuses acute or acutish; costae very stout, elevated upon both sides, below similar to the rachis, above (together with the veins, veinlets, and leaf tissue above) glabrous; main veins about 40 pairs, the venation (including areolation) otherwise similar to that of H, mexicana, except as to the number of veinlets (these 4 to 6 pairs to each group); sori 4 to 6 pairs, distinctly inframedial upon the véinlets; indusium light brown, irregu- larly semicircular, 2 or 3-lobate, the lobes pateriform; receptacle elongate-capitate, minutely squamulose-pubescent; leaf tissue rigidly membrano-herbaceous, light green and lucid above, very much paler below. Type in the U. 8. National Herbarium, no. 830363, collected in Guatemala by O. Salvin (without number), distributed from the Royal Gardens, Kew, as H. subincisa. Hemitelia guatemalensis is to be compared with H. mericana and H. lucida. Only the type specimens have been seen. EXPLANATION OF PLATE 24.—See p, 39. 17. Hemitelia choricarpa Maxon, sp. nov. PLATE 24, d. Caudex and stipe wanting; lamina apparently about 1.5 meters long, 40 to 50 em. broad, pinnate, acuminate, the rachis very firm, rather slender, yellowish brown, sharply bicarinate-sulcate and glabrous above, the under surface loosely crispate- tomentulose, the persistent whitish or yellowish hairs short and somewhat spreading; pinne subopposite to nearly alternate, 5 to 6 cm. apart on each side, oblong-linear, ses- sile, the upper ones ascending, obtusely and subequally cuneate or at. the upper side slightly excavate; middle pinne divaricate or ascending, rounded-truncate or sub- cordate at the base, 27 to 32 cm. long, 3.5 to 4 cm. broad, pinnately lobed at the base about one-half the distance to the costa, above this (and nearly throughout) about two- fifths the distance to the costa, beyond this crenate, the acuminate-attenuate extremity obscurely serrulate; costee conspicuously elevated on both surfaces, glabrous above, PLATE 25. 16, Contr. Nat. Herb., Vol. HEMITELIA GRANDIFOLIA WILLD. MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 41 below crispate-tomentulose, glabrescent; major lobes about 27 to 32 pairs, subequal, 7 to 9 mm. broad at the base, obtuse, close, separated by narrow nearly closed sinuses, the margins slightly revolute, subentire, the costules elevated, arcuate toward theapex, glabrous above, glabrate below; veins 10 to 14 pairs to the lobe, oblique, the lowermost ones of adjacent segments joined by a transverse veinlet (forming a narrow elliptical areole), with 1 or 2 veinlets irregularly soriferous and excurrent to the sinus, there meeting the two basal veins, the other veins simple or mostly once-forked below the middle, soriferous at or below the fork or rarely (in the case of the lower veins) beyond the fork; sori distinctly apart, forming a slightly inframedial line extending from below the apex downward nearly to the costa, there joining the sori of the adjacent lobe ata point remote from the sinus, thus forming one end ofa nearly perfect ellipse; indusium ample, membranous, semicircular in outline, 2 or 3-lobed, the lobes shallow, spreading, irregular; receptacle capitate, setiferous; leaf tissue membrano-herbaceous, dark and lustrous above, paler below. Type in the U.S. National Herbarium, no. 830322; collected in forest near Buenos Aires, Costa Rica, February, 1892, by H. Pittier (no. 4835). Hemitelia choricarpa is apparently confined to Costa Rica, the only other specimens seen being from Cafias Gordas, altitude 1,100 meters, March, 1897, Pittier 10966. The characters depended upon in the key to distinguish it from related species are diag- nostic, EXPLANATION OF PLATE 24,—See p. 39. 18. Hemitelia grandifolia (Willd.) Spreng. Syst. Veg. 4: 125. 1827. = PLaTE 25, Cyathea grandifolia Willd, Sp. Pl. 5: 490, 1810. Hemitelia imrayana Wook. Icon. Pl. 7: pl. 669. 1844. Hemitelia horrida imrayana Hook. in Hook, & Baker, Syn. Fil. 28, 1868. Hemistegia willdenovit Fée, Gen, Fil, 351. 1850-52. Microstegnus grandifolius Presl, Abh. Béhm, Ges, Wiss, V. 5: 354. 1848. Hemistegia insignis Fée, Mém,. Foug. 11:99. 1866. Hemitelia insignis C, Chr. Index Fil. 349. 1905, Caudex arborescent, erect, frequently 4 to 5 meters high, about 10 cm. in diameter, at the base (with its copious covering of brownish flexuous adventive roots) about 15 cm. in diameter, densely clothed at the summit with whitish lanceolate scales; fronds ascending, arching, up to 1.7 meters long, the stipes very stout, more or less imbricate, adnate or ascending close to the caudex, whitish-paleaceous at. the base, eventually deciduous, leaving definite spaced quincuncially arranged elongate-oval scars; lamina ample, ovate, 1 to 1.25 meters long or more, 60 to 80 cm. broad, short- acuminate, the rachis very stout, yellowish brown, distantly muricate to smooth, convex or lightly sulcate on the lower side, on the upper side (at least in the lower part) deeply and narrowly sulcate, the ridges rounded, each similarly sulcate at the outer side; scales of the rachis deciduous, ovate or deltoid-ovate, long-acuminate, whitish, with a bright brown median stripe, finely erose; pinnze opposite or suboppo- site, the lowermost somewhat deflexed, ovate-lanceolate, about 25 cm. long; middle pinne spreading, 8 to 10 cm. apart on each side, very narrowly deltoid-lanceolate to narrowly oblong-lanceolate, 30 to 40 cm. long, 6 to 11.5 cm. broad, sessile, scarcely or not at all reduced at the base, at the lower side strongly imbricate upon the rachis, close above, pinnatifid nearly to the costa, or the basal segments free, above this the segments connected by a narrow gradually increasing costal wing 2 to 4 mm. broad or finally about 4 to 6 mm. broad on each side below the crenate-serrate, ultimately biserrate, acuminate apex; costs very stout, yellowish brown to castaneous, glabrous above, below (especially toward the base) freely paleaceous, the scales similar to those of the rachis but smaller and relatively broader, the brownish median stripe often obsolete; segments adjacent or somewhat apart, 25 to 34 pairs, oblong-lanceolate, sometimes dilatate, the inferior basal one often reduced, inequilateral, and invariably 42 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. clasping the rachis, the others falcate or subfalcate, 3 to 6 cm. long, 12 to 16 mm. broad at the sinuses, slightly narrower above, the margins lightly revolute, obliquely crenate to deeply crenate-serrate (in very large specimens), crenate-serrulate at the acuminate to long-acuminate apex; costules elevated, bearing numerous small white bullate scales in the lower part below, otherwise glabrous; veins 15 to 20 pairs below the apex, minutely setulose, subpinnately forked (with about 2 to 4 pairs of branches) or 2 to 5-forked, the branches oblique, arcuate, soriferous toward their extremities; lower- most branches of basal veins of adjacent segments usually joined by a transverse veinlet, forming a narrowly elongate costal areole; basal branches of veins in the basal third of the segment also infrequently joined similarly, forming relatively broad costu- lar areoles; sori rather small, biseriate and slightly supramedial, or in larger segments forming a continuous regular and deeply crenate line about 1 to 2 mm. distant from the margin; indusium bright brown, simple or 2 or 3-lobed, the margins uneven to lacerate; receptacle capitate to subcylindric, setose; leaf tissue firmly herbaceous, lustrous, often discolored in drying, dark above, much lighter below. Type LocaLiry: Martinique (herb, Willd. 20167). DistriBuTION: Apparently confined to the Lesser Antilles—Dominica, Martinique, Guadeloupe, Montserrat, and St. Kitts—at 300 to 1,200 meters elevation. Intusrrations: Hook. Icon. Pl. pl. 669 (as Hemitelia imrayana), Fée, loc. cit. pl. 26 (as Hemistegia insignis). The following specimens are in the U. 8. National Herbarium: MARTINIQUE: Duss ! 1605 (as H. horrida); Duss 4435 (as Hemitelia sp.); Duss 4605 (as H. grandifolia). GUADELOUPE: Duss 4154 (as H. grandifolia); Duss 4155 (as H. horrida); Duss 4449 (as H. insignis); Duss 4451, 4452 (as Hemitelia sp.). Dominica: In forests, Laudat, Eggers 867; Laudat, Lloyd 263. Sr. Krrrs: Forested slopes of Mount Misery, Britton & Cowell 510. MonTserRRAT: Chauers Mountain, altitude 600 meters, Shafer 283. Without locality, Turner. The taxonomic history of this species, which is rather complicated, is briefly as— follows: (1) Grandifolia. The species was first described as Cyathea grandifolia by Willdenow who cited Plumier’s plate 20 and Petiver’s plate 2, figure 10, and gave as the sole locality Martinique. If we are to interpret the species wholly upon the basis of the illustrations cited the name will apply to no other species than that described later by Kunze as H. kohautiana, the type of which (Sieber’s 375, from Martinique) will be seen (PI. 26) to agree closely with Plumier’s plate 20, the latter also representing a Martinique plant. But there is in this instance a Willdenovian type specimen (herb. Willd., no. 20167) of the species grandifolia; and this, by a careful reading of the original description and especially of the part describing the acuminate segments, will be seen to have served for the really diagnostic features of the description. It seems far preferable, therefore, to give greater weight to the specimen than to the figures cited; and this even though a later writer, Pres], has confused the matter by stating ? (by implication) that Willdenow’s type was from Caracas, collected by Bredemeyer. The Willdenow specimen (no. 20167) shows no such data,? and there is far better reason to credit Willdenow’s statement than Presl’s. The other locality cited by Pres] for his Microstegnus grandifolius is Mount Misery, St. Kitts, the specimen col- lected by Breutel. This the writer has not seen; but specimens collected on this 1 The Duss numbers frequently embrace more than one species. The numbers here listed apply only to specimens in the National Herbarium. ? Loc. cit., 354, in describing Microstegnus grandifolius. * Professor Urban writes that it has merely the following locality data: ‘‘Habitat in America calidiore.”’ MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 43 mountain by Britton and Cowell (no. 510, in 1901) are evidently not specifically dif- ferent from the fragment of Willdenow’s type (no. 20167) shown in plate 25, a; and there is, therefore, on grounds of geographic distribution a strong presumption that Presl was in error as to the origin of the Willdenow specimen. Fée’s change of name for the plant to Hemistegia willdenovii was made purely from a supposed nomenclatorial difficulty, arising from his desire to retain both Micro- stegnus grandifolius Presl and Hemistegia grandifolia Presl (‘‘Hemitelia grandtfolia Hook.’’) in the same genus; unfortunately he in error chose the true grandifolia for renaming. (2) Imrayana. Described from Dominica by Hooker on plants collected by Doctor Imray. A specimen at hand from this island differs in no essential particular from more copious material from Guadeloupe and Martinique. (3) Insignis. A name applied by Fée to specimens collected by l’Herminier in Guadeloupe. A considerable number of specimens from Guadeloupe and Martinique give us our best idea of this species. They are clearly of the same species as Will- denow’s type. Hemitelia grandifolia is sufficiently distinct from H. kohautiana and H. obtusa by the data given in the key. This has been made as full as possible for the purpose of pointing out very definitely the characters by which these species, which have so long been confounded by nearly every writer, may be adequately distinguished. EXPLANATION OF PLATE 25.—a, Fragment from the type specimen, herb. Willdenow (no. 20167), from Martinique; b, tip of a small upper pinna of a Martinique specimen, Duss 1605; c, tip of a large lower pinna of a Guadeloupe specimen, Duss 4452. 19. Hemitelia horrida (L.) R. Br.; Spreng. Syst. Veg. 125, 1827. Polypodium horridum L. Sp. Pl. 1092. 1753. Cyathea horrida J. E. Smith, Mem. Acad. Turin 5: 416. 1793. Cyathea commutata Spreng. Anleit. Kennt. Gewiichse 3: 146. 1804 (excluding all reference to Plumier). Cnemidaria horrida Presl, Tent. Pterid. 57. 1836. Actinophlebia horrida Presl, Abh. Béhm. Ges. Wiss. V. 5: 356. 1848. Hemistegia horrida Fée, Gen. Fil. 351. 1850-52. Hemitelia hookeri Presl, Abh. Béhm. Ges. Wiss. V. 5: 350. 1848. Hemitelia hookeriana Schlecht. Bot. Zeit. 14: 474. 1856. ?Hemitelia acuminata Karst.; Schlecht. Bot. Zeit. 14: 474. 1856. Hemitelia commutata Schlecht. Bot. Zeit. 14: 474. 1856. Hemistegia repanda Fée, Gen. Fil. 351. 1850-52; Mém. Foug. 11: 98. 1866. Typge Locauiry: Near Port de Paix, Haiti, Plumier. DistRIBUTION: Santo Domingo, Jamaica, Cuba, Porto Rico, Costa Rica, and doubt- fully Colombia. InLusrrations: Plumier, Traité Foug. pl. 8; Spreng. loc. cit. pl. 4. f. 82; Hook. Sp. Fil. 1: pl. 15; Hook. & Bauer, Gen. Fil. pl. 4. A redescription of this species is scarcely necessary, inasmuch as it is common in the Greater Antilles and is doubtless well represented in all the larger herbaria. Scant material of H. horrida and a consequent failure to recognize the full extent of its variation were, however, responsible for a large amount of speculation and critical comment in earlier times. Thus, Presl in 1848 founded a new species, Hemitelia hookeri, (without description) upon plate 15 of Hooker’s Species Filicum and plate 4 of his Genera Filicum, supposing these to illustrate a species distinct from H. horrida, whereas they represent a condition not infrequently observed in particularly robust individuals of that species. Presl cited also Venezuelan specimens collected by Linden (no. 1572); but if we regard his species as typified by plate 15, as seems proper, it becomes a straight synonym of H. horrida, whatever may be the identity of Lin- den’s no, 1572. 44 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Schlechtendal published! in 1856 a very elaborate analysis of “‘H. horrida,” as understood by various writers and as evidenced by living specimens and the rela- tively small amount of herbarium material available. His conclusion was to recog- nize tentatively five specific forms as follows: “1. Hemitelia horrida R. Br.=Polypodium horridum Lin. Plum. Fil. p. 9. t. 8. Plum. Amer. p. 3. t. 4. 2. Hemitelia commutata=Cyathea commutata Spreng. Einleit. in d. Stud. d., krypt. Gew. 8. 147. fig. 32. a. b. . Hemitelia hookeriana=H. horrida Wook. Spec. Filic. 1. p. 30. t. XV. excel. synon. 4. Hemitelia Imrayana Hook. Tc. pl. t. 669. Sp. Filic. 1. p. 33. 0. Hemitelia acuminata Karsten in litt.=H. horrida hortorum et nonnull. auct.’’ He suggested, however, that a better knowledge of these in the future might result in their recognition merely as forms of a single highly variable species. As to the status of these it may now be said that: (1) Plumier’s plate 8 (the type of Polypo- dium horridum 1.) shows crudely but unmistakeably the typical form of the West Indian plant known commonly as /emitelia horrida; (2) H. commutata (Spreng. ) Schlecht., leaving out all reference to Plumier’s plate 14 and judging the species by Schlechtendal’s own figure, is assuredly H. horrida; (3) H. hookeriana Schlecht. is only an extreme development of the ordinary West Indian J. horrida, matching perfectly certain material from Jamaica which apparently owes its form to unusually favorable conditions of growth; (4) I. imrayana Hook. is identical with H. insignis Fée, a Lesser Antilles species? which (see page 42) must be known under the still earlier name H. grandifolia (Willd.) Spreng.; and (5) HW. acuminata Schlecht. must be regarded as doubtfully a synonym of H. horrida. The original specimens of H, acuminata (a name first applied by Klotzsch) are said to have been collected near Galipan, Colombia, by Moritz (no. 290) and were listed? as Jf. horrida. These and Valentini’s Costa Rican specimens, mentioned by Schlechtendal,* have not been seen; but that 1. horrida really occurs in Costa Rica is evident to the writer from an examination of the two Costa Rican collections listed below, There is no reason to suppose that the species may not extend also to Colombia. The presence of scales, mentioned by Schlechtendal, suggests some doubt as to the reference of /7. acumi- nata to H. horrida, however; for in undoubted H. horrida the presence of any scales whatever, even in the most immature fronds, is exceedingly rare. Except for their very delicate, thin, and readily abraded whitish-tomentulose covering, the vascular parts of the pinnge are normally glabrous. Indeed, even the thin arachnoid covering is sometimes wholly wanting. rée’s Hemistegia repanda, as redescribed some fourteen years after its original publication, is without much doubt also referable to H. horrida. The following specimens are in the U. 8. National Herbarium: Santo Dominco: Near Barahona, altitude 600 meters, von Tiirckheim 2707. Loma Isabel de Torre, altitude 600 meters, Hggers 2738. Without definite locality, Jaeger 203. . Jamaica: Road to Mooretown, above Port Antonio, Underwood 3479. Near Castleton, Underwood 86. Near Port Antonio, Fredholm 3340. John Crow Mountains, Britton 3986; Iarris & Britton 10697, 10709. Cuna Cuna Gap, altitude 750 meters, Clute 266. Second Breakfast Spring, near Tweedside, altitude 600 meters, Maxon 869. Wet rocky banks of stream and ravines in ioe) ' Bot. Zeit. 14: 449-454; 465-475. 1856. * Schlechtendal, in error, gives H. imrayana as from Santo Domingo, instead of from Dominica. 3 Linnaea 20: 440. 1847. * Bot. Zeit. 14: 465. 1856. Contr. Nat. Herb., Vol. 16 PLATE 26. HEMITELIA KOHAUTIANA (PRESL) KUNZE MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 45 forest, Mansfield, altitude 300 to 500 meters, Maxon 2387, 2456. Without definite locality, Hart 174. Cusa: Mountains near Taco Taco, Baker 3835. Arroyo Trinitario, Rio Negro, Trinidad Mountains, Santa Clara, altitude 550 meters, Britton & Britton 5187. El Yunque Mountain, near Baracoa, altitude 300 to 600 meters, Underwood & Earle 626; Pollard & Palmer 118, 154. Pinal de Santa Ana, Yateras, Oriente, altitude 800 meters, Eggers 5030. Upper slopes and summit of Gran Piedra, Oriente, altitude 900 to 1,200 meters, Maxon 4034. Josephina, north of Jaguey, Yateras, Oriente, altitude 575 meters, Maron 4104. Santa Ana, about 6 miles north of Jaguey, Yateras, Oriente, altitude 600 to 625 meters, Maxon 4187. Monte Verde, Yateras, Oriente, altitude 575 meters, Maxon 4336. Without definite locality, Wright 888. Porto Rico: Maricao, Sintenis 417. Mount Jimenez, Sierra de Luquillo, Sin- tenis 1507. Cayey, in forests, Sintenis 2490b. Adjuntas, in forests of Mount Cienega, Sintenis 4168. Lares, in forest at Buenos Aires, Sintenis 6088. Road from Utuado to Lares, Underwood & Griggs 71. Hacienda Perla, north side of Sierra de Luquillo, altitude 150 meters, Heller & Heller 1043. Mount Morales, near Utuado, Britton & Cowell 831. Near Mayaguez, Britton & Marble 551; Cowell 581. Barranquitas, Hioram 277. Costa Rica: Banks of a stream near Santa Barbara, Pittier 1679. Alajuela, alti- tude 900 meters, Alfaro 108. 20. Hemitelia kohautiana (Presl) Kunze, Bot. Zeit. 2: 298. 1844. PLATE 26. Cnemidaria kohautiana Presl, Tent. Pterid. 57. 1836, name and figure. Hemistegia kohautiana Presl, Abh. Bohm. Ges. Wiss. V. 5: 355. 1848, name only. Hemistegia grandifolia Presl, Abh. Béhm. Ges. Wiss. V. 5: 355. 1848, in part, as to Plumier reference, not Hemitelia grandifolia (Willd.) Spreng. 1827. Low-arborescent, the caudex up to 1.4 meters long and closely covered (at least below) with long dark adventitious roots; fronds numerous; lamina ample, probably 1.5 meters long or more, about 60 cm. broad, deeply bipinnatifid, acuminate, the rachis stout, brownish-stramineous, deeply trisulcate above, lightly and obtusely sulcate below, conspicuously but deciduously paleaceous, the scales linear-lanceolate to elongate- deltoid, all very long-attenuate, finely erose-fimbriate, whitish, or the larger ones with a narrow bright brown median stripe; rachis also yellowish scabrid pilose below, very rough; pinnee opposite or nearly so, 5.5 to 7 cm. apart, very narrowly oblong-lanceolate, 30 to 33 em. long, 4 to 5 cm. broad, spreading, sessile, not narrowed at the base, here slightly apart from the rachis at the upper side, subcordate below and either contigu- ous to the rachis or partially overlying it, deeply and almost equally pinnatifid through- out, the costal wing 4 to 5 mm. broad upon each side, scarcely broader below the crenate and ultimately serrulate acuminate apex; costa stout, yellowish brown, sul- cate and glabrous above, below faintly canaliculate, 1 to 2 mm. broad, copiously clothed with spaced spreading scales similar to those of the rachis or relatively broader, falcate, subflexuous, whitish and without a median stripe; segments close, about 28 pairs, nearly oblong, slightly broadest at the base, 9 to 11.5 mm. broad at the very narrow and sharply acute sinuses, subfalcate, subentire in the lower half, faintly serrulate above the middle, sharply so at the rounded apex; costules elevated, glab- rous above, copiously paleaceous below, the scales minute, whitish, bullate, ending in a capillary point; veins 13 to 16 pairs, elevated, mostly once-forked, the basal branch of each basal vein joined by a short transverse veinlet to the opposed basal branch of the basal vein of the adjacent segment, a narrowly elongate basal areole thus formed along the costule, all four branches excurrent to the sinus; second pair and succeeding veins mostly once-forked at or near the base (sometimes twice-forked), the branches rather close, oblique, soriferous beyond their middle; sori relatively small but contig- uous, extending in a close subflexuous supramedial line from the apex downward to 46 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, a point on the costal wing usually about one-third the distance from costa to sinus, there meeting obtusely the sori of the adjacent segment; indusium membranous, yellowish brown, semicircular or dimidiate, subentire to deeply and irregularly cre- nate, repand, shallow; receptacle globose-capitate, setiferous; leaf tissue herbaceous, dark green and lustrous above, yellowish and much paler below. TypPE LocaLiry: Martinique, Sieber (Fl. Mart. 375), DisrriputTion: Mountains of Martinique and Guadeloupe, at from 500 to 1,480 meters elevation. Intusrrations: Plumier, Traité Foug. pl. 26; Petiver, Pter. Amer. pl. 2. f. 10; Presl, Tent. Pterid. pl. 1. f. 18; Hook. Sp. Fil. 1: pl. 14 B (as H. grandifolia). Plumier’s plate 26 above cited was mentioned by Willdenow as illustrating his grandifolia; but, as shown under the discussion of that species at page 42 H. grandifolia is preferably to be interpreted by means of the Willdenow typp speci- men, which is of the species known hitherto as /. insignis or H. imrayana. On account of the white scales of the under surface, larger specimens of this species might be confounded with true H. grandifolia, except for the simpler venation and the obtuse or, at least, nonacuminate segments. Its white scales will at once dis- tinguish it from #. obtusa which invariably has brown scales, and these of a very different character. EXPLANATION OF PLATE 26.—From a photograph of a portion of Sieber 375, Martinique, the type collec- tion. Natural size. (Specimen in herbarium of the Missouri Botanical Garden.) 21. Hemitelia obtusa Kaulf. Enum. Fil. 252. 1824. Cnemidaria obtusa Presl, Tent. Pterid. 57. 1836. Hemistegia obtusa Presl, Abh. Bohm. Ges. Wiss. V. 5: 355. 1848. Hemitelia bullata Christ, Bot. Jahrb. Engler 24: 81. 1897. Rhizome of mature individuals unknown, probably ascending or forming a short upright caudex; fronds apparently 2 to 2.5 meters long, the stipe stout, clothed at the base with bright yellowish brown linear-lanceolate long-acuminate scales, sharply short-aculeate, hght castaneous to yellowish brown, muricate upwards, deciduously paleaceous; lamina broadly oblong or ovate-oblong, apparently 1 to 1.5 meters long, 40 to 60 cm. broad, abruptly acuminate, bipinnatifid, the rachis stout, deeply sulcate and glabrescent above, below lightly and obtusely sulcate, smoothish, laxly yellowish- pubescent with long flexuous flaccid hairs; pinnzee numerous, close, the lower ones deflexed; middle pinnze divergent, opposite or subopposite about 4 to 6 em, apart, oblong-ligulate, 20 to 35 cm. long, 3 to 4 (rarely 4.5) em. broad, not or scarcely narrower at the base, close to the rachis at the upper side, subcordate below and com- monly overlying the rachis, sessile, straight or lightly falcate, gradually acuminate in the apical third or fourth, (the extreme apex sharply serrate,) pinnatifid two- thirds to three-fourths the distance to the costa, the costal wing about 3 to 5 mm. wide on each side or slightly broader toward the apex; coste glabrous and minutely but sharply sulcate above, below yellowish to brownish, stout, conspicuously ele- vated, at first freely clothed (at least toward the base) with shining subbullate ovate or oblong-ovate ferruginous or yellowish brown flaccid scales with lighter fibrillose margins, some of the scales persistent at the sides; segments about 24 to 30 pairs, oblong, lightly falcate, 9 to 12 mm, broad at the base, mostly close, the sinuses narrow and sharply acute, or broader in drying and acutish, the margins usually revolute, lightly crenate-serrate above the sinus, toward the apex crenate-dentate, the teeth not prolonged; costules sharply elevated, glabrous above, below obscurely setulose- glandular and bearing numerous small bullate brownish fibrillose scales; veins 13 to 16 pairs, elevated, glabrous above, minutely glandular-setulose below, once-forked, the basal one having its basal branch connected with a similar branch from the adja- cent segment by a short transverse veinlet, the costal areole very narrowly elongate, ull four branches extending to the sinus or to a point immediately above; second MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 47 pair and succeeding veins once-forked close to the sinus, the branches divergent, distant, soriferous above their middle; sori rather small, adjacent (sometimes con- tiguous with age), forming a distinctly supramedial line extending one-half to two- thirds the distance to the apex (rarely to the apex) and downward toa point on the costal wing about equidistant between the costa and the sinus, indusium grayish brown, semicircular or dimidiate, subentire to crenately lobed, shallow; receptacle globose, squamulose-setiferous; leaf tissue rigidly herbaceous, dark green above, lighter below, lustrous on both surfaces. Type Locatiry: ‘Habitat in Antilles.” DisrriputTion: Apparently confined to Grenada and St. Vincent. The present species has been misidentified with nearly as much frequency as has H. grandifolia (and usually under that name), although Kaulfuss’s description is definite enough. In addition to Kaulfuss’s original plant Presl cites specimens col- lected by Guilding in St. Vincent, in which on the basis of material at hand he is probably correct. Grenada specimens collected by Eggers (no. 6035) were first determined by Christ as 1. grandifolia, but were subsequently made the type of his new species H. bullata. They are perfectly typical examples of H. obtusa, as here understood. In leaf outline and venation //. obtusa resembles H. kohautiana rather closely; but it is strikingly different in its fewer and distinctly brownish scales of the under surface and in its almost nonpaleaceous, smoothish, and yellowish pubescent rachises.! The following specimens have been examined: Sr. Vincent: H. H. & G. W. Smith 854, 1715; Eggers 6731. GRENADA: Eggers 6035; Sherring; Broadway. DOUBTFUL SPECIES. 1. Hemrrevta cructata Desv. Mém. Soe. Linn. Paris 6: 320, 1827. The original description is as follows: “Pinnis oppositis, sessilibus, lineari-lanceolatis subacuminatis, patentibus, pro- funde crenatis: laciniis subimbricatis incurvis, obtusisque apici latere acutiusculis obscure denticulatis: costis rachique nudis; caudice arborescente? “Habitat in America calidiori. Media inter H. grandifoliam et speciosam.”’ An excellent photograph of the type specimen, which is preserved in the Muséum d’Histoire Naturelle at the Jardin des Plantes, Paris, indicates a species at least very closely allied to that here recognized under the name H. spectabilis Kunze. In the shape of its pinnee and in venation the specimen appears to agree with the Trinidad plants here cited under H. spectabilis and to differ only in its subimbricate segments. Without a direct comparison of the specimen itself with J. spectabilis it appears inadvisable to substitute the earlier name; but it is more than likely that the two relate to phases of the same species. At any rate its relationship is clearly with H. spectabilis, as here understood. 9. HEMISTEGIA ELEGANTISSIMA I’ée, Mém.Foug, 8: 110. 1857. Founded upon a Mexican specimen collected by Linden, without number; not itlentified by the writer. The description, brief though it is, does not accord with any of the species here recognized. 1The plant figured by Hooker, Sp. Fil. 1: pl. 14 A. as H. obtusa is neither H. obtusa nor any species closely related to it. The illustration agrees exactly with the Trinidad material here taken up under the name J/7. spectabilis, but it does not show any secondary areoles (i. e., along the costules of the segments). This, however, is not a constant feature of that species and is only observed here and there. See under H. spectabilis. 48 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 3. Hemiterta Munira (Willd.) Hook.; Kuhn, Linnaea 36: 162. 1869. Cyathea munita Willd.; Kaulf. Enum. Fil. 260. 1824, nomen nudum. Hematelia munita Hook. Sp. Fil. 1: 32. 1844, nomen nudum. Hemistegia munita Presl, Abh. Bohm. Ges. Wiss. V. 5: 355. 1848. The present species, based upon Willdenow’s no. 20168, was not described until 1869, by Kuhn, although mentioned by several earlier writers under different names as indicated above. Through the kindness of Dr. I. Urban the writer has examined a small portion of Willdenow’s specimen, which has as its type locality simply “America.”? It conforms well with Kuhn’s description and represents either a valid species or a nearly sterile state of H. obtusa Kaulf.; probably the latter, although it is not matched exactly by other specimens. The scales are brownish, as in that species, but very few and minute; also, the segments are more deeply serrate and the sinuses much narrower than usual. It is, at least, closely allied to IJ. obtusa, and the type should be compared closely with undoubted specimens of that species. 4, HEMITELIA spECTABILIS Kunze, Linnaea 21: 233. 1848. Hemistegia spectabilis Fée, Gen. Fil. 351. 1850-52. Actinophlebia obtusa Presl, Abh. Bohm. Ges. Wiss. V. 5: 356. 1s48, not Hemitelia obtusa Kaulf. 1824. Type LocaLiry: Surinam, Kappler 1771. Disrripution: French and Dutch Guiana, Trinidad, and Venezuela, according to Kunze. Intusrration: Hook. Sp. Fil. 1: pl. 14.4 (as I. obtusa). So far as can be ascertained this species, which is here identified with some uncer- tainty, is wholly South American, the Trinidad flora being considered as belonging to that continent. Kunze included in his concept of the species plants from several widely separated regions: Material collected in French Guiana by Leprieur and at first’ referred doubtfully to H. obtusa; better specimens received later from Dutch Guiana and Trinidad; and, finally, material collected near Caracas by Linden and by Karsten. Mettenius ? subsequently redescribed the species in full, citing it only from Dutch Guiana. Principally on the basis of the latter diagnosis, which does not conflict with that of Kunze, the name is here applied with reservation to the following material in the National Herbarium: TRINIDAD: Without locality, Fendler 25 (4 sheets). Near Valencia, Nov., 1883, Eggers 1423. Without locality, ex herb. Bot. Gard. Trinidad, 195. Venecuena: El Valle, Island of Margarita, August 16, 1901, Miller & Johnston 164, San Juan Mountain, Island of Margarita, altitude 500 meters, July 16, 1903, Johnston 191 (in part). Whether or not these specimens actually pertain to I, spectabilis, they at least represent a species distinct from any of the North American flora, and one to which no other name appears to apply. They accord well with the descriptions by Kunze and Mettenius already mentioned. Hooker's figure cited above also agrees perfectly, It was probably drawn from Lockhart’s Trinidad material mentioned by him.’ (Seg under /T. obtusa.) According to Christensen’s Index Filicum []emistegia spectabilis Fée is an equivalent of Hemitelia subincisa. Fée published no description of it but cited the following synonymy: “ Hemithelia obtusa, Hook., fragm., *non Klfss.; Hemithelia [Cnemidaria | subincisa, Kze.”? Thus, although he apparently did not intend it to be a transfer of Hemitelia spectabilis Kunze to the genus Hemistegia and, in fact, makes no reference to Kunze’s species, it is nevertheless on the basis of Hooker’s illustration a probable synonym of Hemitelia spectabtlis. 1 Bot. Zeit. 2: 297. 1844. ? Fil. Hort. Lips. 111. 1854. * This is substantiated by a recent letter from the Director of the Royal Gardens, Kew. MAXON—STUDIES OF TROPICAL AMERICAN FERNS, 49 Presl’s Actinophlebia obtusa was founded wholly upon Hooker’s plate 14A which was clearly a misidentification of H. obtusa Kaulf., 1824. Hooker’s figure, moreover, which is here regarded tentatively as typifying H. spectabilis, is itself one of the several original elements of Kunze’s H. subincisa; but it can scarcely apply to that species as typified here on the basis of Péppig’s Peruvian plant. Kunze, in publishing H. spectabilis, writes: ‘Est quasi media inter meam H. subincisam et H. obtusam,?’— with mention of his earlier (1844) reference of Hooker’s plate I4A. 5. HEMITELIA SUBINCISA Kunze, Bot, Zeit, 2: 296, 1844, Cnemidaria speciosa Presl, Tent. Pterid. 57. pl. 1. /. 16. 17. 1836, not Cyathea speciosa H. & B.; Willd. 1810. Hemistegia speciosa Fée, Gen. Fil. 351. 1850-52. Type Locauity: Peru, Péppig. Distripution: Venezuela to northern Brazil and Peru (according to Underwood MS.). ILLustTRATION: Presl, loc. cit. pl. 7. f. 16. 17. The ground taken by Kunze, in his long review ! of Hooker’s treatment of Hemitelia in the Species Filicum, for establishing Hemitelia subincisa is essentially that taken by Presl, both authors agreeing that Kaulfuss erred in his identification of Cyathea speciosa H. & B. (See under /. speciosa, page 30). But just what herbarium material Kaulfuss had in hand in transferring Cyathea speciosa H. & B. to Hemitelia can not be stated, nor can the source of Presl’s information. In the Presl herbarium at Prague, however, is a specimen of ‘‘ Cnemidaria speciosa” collected in Peru by Péppig. This very likely not only formed the basis of Presl’s figures 16 and 17, but is probably a plant of the same Péppig number which gave Kunze many of the data for his new J/. subincisa. A fragment in the Underwood Fern Herbarium does not agree with Presl’s figure 16, but accords perfectly with figure 17. Possibly figures 16 and 17 belong to different individuals or different species. In any case, it seems desirable for present purposes to typify the species on figure 17, which apparently represents Péppig’s Peruvian plant as found in Presl’s own herbarium. The Brazilian plant, as repre- sented by Martius’s figure,” seems to be the same. Hooker’s plate 14A, published as ‘‘//. obtusa” is also cited by Kunze for his 7H. subincisa; but it is drawn from a Trinidad specimen and represents a species distinct from H. subincisa, as the latter is typified in this paper. It is here regarded as repre- senting H. spectabilis. Hemitelia subincisa has been credited to Guatemala and other parts of tropical North America, but so far as can be ascertained it is altogether South American. FURTHER NOTES ON THE WEST INDIAN SPECIES OF POLYSTICHUM. Since the writer’s revision of the West Indian species of Poly- stichum in the last paper of this series (1909) considerable addi- tional material has been received, some of it showing extension of ranges, as here recorded. The single new species to be described is rather closely related to P. dissimulans, yet offers differences which seem to be specific. Polystichum ambiguum Maxon, sp. nov. PLATE 27. Fronds 4 or 5, laxly arching, 60 to 74 cm. long, long-stipitate (the stipe as long as the lamina or longer), Rhizome decumbent, about 5 cm. long, 1.5 cm. in diameter, woody, bearing numerous coarse freely branched roots, and sparingly clothed with thin dark to light brown oblong-lanceolate scales about 1 cm. long; stipes stoutish, 31 to 38 1 Bot. Zeit. 2: 294-299, 1844, 2 Icon. Pl. Crypt. pl. 48. f. 2. 50 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. cm. long, stramineous, subquadrangular, sulcate, bearing a few large flaccid yellowish brown scales near the base, elsewhere nearly glabrous, or with a few linear or fibrillose tortuous scales above, these extending along the rachis but readily deciduous; lamina narrowly ovate, deeply bipinnatifid nearly throughout, 29 to 36 cm. long, 10 to 17 cm. broad, comprising 15 to 17 pairs of spreading mostly falcate pinnze; middle and lower pinne 6 to 8.5 em, long, 1.5 to 2em. broad at the middle, subpinnate, strongly ineequi- lateral at the base, the superior basal segment much the largest, free, rhombic-ovate from an unequal narrowly cuneate base, the inferior one minute, free, the next 5 to 8 pairs narrowly to broadly rhombic-ovate, 8 to 12 mm. long, very oblique (the distal margin lying close to the narrowly alate secondary rachis), the apical segments much narrower, fully adnate and strongly decurrent, finally evident only as deep serra- tions at the acuminate apex; all the segments sharply acuminate but scarcely spines- cent; apical pinne 1 to 2 cm. long, inserted 1 to 1.5 cm. apart; leaf tissue membrano- chartaceous, the under surfaces very sparingly and minutely fibrillose-paleaceous, mainly along the veins; rachis stout, very narrowly alate in the upper part, terminat- ing in a large viviparous bud 1 to 2 cm. above the apical pinnae; venation concealed, mostly flabellate, repeatedly dichotomous, the segments without definite midribs; sori large, irregularly biserial, the larger segments with 2 to 5 pairs. Type in the U. S. National Herbarium, nos, 520201 and 520202, collected upon rocky slopes bordering humid forests in the immediate vicinity of Holly Mount, Mount Diabolo, Jamaica, altitude about 750 meters, May 25 to 27, 1904, by William R. Maxon (no. 2283). The relationship of the present species is clearly with P. dissimulans. From this it differs mainly in the paler scales of the rhizome, in its different leaf shape, fewer pinne and minute inferior basal pinnules, in its flagelliform (and not foliose) apex, and in its sharply acuminate, rather than rigidly spinescent, segments, - P. dissim- ulans is the most rigidly coriaceous of all the West Indian allies of P. triangulum while P. ambiqguum has singularly flaccid fronds for a member of thisgroup. P. hetero- lepis, though superficially resembling P. ambiguum to a certain extent, differs in nearly all essential details. EXPLANATION OF PLATE 27, A middle section of the type specimen. Polystichum plaschnickianum (Kunze) Moore. This species, known hitherto only from Jamaica, has been collected recently in Santo Domingo by von Tiirckheim (no. 3038). The specimens, which are typical, are from the vicinity of Constanza, altitude 1,350 meters. Polystichum polystichiforme (I*ée) Maxon. Known previously only from Cuba and Jamaica. Collected recently in Porto Rico by Brother Hioram, his specimens (no. 245) from Mount Torresilla, July, 1911. Polystichum triangulum (L.) Fée. This species, mentioned in the last paper as inhabiting only Santo Domingo, Cuba, and Jamaica, occurs also in Guatemala. The record rests upon plants collected by von Tiirckheim near Coban, Alta Verapaz, altitude about 1,350 meters, on rocks, and distributed by Captain Smith as no. 851. Specimens from the vicinity of Constanza, Santo Domingo, altitude 1,190 meters, February, 1910, von Tiirckheim 2933, distributed as Polystichum triangulum vay. iliei- folium Fée are not Polystichum ilicifolium Fée. They are, rather, referable to P. trianqulum, but represent an unusually spiny form ef the species. Polystichum wrightii (Baker) C. Chr. in herb. Polypodium wrightit Baker in Hook. & Baker, Syn. Fil. 304. 1867. Dryopteris sauvallei C. Chr. Ind. Fil, 291, 1905. Polystichum longipes Maxon, Contr. Nat. Herb. 18: 34. pl. 6. 1909. The above synonymy relates wholly to a Cuban species known only upon Wright's no. 3924. Baker, strangely enough, first described the species under his section Contr. Nat. Herb., Vol. 16. PLATE 27. POLYSTICHUM AMBIQGUUM MAXON, Contr. Nat. Herb., Vol. 16. PLATE 28. PTEROPSIS UNDERWOODIANA MAXON, MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 51 “Euphegopteris” of Polypodium. Christensen, supposing it from this incomplete description and wrong generic position to be a Dryopteris, transferred it to that genus as D. sauvallei, the new species name being necessary in Dryopteris because of an earlier D. wrightti of Kuntze (1891). Subsequently he examined specimens at Stock- holm, and noting their true affinity, called them Polystichum wrightti, aname which must take precedence over P. longipes, published in ignorance of Baker’s diagnosis. THE AMERICAN SPECIES OF PTEROPSIS. In publishing the new genus Ananthacorus several years ago! the writer indicated briefly the grounds upon which the generic name Pteropsis (Desvaux, 1827) should be taken up to replace Drymo- glossum (Presl, 1836). Two species of this genus have been known previously from America, one from Ecuador, the other from Mar- tinique. A third, from Costa Rica, was detected by Dr. L. M. Under- wood in 1904, but apparently never named or described by him. The three species are: 1. Pteropsis wiesbaurii (Sodiro) Maxon. Drymoglossum wiesbaurii Sodiro, Vasc. Crypt. Quit. 419. 1893. Known only from Ecuador, the type being from tree trunks along the Rio Chimbo, altitude 300 to 500 meters. 2. Pteropsis martinicensis (Christ) Maxon. Drymoglossum martinicense Christ, Bot. Jahrb. Engler 24: 137. 1897. Apparently confined to Martinique; the original specimens from trees and rocks near Vauclin, Duss 250b. 3. Pteropsis underwoodiana Maxon, sp. nov. PLATE 28. Rhizome sarmentose, very slender, about 1 mm. in diameter, sparingly paleaceous, the scales subappressed, grayish, about 1 to 1.5 mm. long, linear-oblong to oblong- ovate, acute, minutely erose, membranous, with thin cell walls. Sterile fronds sessile, lanceolate, acuminate, 8 to 13 cm. long, 2.2 to3.8 cm. broad above the broadly cuneate, usually equilateral base, strongly costate, the stramineous midvein and slender irreg- ularly reticulate veins elevated and evident upon both sides throughout; leaf tissue firmly membrano-chartaceous, inconspicuously whitish-glandular above, bearing upon both surfaces numerous but distant minute punctiform ovate to suborbicular scales, these centrally peltate, with narrowly erose-fimbriate whitish margins. Fertile fronds 9 to 11 cm, long, short-stipitate, the stipe (1 to 1.5 cm. long) stout, appressed- paleaceous, the lamina linear, narrowly long-cuneate, 8 to 9cm. long, 4 to 5 mm. broad; sporangia arising in a dense line about midway between the costa and margin, spreading to the margin and at maturity almost completely obscuring the costa below the short linear-cuspidate apex. Type in the U.S. National Herbarium, no. 827444, collected near Suerre, Llanuras de Santa Clara, Costa Rica, altitude 300 meters, February, 1896, by John Donnell Smith, no. 6941; distributed as “ Acrostichum amygdalifolium Mett.’’ There are speci- mens of the same number in the Underwood Fern Herbarium, New York Botanical Garden. The American species may be distinguished by the following key: Lamina of sterile fronds obovate, coriaceous, about 3 cm. long, 1 cm. broad, densely covered with minute appressed stel- late scales; fertile fronds plicate...............2........ 2. P. martinicensis. 1 Contr, Nat. Herb. 10: 486. 1908. 32870°—-12 —-3 52 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Lamina of sterile fronds lanceolate to oval-lanceolate, membra- nous or membrano-chartaceous, 2 to 4 times as large, one or both sides bearing minute scattered suborbicular to ovate scales; fertile fronds not plicate. Fertile fronds 3 to 4mm. broad, long-stipitate, the stipe 3 to 5 cm. long; sterile fronds oval-lanceolate, 6 to 9 cm. long; 2 to3cm. broad .........22...-.-2220------ 1. P. wiesbaurit. Fertile fronds 4 to 5 mm. broad, short-stipitate, the stipe | to 1.5 em, long; sterile fronds lanceolate, 8 to 13 cm. long; 2.2 to 3.8 em, broad..............--..-.-.-... 3. P. underwoodiana. TWO UNUSUAL FORMS OF DICRANOPTERIS. Under the head of ‘‘ Doubtful or Extralimital Species” the writer, in treating recently the North American species of Dicranopteris,' made mention of two peculiar forms as follows: Mertensia gleichenioides Liebm. Vidensk. Selsk. Skr. V. 1: 296. 1849. (Gleichenia lieéhbmanni Moore, Index Fil. 379. 1862.) A remarkable form, accurately described by Liebmann from specimens collected by him near Cuaba, Vera Cruz, Mexico, and apparently not since collected. In general appearance the specimens differ widely from the usual type of Dicranopteris in the direction of Gleichenia, but not in venation and other characters. In minute characters they appear to represent a species not otherwise knowr, but in gross morphology the plant is almost certainly atypical and possibly indicates a reversion toward a general ancestral form, Plants similar in form, but very different in vestiture, have been collected in Jamaica by Professor Under- wood and the writer; these were growing with D. bifida, and from their minute characters must be reckoned a form of that species. The present note is for the purpose of directing further attention to the peculiar morphology of these plants. Plate 29 represents at about two-fifths natural size the Jamaican plants referred to (Mazon 936). They were collected by the writer in company with Prof. L. M. Underwood upon the dryish, brushy slopes of an abandoned coffee plantation, altitude about 750 meters, above Tweedside, which is near Mount Moses, in the Blue Mountains. Surrounding them upon all sides was a typical growth of the common tropical American species called Dicranopteris fulva (Desv.) by Doctor Underwood ? and recently redescribed * by the writer under an earlier species name as D. bifida (Willd.) Maxon. The unusual interest attaching to these specimens was perhaps not fully appreciated at the time; at any rate nothing was noted beyond the fact that they covered an area of only a few square feet in the midst of normal D. bifida. Plants of similar form were not encountered elsewhere in Jamaica, although D. bifida is the commonest species of the genus at mid-elevations. In minute characters the specimens are evidently identical with ordinary D. bifida, which in its several forms is one of the most readily recognized species of the genus in North America, its 'N. Amer. Flora 16!: 53-63. 1909. 3N. Amer. Flora 16!: 60. 1909. ? Bull. Torrey Club 34: 255. 1907. Contr. Nat. Herb., Vol, 16. PLATE 29. DICRANOPTERIS BIFIDA (WILLD.) MAXON. (A monstrous form.) MAXON—STUDIES OF TROPICAL AMERICAN FERNS, 53 segments being covered beneath by a rusty, usually dense, entan- gled tomentum, which, however, with age frequently becomes bleached and matted, or nearly disappears. The peculiarity of the Jamaican form here figured consists mainly in its having its simple pinne very narrowly linear (3 to 5 mm. broad) and merely subentire to broadly crenate, instead of pectinate, as in the normal form. The veins, which are short, are once-forked, the branches either simple or one or both of them again forked. The sori are dorsal upon the veinlets, as in all species of Dicranopteris, instead of terminal, as in Gleichenia. A dull brownish rusty tomen- tum closely invests the under surface of the pinne throughout from the narrowly revolute margins to the rachis. The primary internodes which subtend the pinne are precisely like those of the normal fronds of the species (Mazon 937), which were collected at the same time and place, except that they have in several instances a crenate or crenately lobed wing on the lower side, as well as upon the upper. The presence or absence of reduced segments bordering the primary and secondary internodes of the lateral branches of D. bifida is, how- ever, an unusually variable feature. Of almost identical form, but of very different covering below, is the plant described by Liebmann as Mertensia gleichenioides, Mertensia being used by him as the equivalent of Dicranopteris, and the species name gleichenioides in allusion to the general resemblance which the plant offers to true Old World Gleichenia. The lightly but broadly crenate pinne are a little more slender than in the monstrous form from Jamaica, not exceeding 4 mm. in width, and the margins are for the most part strongly revolute. The rachises of the pinne are clothed below with delicately lacerate pale ferruginous scales, and the veinlets of the under surface are covered with minutely dissected subpersistent scales, their capillary divisions exceedingly delicate and in mass strongly suggesting a tomentum. The veins are mostly once-forked, each of the branches again once or twice forked, the veinlets thus subfasciculate, a group to each broad crenation. That Liebmann’s plants represent an abnormal state of some Mexican species, as the monstrous state here figured does of D. bifida, is entirely probable, but the writer is unable to identify it with any previously described. For a very careful sketch of the two specimens constituting Liebmann’s type, and for a pair of pinne of the type, forwarded from the Botanisk Museum, Copenhagen, to the U. S. National Museum, the writer is indebted to the courtesy of Mr. Carl Christensen. As to the significance of the peculiar form shown by these two collections of different species, speculation is perhaps idle; yet it seems not unlikely that they represent a reversion to a more general- ized ancestral type, rather than a chance variation. And the suppo- sition that they may, perhaps, be an atavistic expression is doubtless 54 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. strengthened by their separate occurrence in regions far apart. Minor variations in form, size, and vestiture of the segments, in the highly complex scheme of branching, and in the production of seg- ments upon the internodes of the lateral branches—all of these being features observed commonly in the field—seem to indicate that the species of this genus are in a more or less unstable state; and this renders the more significant so pronounced a departure from the normal form. Members of this family are said to be difficult of culti- vation, which, together with the usual incompleteness of herbarium material, may account for the scant attention they have received. Nevertheless, the group is one of the greatest interest and one which, in the writer’s opinion, would well repay critical investigation of the living plants, more especially a comparative study of those species showing radically diverse methods of branching. Following such a study it is not unlikely that Dicranopteris, instead of being again merged in Gleichenia, will itself be subdivided into several genera. THE AMERICAN SPECIES OF CIBOTIUM. In working over Cibotium for the forthcoming second part of volume 16 (Pteridophyta) of the North American Flora, it has been found that the American species have, if anything, been more fre- quently misidentified than the Old World material of the genus. The main reason for this appears to be that no one has given the American species careful attention. Thus, Kuhn properly distinguished two new species, C. guatemalense and C. wendlandi, in 1869, but failed to point out that Hooker’s illustration of “‘ Cibotium schiedei,” plate 30A of the Species Filicum, really pertained to one of these, C. wendlandi. The confusion attending the illustration of C. regale, itself a valid species, is explained farther on, under that species. As a matter of fact the material available at any one herbarium is probably scant; and the distinctive points of difference among the several species, if evident to individual students, have at least never been pointed out. Full descriptions of the four species already mentioned will appear shortly. In the meantime illustrations of these, with the notes here given by way of comparison, may be helpful. The Cibotiwm horridum of Liebmann is found not to belong to this genus or its tribe. The genus Cibotium of Kaulfuss is often credited to his Enumeratio Filicum (1824). It was, however, published! four years earlier in a little pharmaceutical journal, the only file of which known to the writer is that in the Library of the Surgeon General, in the Army Medical Museum, Washington, D. C. The original description is as follows: “Die Fruchthaufen sind in gewélbten, an einem Punkte auf der Unterseite des Laubes befestigten, lederartigen Schleierchen eingeschlossen, die sich von oben 1 Kaulfuss in Berl. Jahrb. Pharm. 21: 53. 1820. PLATE 30. Contr. Nat. Herb., Vol. 16, CIBOTIUM SCHIEDE! SCHLECHT. & CHAM. MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 55 mit einem bleibenden, nach der Rippe der Lappchen zu sich neigenden Deckel éffnen. Nur Eine, wahrscheinlich baumférmige, Art, Cibotium Chamissoi, von der Siidsee.”’ In America the genus is apparently restricted to the northern con- tinent. The species may be distinguished by means of the following key: KEY TO THE SPECIES. Coste glabrous or readily glabrescent; leaf tissue conspicuously ceraceo-pruinose below. Sori mostly distant or subdistant, usually extending beyond the margin in the plane of the segment................ 1. C. schiedet. Sori contiguous to imbricate, appearing dorsal, i. e., not ex- tending beyond the margin. Lamina deeply tripinnatifid; veins 7 to 9 pairs to the segment; sori nearly parallel to the costule.....-. 2. C. regale. Lamina tripinnate or subtripinnate; veins 8 to 15 pairs to the segment; sori obviously oblique to the cos- tule, closer..........-------2-02 eee ee ee eeeeeee- 3. ©. quatemalense. Coste thickly invested with long persistent antrorse hairs; leaf tissue not obviously ceraceo-pruinose below......-....---- 4. C. wendlandi. 1. Cibotium schiedei Schlecht. & Cham. Linnaea 5: 616. 1830. PLATE 30. Dicksonia schiedei Baker in Hook. & Baker, Syn. Fil. 50. 1866. Tyre Locauity: Hacienda de la Laguna, Mexico, Schiede 801. DistRiBUTION: Humid mountains of Oaxaca and Vera Cruz, at 600 to 1,200 meters elevation. InLustraTIons: Presl, Tent. Pterid. pl. 11. f. 9. Cibotium schiedei, which was the first species of this genus to be described from North America, is apparently confined to Mexico.! It has long been in cultivation and need not be confused with any other. The fertile segments, especially the larger ones, are manifestly dentate-crenate and contain only 6 to 8 pairs of veins, of which the fertile ones are almost invariably simple and the sterile ones usually once forked at a slight angle. The most distinctive feature lies in the distant or subdistant produced sori (2 to 7 pairs) which commonly extend outward in the plane of the lamina, but in a few instances (in extreme age) are bent back under the segment. This character, while helpful in distinguishing the species, is possibly not fundamental and is, no doubt, correlated directly with the thin, chartaceous leaf tissue. In all the other American species the sori are closer and have the appearance of being erect and dorsal, since the segments, on account of their coriaceous or at least herbaceous texture, have the margins strongly revolute at maturity or in drying. The following specimens have been examined: Mexico: Hacienda de la Laguna, Schiede 801. San Francisco, Mirador, Lieb- mann. Orizaba, Miiller. Cordoba, Kerber 90a; Fink 13; Bourgeau 2378. Barranca de Tenampa, Zacuapan, Vera Cruz, September, 1906, C. A. Purpus 1976. Zacuapan, Vera Cruz, November, 1908, C. A. Purpus 1976a. (Also numerous specimens from the Botanical Gardens of Kew, Berlin, and Leipsic.) EXPLANATION OF PLATE 30.—a, Pinnule from a cultivated plant, er hort. Lips.; 6, portion of pinnule of type collection (Schiede 801), in Underwood Fern Herbarium; c, pinnule from very old specimen of Purpus 1976; d-g, Purpus 1976a, d representing one of the larger inferior pinnules, e and f two superior pinnules from near the base of the pinna (at a point opposite d), g a nearly sterile pinnule from near the middle of a pinna, Only a shows the upper surface, All are at natural size. ! Skinner’s plant from Guatemala, referred here by Hooker, is C. wendlandi. See under that species, p. 57. 56 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, 2. Cibotium regale Versch. & Lem. Ill. Hort. 15: under pl. 548, 1868. Puarer 31. Dicksonia regalis Baker in Hook. & Baker, Syn. Fil. ed. 2. 461. 1874. Type Locauity: Described from cultivated specimens collected in Mexico by Ghiesbreght, altitude 1,500 to 1,800 meters. DistriBuTION: Chiapas, Mexico. ItLustration: Ill. Hort. pl. 548, in part (colored figure; also fig. 52). The original description of this species states no exact locality for the specimens collected by Ghiesbreght, but gives the altitude as from 5,000 to 6,000 feet. Two large pinne of Ghiesbreght’s collection in the D. C. Eaton herbarium, however, have the following data: “‘No. 351. Terre temperée. Etat de Chiapas. Fougére arbore- scente. Tronc de 3 44 pieds de haut. Frondes de 5 4 6 metres de longeur. Croix au bord des ruisseaux, Juillet et Aout.’’; all but the number being apparently in Ghiesbreght’s hand. Three pinnules of this, which is doubtless the type collection, are shown in plate 31 and will give an excellent idea of the species.' Cibotium regale has nearly the form of C. wendlandi, but differs conspicuously in its acuminate (not aristate) segments, in having the costee and costules very sparingly silky-pubescent below (the hairs long and very readily deciduous), and in having the under surface of the segments conspicuously pruinose (not greenish), In cutting it is quite dissimilar to C. guatemalense. It is more nearly related, perhaps, to C. schiedei, but differs in many respects, notably in its more numerous pinnules, greater size, different texture, revolute margins, and the different direction and position of the sori, and in having the veins of the sterile segments mostly two or three times forked, not mostly once forked. EXPLANATION OF PLATE 31.—Three pinnules of Ghiesbreght 351, from Chiapas. Natural size. 3. Cibotium guatemalense Reichenb.; Kuhn, Linnaea 86: 152. 1869. Pratre32, f, g. Dicksonia guatemalensis Baker in Hook. & Baker, Syn. Fil. ed. 2. 461. 1874. Type Locauity: Guatemala, Wendland. DistrisutTion: Apparently confined to the humid mountain region of eastern Guatemala, altitude 1,500 meters or less. Cibotium guatemalense is readily distinguished from its allies by the key characters noted above. It has very large, nearly or quite tripinnate fronds, with both pinne and pinnules very much larger than those of C. scheidei, from which it differs con- spicuously also in its imbricate, differently placed sori and its more numerous veins (8 to 15 pairs). The oblique position and crowding of the sori separate it immediately from C. regale. The species has been reported from Costa Rica upon the basis of two different collections. The first of these (Warscewicz 43), according to a pinnule in the Under- wood Fern Herbarium, indicates an undescribed species very closely allied to C. wend- An illustration is practically essential to a clear understanding of this species, owing to the very faulty original figures and the confusion existing between Lemaire’s legends for the detailed drawings and his ‘explanation of the analytical figures.”’ Figure 5 of plate 548 is presumably “‘fig. 1” of the ‘‘explanation,”’ and probably is intended to represent Ghiesbreght’s plant, as is stated. Figures 1, 2, 3, and 4 of plate 548 apparently represent the two ‘‘pennules” and ‘‘a, b, & c” mentioned in the “‘explanation,’”? which are said to be redrawn from Hooker’s plate 30A of the Species Filicum, and are obviously copied from that. A comparison of plate 548 with Hooker’s plate 30A shows that the disagreement of the numbers which really appear on plate 548, with the letters and numbers of Lemaire’s ‘‘explanation,” is due to an error of the artist, who copied not only Hooker’s detailed illustrations but also the identical numbers which accompanied them in the original! The drawing labeled 5 in plate 548 is, as mentioned above, doubtless meant for C. regale; but its resemblance to Hooker’s figure (fig. 1 of pl. 548) is too close to offer any distinctive features whatever. It should be borne in mind also that the plant of Hooker’s plate 30A is not C. schiedei, but C. wendlandi. (See under the latter species, p. 57.) Contr. Nat. Herb., Vol. 16. PLATE 31. ae : fe he, “J \> © . f " SG, ii ae CIBOTIUM REGALE VERSCH. & LEM. PLATE 32, 16. Contr. Nat. Herb., Vol. , hie we » Ve S™ Ke CIBOTIUM WENDLANDI METT. AND C. GUATEMALENSE REICHENB. MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 57 landi; it is, at any rate, not C. guatemalense. The second is of a plant collected by Wercklé and listed! by Christ as Cibotiwm guatemalense. Several specimens of this in the U. 8. National Herbarium, received from Doctor Christ and so labeled in his hand, are Dicksonia navarrensis Christ. So far as can be ascertained, C. guatemalense is strictly confined to Guatemala. The following specimens have been examined: GUATEMALA: Santa Cruz, Alta Verapaz, altitude 1,380 meters, John Donnell Smith 1505 (5 sheets), Rio Frio, near Santa Cruz, Alta Verapaz, von Ttirckheim II. 2113 (4 sheets). Without locality, Salvin & Godman (ex herb. Kew). 4. Cibotium wendlandi Mett.; Kuhn, Linnaea 36: 151. 1869. PLATE 32, a-e. Dicksonia wendlandi Baker, in "Hook. & Baker, Syn. Fil. ed. 2. 460. 1874. Type LocaLity: Guatemala, Wendland. DistriBUTION: Probably confined to Guatemala. InLustRATION: Hook. Sp. Fil. 1: pl. 30A (as C. schiedet).? Cibotium wendlandi differs widely from the other American species of this genus in the persistent, appressed-hairy covering of the costae and costules, and in its greenish under surfaces. A close examination, nevertheless, shows the underside of the leaf tissue to be very minutely papillate, though not pruinose. Christ has reported® this species from Chiapas, Mexico, upon specimens collected by Munch, and has subsequently 4 given the further data: “El Zontehuitz, altitude 2.858 meters (Munch 104).’’ At the latter reference he lists also a “‘very similar” Costa Rican plant (Tonduz 10697) which, though referred to C. wendlandi, is said to “approach” C. guatemalense. This number (10697) is again mentioned’ by him the following year, under C. wendlandi, as distinct from C. guatemalense; and again in 1907° (here incorrectly as no. 10797) as C. wendlandi, ‘‘distinguished from C. guate- malense by its hairy surface and greater dimensions.’’ The plant in question (no. 10697) is represented by two excellent specimens in the U. 8. National Herbarium, one of these received from Doctor Christ. It is not a Cibotium, but an apparently undescribed species of Dicksonia. Under the circumstances the Mexican record (Munch 104) must be considered as exceedingly doubtful, not only for the species C. wendlandi but for the genus Cibotium as well; the plant is probably a Dicksonia. Cibotium wendlandi is apparently a species of the semiarid regions of the Pacific coast, which probably accounts for the denser vestiture of the leaf surfaces. The following specimens have been examined: GUATEMALA: Without locality, Skinner 22 (ex herb. Kew); Wendland (ex herb. Berol.). Guatemala, Depart. Guatemala, altitude 1,350 meters, John Donnell Smith 2423 (4 sheets).” EXPLANATION OF PLATE 32.—a-e, Cibotium wendlandi; a, fragment (from Kew) of Skinner’s Guatemalan Specimen; b, fragment (from Berlin) of Wendland’s Guatemalan specimen (type collection); c-e, Guate- mala, John Donnell Smith 2423, c being a small subapical pinnule, d and e the fifth pair from the base (inferior and superior, respectively); f, g, Cibotium guatemalense; {, Guatemala, John Donnell Smith 1506, a middle inferior pinnule; g, Guatemala, von Tiirckheim II. 2113, one of the larger superior pinnules. All at natural size. 1 Bull. Herb. Boiss. II. 5: 251. 1905. 2 This illustration has such slight resemblance to C. schiedei that the writer wrote to Kew, asking the source of the material figured. The reply (May 2, 1910) states that the figure was probably drawn from Skinner’s no. 22. A pinnule of this, for- warded at the same time, is C. wendlandi. 3 Bull. Herb. Boiss. IT. 5: 251. 1905. * Loc. cit. I]. 5: 734. 1905. 5 Loc. cit. II. 6: 189. 1906. 6 Loc. cit. IT. 7: 273. 1907. 7 Captain Smith states in a recent letter that the exact locality is a barranca or deep ravine bounding a little hacienda called Aceituno, not far outside Guatemala City. 58 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. EXCLUDED SPECIES. CrBoTIuM HORRIDUM Liebm. Vid. Selsk. Skr. V. 1: 279. 1849. This species is represented in the U. S. National Herbarium by three pinnules and part of a secondary rachis of Liebmann’s original material, forwarded from Copen- hagen, and said to have been collected by Liebmann in June, 1842, “in sylva montana prope Teotalcingo, Chinantla, Dept. Oaxaca, alt. 4-5000’.” The most cursory exami- nation of the scales of the rachis shows them to be relatively short, rigid, spinescent- ciliate, 5 to 10 cells broad, and thus of a totally different type from those of Cibotium, which are capillary, flaccid, and only a single cell broad. Their agreement with scales of the secondary rachis of Cyathea princeps (Linden) E. Meyer is so close, and the secondary rachis and the pinnules agree so closely in every particular, that Lieb- mann’s species must undoubtedly be written as a synonym of Cyathea prince ps as recently redescribed by the writer.!| Liebmann’s specimens, according to his description and the fragments received, are sterile, as might be expected in young plants of a Cyathea attaining the great size of C. princeps and in plants of such small size as that attributed by Liebmann to Cibotium horridum. TWO NEW SPECIES OF NOTHOLAENA. In a recent examination of the Mexican material of Notholaena in the U. S. National Herbarium the following two new species were detected: Notholaena leonina Maxon, sp. nov. Fronds 4 to 8 in number, 4 to 11 em. high, fasciculate. Rhizome relatively stout, creeping or ascending, 1 to 1.5 cm. long (incomplete), about 4 mm. in diameter, very thickly clothed with densely imbricate, lance-acuminate to linear-subulate, dark brown scales (2.5 to 3.5 mm. long) with yellowish brown borders, the margins dis- tantly and delicately glandular-papillate (especially toward the apex), otherwise subentire; stipes 2.5 to 7 cm. long, very slender, blackish, terete, slightly scaly toward the base, the scales broader than those of the rhizome, ovate, long-acuminate, yellowish brown, concolorous or with darker tips; lamina deltoid to deltoid-oblong, acute or slightly produced, 2 to 4.5 cm. long, 1.7 to 3.5 em. broad, bipinnate or rarely tripinnatifid in the basal part, simple above, the apex simply pinnatisect, the rachis similar to the stipe but lightly sulcate ventrally; major pinne 3 to 5 pairs (those above simple, linear-oblong to oblong), subopposite, inserted 7 to 12 mm, apart, the basal ones deltoid, with 2 to 4 pairs of spreading pinnules (or segments) below the usually trilobate apex, these elongate-oblong, simple and at least partly adnate, or the basal ones sessile and with 1 or 2 pairs of minute segments or lobes; pinnules or segments in general 1.5 to 2 mm. broad, flat, rigidly herbaceous, grayish green, together with the rachises densely ceraceo-papillate throughout; coste of the seg- ments wholly concealed above, evident below orly toward the base; margins closely revolute about one-third the distance to the costa (or less at maturity), unchanged, only partially concealing the dark brown sporangia. Type in the U. 8. National Herbarium, no. 834605, collected near Monterey, State of Nuevo Leon, Mexico, February 17 to 26, 1880, by Dr. Edward Palmer (no. 1381) the specimens received from Capt. John Donnell Smith. Known to the writer only from the type number, which apparently was not gen- erally distributed to herbaria; at least it is wanting in the National Herbarium set received originally, and is not cited by Baker,? who does, however, list numbers 1382 and 1383 of the same collection as N. pringlei. Eaton® listed no, 1381 as N. candida Hook. d 1N. Amer. Flora 16': 78, 1909. 3 Proc. Amer. Acad. 18: 185. 1883. ? Annals of Botany 5: 482. 1891. MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 59 The systematic position of N. leonina is next to N. pringlei Davenp. and N, bryo- poda Maxon. From the former, which it resembles superficially, it differs in the delicately glandular-papillate (not coarsely and irregularly denticulate) tips of the rhizome scales, in its very slender terete blackish (not stout sulcate yellowish brown) stipes, in the lesser degree of subdivision of the lamina, and in its flat seg- ments and less coriaceous leaf tissue. N. pringlei is usually much larger, and has shorter segments; and even in its younger states commonly has the lamina tripinnate in the lower part. The difference in color and thickness of the stipes is marked. Notholaena bryopoda differs from both species in the complete absence of any cera- ceous covering to the lamina, and in numerous other particulars, Notholaena rosei Maxon, sp. nov. Fronds 6 to 8, rigidly erect, 30 to 48 em. high, loosely fasciculate. Rhizome suberect, stout, multicipital, 1.5 to 3 cm. in diameter, 4 to 5 cm. long, densely paleaceous, freely radicose, and bearing numerous imbricate stipe-bases of old fronds; scales of the rhi- zome closely imbricate, lance-acicular, 2 to 2.5 mm, long, very rigid, the center yellow- ish brown, the apex and sides lustrous blackish brown and strongly thickened, with a delicate lax yellowish white araneose-ciliate outer border; stipes stout, 1.5 to 2 mm, in diameter, 8 to 13 cm. long, brownish stramineous from a dull castaneous brownish flexuous base, subterete, lightly canaliculate along the anterior face, sparingly and deciduously scaly, densely puberulous with glandular-capitate hairs; lamina 20 to 35 em. long, 7 to 10 cm. broad near the middle, lanceolate, acuminate, deeply tripinnati- fid nearly throughout, the rachis similar to the stipe but with a conspicuous narrow ventral furrow; pinne few, distant, of a deltoid type, mostly inequilateral, ascending, stalked (1 to 3 mm.), the basal pair subopposite and reduced (about 2 cm, long), the second pair 5 to 6 cm. distant; middle pinne subopposite, 3 to 5 cm. apart, deltoid- ovate, 4 to 5 cm. long, 1.8 to 2.3 cm. broad, comprising about 8 to 10 pairs of distant subsessile to adnate pinnules below the pinnately lobed acuminate apex; larger pinnules oblong-acuminate to narrowly deltoid-oblong and subcaudate, 10 to 15 mm. long, 3 to5 mm. broad, pinnatifid (often nearly to the midvein), the lobes about 4 to 6 pairs; upper pinne simply pinnate, 1 to 2 cm. apart; leaf tissue herbaceous, bright yellowish green and glabrous above, below densely white-ceraceous, partially conceal- ing the costze of the pinnules; margins yellowish, a very narrow border slightly meta- morphosed and partly covering the sporangia before maturity, early thrust back; sporangia dark brown, relatively thick. Type in the U.S. National Herbarium, no. 451280, collected on a rocky hillside near Chapala, State of Jalisco, Mexico, October 5, 1903, by J. N. Rose and J. H. Painter (no. 7665). The only other specimens seen are very immature plants of no. 701 of Dr. Edward Palmer’s 1886 collection, from the same place. These are mentioned by Davenport! as identical with Pringle’s ‘2830,’ which is there described as Notholaena lemmoni var. straminea Davenport, var. nov. Mr. Pringle’s type specimens (‘‘2830”), which are said to have come from rocky hills near Guadalajara, Jalisco, December, 1888, have not been seen by the writer. They are not at the Gray Herbarium, nor at the Daven- port Herbarium in Boston; moreover the number 2830 was given by Mr. Pringle (perhaps subsequently) to a flowering plant (Asclepias mexicana) which was actually distributed in his regular series. Judging from the brief description, Palmer 701 is the same as Pringle ‘‘2830,’’ the type of var. straminea; but this is by no means certain. Possibly no. “2830” may be contained in the Pringle Herbarium. This is now the property of the University of Vermont and available for study only to investigators who will consult it in Burlington, Vt. Under the circumstances it seems advisable not to make up the varietal name for this species. Notholaena rosei is so dissimilar in every respect from N. lemmoni that one wonders upon what common ground the comparison of relationship could have been instituted 1 Garden and Forest 4: 519. 1891. 60 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. by Mr. Davenport. In general leaf shape only it resembles somewhat N. rigida, but the lamina is subtripinnate, instead of bipinnate. In most other characters, and particularly in its puberulous, dull stramineous (not lustrous dark castaneous) stipes and rachis, it is widely different. The rhizome and rhizome scales of the two species are wholly unlike. NV. rosei appears to have no very near relatives, MISCELLANEOUS NOTES AND CHANGES OF NAME. Adiantopsis rupicola Maxon, Contr. Nat. Herb, 10: 485. 1908. Two recent collections of this very distinct Cuban species may be reported, as follows: Bafios San Vicente, province of Pinar del Rio, September 12-16, 1910, Britton, Britton & Gager 7497; trail from Buenaventura to San Juan de Guacamalla, on rocky hillside, December 16, 1910, P. Wilson 9349. Cheilanthes aurea Baker in Hook. & Baker, Syn. Fil. ed. 2. 476. 1874. Tyre Locaity: Matagua Valley, Guatemala, Salvin & Godman. DistripuTion: Apparently known only from Guatemala. Innustration: Hook. Icon. Pl. pl. 1637. The original specimens have not been seen by the writer, but the following plant accords perfectly with the diagnosis and later plate: Along the Rio Carracal, near Quetzaltepeque, Guatemala, altitude 1,000 meters (rare), 10, 7, 1882, Lehmann 1689. This number seems to have been omitted by Hier- onymus from his report on this collection. The specimen at hand, received from Captain Smith, was determined by Christ as Cheilanthes microphylla Sw. Cheiroglossa palmata (L.) Presl. In addition to the illustrations for this species cited in the North American Flora! may be mentioned plate 4 of Hooker’s Icones Plantarum (1837), which shows a small plant said to have come from the island of Bourbon, off the East African coast. In America the plant ranges from southern Florida throughout the West Indies to Mexico and Brazil. Its unusual distribution, which is common to a few other species of pteridophytes, is discussed by Doctor Christ in his recent work, Die Geographie der Farne, 1910. Dryopteris germaniana (Fée) (©. Chr. Christensen, in his second paper on the ferns of the group of Dryopteris opposita,? has extended the range of D. germaniana to Cuba, the species having been known previously only from Guadeloupe. Agreeing exactly with this Cuban specimen ( Mazon 4059) is a plant received recently from Porto Rico, collected at Barranquitas, July, 1911, by Brother Hioram (no. 270), Goniophlebium eatoni (Baker) Maxon. PLATE 33. Polypodium ghiesbreghtti D.C. Eaton, Proc. Amer. Acad. 8: 618. 1873, not Linden, 1867, Polypodium eatoni Baker, in Hook. & Baker, Syn. Fil, ed. 2. 511. 1874. Gontophlebium pringlei Maxon, Proc. U. 8. Nat. Mus. 2°7: 953. pl. 48. 1904. An examination of the type specimens of P. ghiesbreghtii D. C. Eaton in the Eaton Herbarium, collected in Chiapas by Ghiesbreght (no. 273), shows them to be identical with G. pringlei, described several years ago upon specimens from the vicinity of Jalapa, Vera Cruz, Pringle 11855. In publishing G. pringlei the writer commented upon the peculiar character of one or more pairs of the basal pinnz, as follows: “Two additional sheets in the United States National Herbarium differ in having the second pair of pinne like the first—that is, free and cordate-clasping at the base 116: 13. 1909, * Smiths. Misc. Coll. 52: 365-396. 1909, Contr. Nat. Herb., Vol. 16. PLATE 33. ’ os ce SBrenee | yeeseoee eee” oSaeereeeg} ee ahd on ae pee GONIOPHLEBIUM EATONI (BAKER) MAXON. Contr. Nat. Herb., Vol. 16 PLATE 34. GONIOPHLEBIUM RHACHIPTERYGIUM (LIEBM.) MOORE. MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 61 below the midvein and fully adnate above, the adnate upper portion of the first pair overlapping the clasping base of the second, there being only a very slight connecting foliar wing along the rachis, In one of these (no. 460766), moreover, this condition occurs in the case of the third pair of pinne also, and the foliar wing becomes well developed only between the third and fourth pairs of pinne and between succeeding pinne. In all four sheets examined the superabundance of leafy tissue, which must have given a remarkable fluted appearance to the living fronds, is very noticeable.” One of the Ghiesbreght plants shows this peculiar condition so admirably that a \botographic illustration is given herewith (pl. 33). The species has been reported! by Doctor Christ also from Chiapas, the specimen collected by G. Munch. Goniophlebium rhachipterygium (Liebm.) Moore, Index Fil. 396, 1862. Pare 34. Polypodium rhachipterygium Liebm. Vid. Selsk. Skr. V. 1: 191. 1849. Polypodium stenoloma D. ©. Eaton, Proc. Amer. Acad. 8: 618. 1873. Polypodium donnell-smithii Christ, Bull. Herb. Boiss, IT. 6: 291. 1906. Doctor Christ has recently ? pointed out the identity of P. donnell-smithii, described from Guatemalan specimens collected by von Tiirckheim (Donnell Smith 8823), with the earlier P. stenoloma D. C. Eaton, founded upon Ghiesbreght 386, from Chiapas, Mexico. A still earlier publication of the species is that by Liebmann, mentioned above, whose specimens came from Teotalcingo, Chinantla, Oaxaca, Mexico, June, 1842. A very complete sketch of these, together with a fragment, both sent by Mr. Christensen at the writer’s request, leaves no doubt as to their identity with the plants described subsequently by Professor Eaton and by Doctor Christ. The illustration presented herewith (pl. 34) is of the type specimens of P. stenoloma (herb. D. C. Eaton), showing the plants at about one-half natural size. Lycopodium dichaeoides Maxon, Proc. Biol. Soc. Washington 18: 231. 1905. Besides the two collections from Alta Verapaz originally mentioned, this species is known to the writer only upon Captain Smith’s no. 958, collected at Pansamala, Alta Verapaz, Guatemala, at about 1,200 meters elevation, by von Tiirckheim, in July, 1886. The species, though obviously related to L. aqualoupianum, is well marked by its short strobiles and its short, rigid, achene-like sporophyls. Notholaena rigida Davenp. The original specimens are from limestone ledges, Sierra de la Silla, Nuevo Leon, Mexico, May 31, 1889, Pringle 2599. The only other plants of this species seen by the writer are those collected near Victoria, Tamaulipas, Mexico, altitude 320 meters, February to April, 1907, by Dr. Edw. Palmer (no. 142). The species is strongly char- acterized by its lustrous, dark castaneous stipes and rachis and by its branched, lig- neous rhizome, the latter closely invested with very rigid, opaque, blackish scales. Pellaea notabilis Maxon, Contr. Nat. Herb. 10: 500. 1908. A second record for this species rests upon two dwarfed fronds in the herbarium of the New York Botanical Garden: These were collected near San José, Tamaulipas, Mexico, altitude 600 to 1,100 meters, by Prof. James F. Kemp, of Columbia University, in 1902. They measure only 8 and 12 cm. high, respectively, and are thus much smaller than the original specimens, which are also from Tamaulipas. Polypodium duale Maxon, nom. nov. Acrostichum serrulatum Swartz, Prodr. Veg. Ind. Occ. 128. 1788. Polypodium serrulatum Mett. Fil. Hort. Bot. Lips. 30. 1856, not Swartz, 1801. The above change of name for the diminutive tropical fern described from Jamaica by Swartz in 1788 as Acrostichum serrulatum is made necessary by the use of the same 1 Bull. Herb. Boiss. I]. '7: 413. 1907. 2 Bull. Soc. Bot. Genéve 1: 220. 1909. 62 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. species name under Polypodium by Swartz (in 1801) for a plant now known as Dryop- teris serrulata. The synonymy will be indicated in full in some notes on Polypodium duale and related species, to be published in the next paper of this series. Two other species names! subsequently applied are not valid under Polypodium. Polypodium heterotrichum Baker. A rare species hitherto recorded, apparently, only from Jamaica, where it occurs on trees in the higher forested slopes of the Blue Mountains, at 2,100 meters. Speci- mens have been received recently from Barranca Trinidad, State of Hidalgo, Mexico, May 10, 1904 (Pringle 13494). Polypodium jenmani Underw. nom. nov, “Polypodium lasiolepis” Jenman, Bull. Bot. Dept. Jamaica II. 4: 118. 1897, not Mett. 1869. The Jamaican plant described by Jenman as ‘‘ Polypodium lasiolepis Mett.’’ is not very closely related to P. lasiolepis of the Lesser Antilles, which is itself (together with P. grenadense Jenman) apparently a synonym of P, tenuiculum Fée. Jenman’s description is very complete. He compares the species with P. pendulum Swartz, pointing out, in part, that it differs from that species in its stronger rhizome, close and more decurrently adnate, ciliate segments, slightly hairy surface, terminal superficial sori, absence of glands, and different apex and base. He adds: ‘‘It is also ereet in growth, and the veins and midrib of the pinne are not raised on the upper side.”’ As a matter of fact, the relationship to P. pendulum is not very close, but that is a species very much misunderstood. The following specimens of P. jenmani are in the National Herbarium: Jamaica: At base of tree, slopes above Tweedside, altitude about 900 meters, Maxon 961. On trees, near Mabess River, altitude 900 meters, Maron 1535 (= Underwood 2606). Polypodium leptostomum Fée, Mém. Foug 7: 58, pl. 2. f. 2. 1857. ?Polypodium productum Maxon, Contr. Nat. Herb, 18: 11. 1909, not Christ, 1907. In assigning the name P, productum to a supposed new Guatemalan species in 1909 the writer overlo»ked the earlier use of this name for a Philippine species by Christ.? A new name for the Guatemalan plant does not appear necessary at present, however, inasmuch as it resembles very closely P. leptostomum Fée, founded upon plants from Orizaba (W. Schaffner 210) and may prove identical with it. Specimens of the type collection bave rot been seen by the writer. ' Micropteris orientalis Desv. Mém. Soc. Linn. Paris 6: 217. 1827, not P. orientale Gmel, 1791. Xiphopteris extensa Fée, Mém. Foug. 11: 14. 1866, not P. extensum Forst. 1786. * Philippine Journ, Sci. C. Bot. 2: 178. 1907. THE NORTH AMERICAN SPECIES OF NYMPHAEA. By Gerrit S. Miter, Jr., and Paut C, STANDLEY. INTRODUCTION. NOTE BY J. N. ROSE. There are some groups of plants whose taxonomy can be fairly well understood almost solely from ordinary herbarium material. There are others in which it is impossible to understand the true relationships from such material alone, and some even in which herbarium specimens are almost. useless. Among the last are the Cactaceae, many of the tropical Euphorbias, the Crassulaceae, and in general all the succulent plants. To be properly understood such plants should be seen growing, or should be studied from material preserved in alcohol or formalin, since in dry specimens most of the important characters are distorted beyond all possibility of recogni- tion. Among plants of this kind the Nymphaeaceae must be included. Most of our knowledge of the genus Nymphaea, judging from the literature of the North American species at least, has been derived from the study of dried herbarium material. Asa result, different authors have arrived at very different conclusions. Impor- tant peculiarities of habit and structure not discernible in such specimens have been overlooked, and our knowledge of the genus has not kept pace with the advances made in some other groups. About ten years ago Mr. Gerrit 5. Miller, jr., became interested in the genus Nymphaea from field observations of the plants occurring in central New York and in the vicinity of Washington, currently regarded as belonging to one species, Study of fresh and formalin- preserved material showed that there were important and easily recognizable differences between the northern form and that found farther south, differences in habit, color, and structure, scarcely to be detected in dried specimens. He published a brief paper in 1902? stating these differences and recognizing the northern plant as a distinct species. About the same time he began to bring together fresh material from all parts of North America to facilitate complete knowledge of the genus as represented there. This attempt was remarkably successful. The plants wherever they occur are well known as ‘‘yellow pond lilies,” hence, not only professional botanists, 1 Proc. Biol. Soc. Washington 15: 11-13. . 63 64 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. but those with little or no botanical knowledge, were able to secure the specimens desired. Numerous persons connected with the various branches of scientific work in Washington as well as botanists and others throughout the country assisted, so that finally a very large suite of specimens was brought together. This material con- sists at present of about two hundred and twenty jars of specimens preserved in formalin, besides an exhaustive series of dried specimens, the whole being deposited in the National Herbarium. Canada, Alaska, and practically every State in the Union are represented, except the Territories of New Mexico and Arizona, where the genus is not known to occur. This material was forwarded to Washington, sometimes in formalin, but more often fresh and merely wrapped in oiled paper. As soon as specimens were received careful notes were made on the color of the various parts. Tracings of some of the leaves were drawn, seeds were selected from the ripe fruits and dried, and in general a careful study was made of any peculiarities exhibited by the fresh plants. All but one of the species finally recognized were thus studied in the fresh condition. By the year 1904 enough material had been brought together to permit the determination of all but two of the nineteen forms of Nymphaea now known to occur in America. At this time pressure of other duties, necessitating several prolonged absences from Washington, made it impossible for Mr. Miller to continue the task of completing and publishing a revision of the genus, though he had written out the key to the species essentially in the form in which it now stands. At Mr. Miller’s request, Mr. J. H. Painter prepared to take up the work of elaborating the paper and procuring such additional material as was needed, but in December, 1908, this was interrupted by Painter’s tragic death. During the past year it has been resumed and brought to completion by Mr. Paul C. Standley, under whose authorship jointly with Mr. Miller’s the paper is now printed. AUTHORS’ INTRODUCTION. MATERIAL STUDIED. As stated by Doctor Rose, dry herbarium material is of little value for the study of this genus. When the species are once known, how- ever, most of them can be recognized in the dried state. Conse- quently, we have examined all the herbarium material available, with the result that we are able to map the areas of distribution of the better known species with some detail. In addition to showing the distribution of the various species upon outline maps, we have listed, with name of collector and date and locality of collection, all the specimens examined. These include all the material in the herbaria MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 65 of the University of Pennsylvania, the Philadelphia Academy of Natural Science, the New York Botanical Garden and Columbia College, the Gray Herbarium, and the Missouri Botanical Garden (this containing the types of the Engelmann Herbarium), together with certain material from the University of California, that in the private herbarium of Dr. E. L. Greene, and, finally, all that in the National Herbarium. To those who have charge of these collections we are deeply indebted, either for the loan of the specimens or for facilities for their study. In addition, we are under obligations to the dozens of people all over the country who have responded to requests for living material. Their names, which are so numerous as to prohibit their full enumeration here, will be found in the lists of specimens examined. Without their assistance the work could not have been completed. HISTORY OF NAMES. NAMES APPLIED TO GENERA. Blephara J. E. Smith, Mem. & Corr, 1: 577. 1882. Before publishing the name Nuphar, Smith sent a diagnosis of the genus under the name Blephara to the Bishop of Carlisle in a letter dated November 17, 1808. This letter is printed in the ‘‘Memoir and Correspondence,” thus giving the name Blephara a definite status. The type is given as Nymphaea lutea. Nuphar J. E. Smith in Sibth. Fl. Graec. Prodr. 1: 361. 1808 or 1809 (title page dated 1806, but part containing this name not printed before December, 1808). Type, by monotypy, Nymphaea lutea L. A synonym of Nymphaea L., as restricted by Salisbury in 1806, and of Nymphozanthus L. C. Richard, May, 1808. Nymphaea L. Sp. Pl. 510. 1753. Type, N. lutea L. The genus originally contained the species lutea (misprinted lusea), alba, lotos, and nelumbo, representing the modern genera Nymphaea (lutea), Castalia Salisb., 1806 (alba and lotos), and Nelumbo Adans., 1763 (nelumbo). No type was designated, nor was any clue furnished to the author's intention. Under the American code of Botanical Nomenclature the types of the genera of Linneus’ Species Plantarum are to be determined through the citations given in his Genera Plantarum (1754). On page 227 of this work are cited under Nymphaea Tournefort’s plates 137 and 138. These represent the white-flowered Castalia alba and the yellow-flowered Nymphaea lutea. Since Linneus gives no specific indication that either was the type of his genus, the first of the Linnean species common to the two works is to be regarded as the type. This is Nymphaea lutea L. 66 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. In 1806 Salisbury,’ the first reviser of the genus Nymphaea restricted the name to a group containing lutea only of the original Linnean species, thus fixing the type, according to the practice of zoologists, Two years laterJ. E.Smith,? deliberately setting aside Salisbury’s per- fectly valid action, and overlooking the Nymphozanthus of Richard, May, 1808, re-restricted Nymphaea to the group represented by alba and dotos, and applied a new name, Nuphar, to lutea and its allies. Although not justified by any rules of nomenclature, Smith’s treat- ment of the subject received the general sanction of botanists for nearly eighty years. In 1887 and 1888, however, Greene * and Brit- ten * called attention to the errors and restored the correct names. Nymphona Bubani, Flora Pyrenaea 3: 259. 1901. A substitute for Nuphar proposed on account. of philological prejudices, the ancients having applied the wor¢ “nuphar” to the plant’s root. It is a synonym of Nymphaea. Nymphozanthus?® LL. ©. Richard, Démonstr. Bot. 68. 1808. A name based on the yellow-flowered species of the Linnean genus Nymphaea. As it was published in May, 1808, it antedates the Nuphar of Smith by at least seven months (see Britten, Journ. Bot, 26; 7. January, 1888). Ropalon Raf. New FI. N. Amer. 2: 17. 1836. A synonym of Nymphaea based on Nymphaea sagittata Pers. NAMES APPLIED TO SPECIES. In addition to the names lutea and pumila originally applied to European members of the genus and erroneously used for American species, thirty names, including the ten published here for the first time, have been based on American plants. They are as follows: Advena. Nymphaea advena Ait. Hort. Kew. 2: 226. 1789. This is the first name based on an American member of the genus. It was applied to a plant brought to the Kew Gardens in 1772 by William Young. As Young collected in South Carolina and in the neighborhood of Philadelphia, Pennsylvania,* there can be no ques- tion as to the applicability of the name to the common erect species characteristic of the Upper Austral zone. It has been more com- monly misapplied, however, to the floating-leaved Boreal plant. ' Ann. Bot, 2:71. ? See Memoir and Correspondence of Sir J. E. Smith. 1:575, 577-581. 1832. ® Bull. Torrey Club 14: 177-179. September, 1887; 257-258, December, 1887; ibid., 15: 84-85. March, 1888. * Journ. Bot. Brit. & For. 26: 6-11. January, 1888. * Spelled Nymphozanthus on p. 63 and in the index, Nymphosanthus on p. 68 where the genus is defined. The second form may be regarded as an obvious misprint. 6 See Britten, Journ. Bot. Brit. & For. 32: 332. N ovember, 1894, MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 67 Americana. Nuphar americana Provancher, FI. Canad. 1: 28. 1862. Although intended merely as a substitute for advena this name is available for the Boreal species now better known as variegata. The diagnosis clearly refers to the northern plant, and the type locality, Lake St. Jean—Georgie, Quebec, is far beyond the range of Nymphaea advena. Arifolia. Nymphaea arifolia Salisb. Ann. Bot. 2:71. 1806. Another substitute for advena. Bombycina. See page 102. Chartacea. See page 94. Erythraea. See page 91. Fletcheri. Nymphaea fletcheri Lawson, Proc. & Trans. Royal Soc, Canada 6: 119, 1888. Aname applied to Nymphaea rubrodisca. The author did not, how- ever, regard the plant as a distinct species, but as a hybrid between N. kalmiana and N. americana (‘‘advena’’). Fluviatilis. Nymphaea fluviatilis Harper, Bull. Torrey Club 83: 234, April, 1906. The only name based on a certain well-marked species, confined, so far as now known, to the State of Georgia. Type locality, near Groveland, Bryan County, Georgia. Fraterna. See page 82. Hastata. Nymphaea hastata Steud. Nom, Bot. ed. 2. 200. 1841. The entry is as follows: ‘‘[Nymphaea] hastata Michx. Nuphar sagittaefolia.”” Hybrida. Nuphar advena var. hybrida Peck, Ann. Rep. N. Y. Mus. Nat. Hist. 34: 53. 1881. This is the earliest name for the peculiar Boreal plant better known as rubrodiscum and minor. In accordance with the American Code of Botanical Nomenclature it is at present set aside in favor of its synonym rubrodiscum. Type locality, Forked Lake, Hamilton County, New York. Kalmiana. Nymphaea lutea B kalmiana Michx. Fl. Bor. Amer. 1: 311. 1803. The first name applied to the smallest of the three eastern Canadian species, but for the present displaced by the later microphylla. No locality further than ‘‘Canada” mentioned. Longifolia. Nymphaea longifolia Michx. Fl. Bor. Amer. 1: 312. 1803. A synonym of N. sagittifolia Walt. 1788, as suspected by Michaux himself. ‘‘Hab. in amnibus Carolinae sept. et merid.’’ Ludoviciana. See page 92. Macrophylla. Nymphaea macrophylla Small, Bull. Torrey Club 25: 465. Sep- tember, 1898. The local race of advena occurring in northern and central Florida. Type locality, Eustis, Lake County. Microphylla. Nymphaea microphylla Pers. Syn. Pl. 2: 63. 1807. Although published later than kalmiana Michx. 1803, this name, under the American Code, stands for the plant to which it was applied. 68 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Minor. Nuphar advena var.? minor Morong, Bot. Gaz. 11: 167. July, 1886. A name originally applied to depauperate specimens of Nymphaea americana from Smith’s Pond near Litchfield, Herkimer County, New York, but afterward transferred by Watson and Coulter to N. rubrodisca (Gray’s Manual, ed. 6. 56. 1889). Orbiculata. Nymphaea orbiculata Small, Bull. Torrey Club 33: 128. April 30, 1896. The only name for a strikingly characterized, large-leaved species known from southern Georgia only. Type locality, a small lake near Thomasville, Thomas County. Ovata. See page 97. Ozarkana. See page 91. Pictum. Nuphar polysepalum pictum Engelm. Trans. Acad. St. Louis 2: 285. 1868. A synonym of polysepalum. Polysepalum. Nuphar polysepalum Engelm. Trans. Acad. St. Louis 2: 282. 1868. The first name based on the large-fruited western member of the genus. After mentioning specimens from various localities Engel- mann says: ‘‘* * * And finally, Dr. Parry gathered ample ma- terial and full notes, which have been largely used in the following description, in Osborn’s Lake in the same region [near Longs Peak, Colorado] at an altitude of 8,800 feet, where it grows with Menyanthes trifoium * * *.” As the description is almost wholly based on the Parry material, Osborn’s Lake must be accepted as the type locality of polysepalum. On page 285, however, the plant from the same lake receives a second name: ‘The flowers of Dr. Parry’s plant are more highly colored than that of Dr. Hayden’s [from a small lake between Henry’s Fork and Snake Fork of the Columbia River] and may preserve the name of var. pictum, which the discoverer has applied to the species.” The name pictum is thus an exact synonym of polysepalum. Puberula. See page 99. Rubrodiscum. Nuphar rubrodiscum Morong, Bot. Gaz. 11: 167. 1886. The first specific name given to the plant originally described aa a subspecies under the name hybrida. The author apparently regarded the plant, which he found on the Vermont shore of Lake Champlain, as a species in process of formation from a hybrid. ‘Intermediate between N. advena [=americana] and the following [microphylla], and produced from a hybrid between them. Still a hybrid in many localities.” , Sagittaefolia. Nuphar sagittacfolia Pursh, Fl. Amer. Sept. 2: 370. 1814. Substitute for sagittifolia Walt. 1788. Sagittata. Nymphaea sagittata Pers. Syn. Pl. 2: 63. 1807. This is merely another name for Nymphaea sagittifolia. Sagittifolia. Nymphaea sagittifolia Walt. Fl. Carol. 155. 1788. The first name based on the plant to which it is currently applied. No type locality is mentioned, but the preface to the Flora is sub- scribed: ‘‘Carolinae Meridionalis, ad Ripis Fluvii Santee.”’ MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 69 Tomentosum. Nuphar advena 8 tomentosum Torr. & Gr. Fl. N. Amer. 1: 58, 1838. This name was based on a specimen of Nymphaea advena collected by Thomas Nuttall in the vicinity of Philadelphia, Pennsylvania, and labeled by him Nuphar tomentosum. It is still extant in the herbarium of the Academy of Natural Sciences of Philadelphia. The supposed pubescence of the petioles and lower surface of the leaves is merely a dense growth of alge. Ulvacea. See page 97. Variegata. Nuphar advena var. variegatum Engelm, in A. Gray, Man. ed. 5, 57, 1867. Under this name, based apparently on specimens from Michigan, Nymphaea americana was first, though imperfectly, distinguished from N. advena. Later it was used in binomial form for the same plant by G.S. Miller.’ SYSTEMATIC TREATMENT. NYMPHAEA L. 1753. Nymphaea L. Sp. Pl. 510. Type, Nymphaea lutea L. (Salisbury, Ann. Bot. 2:71. 1806.) 1808. Nymphozanthus L. C. Richard, Démonstr. Bot. 63-68. Type, Nymphaea lutea L. (Misprinted Nymphosanthusy the correct spelling on p. 63 and in index.) May, 1808. 1808 or 1809. Nuphar J. E. Smith in Sibth. Fl. Graec. Prodr. 1: 361. December, 1808, or early in 1809. Type, Nymphaea lutea L. 1832. Blephara J. E. Smith, Mem. & Corr. 1:576. (Substitute for Nuphar suggested but not adopted.) 1836. Ropalon Raf. New Fl. N. Amer. 2:17. Type, Nymphaea sagittata Pers. 1887. Nymphaea Greene, Bull. Torrey Club 14: 177-179. September, 1887; 257- 258. December, 1887. 1901. Nymphona Bubani, Fl. Pyr. 3: 259. (Substitute for Nuphar.) Type species, Nymphaea lutea L. DistrRiBUTION: I’resh-water ponds and sluggish streams (rarely in damp ground away from water) of the Holarctic Region from northern Europe and Asia to the Mediterranean and Japan and from Alaska and northern Canada to Cuba and north- eastern Mexico. DescripTion: Perennial aquatics with stout, creeping rootstocks; leaf blades entire, with a deep sinus at the base, narrowly lanceolate to orbicular, floating or emersed, on slender or stout, cylindric or flattened petioles; submersed leaves often present, these thin and delicate; sepals 5 to 12, orbicular to oblong or spatulate, con- cave, greenish, tinged with yellow or red; petals numerous, linear to oblong, thick, stamen-like or scale-like, inserted with the very numerous stamens on the receptacle under the ovary; stigma disk-like, with few to many rays; fruit ovoid to columnar, usually ripening above the water; seeds mostly ovoid, yellow or brown, smooth and shining, numerous. The Old World forms of Nymphaea are so imperfectly known that it is impossible to make any satisfactory estimate of the number of species in the genus. The Old World species are recognized by Engler and Prantl as about five. Half a dozen 1 Proc. Biol. Soc. Washington 15:18. February 18, 1902. 70 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. forms have until recently been supposed to represent the specific differentiation of the American members of the group. Their number is now increased to nineteen. KEY TO THE AMERICAN SPECIES. Width of leaf blade less than half the length; sinus less than one-fourth the length of the blade. Sinus usually closed; length of blade 2.5 times the breadth or less; fruit about 2 cm. long; seeds 3 mm. long; stigma rays elliptical; leaves acutish.........2.. 222.0022 2 eee eee eee ll. N. ulvacea. Sinus usually open; length of blade over 3 times the breadth; fruit about 3 cm. long; seeds 4 mm. long; stigma rays linear; leavesobtuse....... 10. . sagittifolia. Width of leaf more than one-half its length; sinus one- half to one-fourth the length of the blade. Calyx of more than 6, usually 9, parts............... 17. N. polysepala. Calyx of usually 6 parts. ’ Petioles conspicuously flattened; leaves floating. Anthers shorter than the filaments. Stigma rays less than 10; flowers less than 20 mm. in diameter; sepals NAITOW. . 2.2... eee eee eee eee eee 1. N. microphylla. Stigma rays more than 10; flowers about 30 mm. in diameter; sepals broad. 22... 2..2.-22.-2222-025- 2. N. rubrodisca. Anthers at least equaling the filaments. Capsules about 40 mm. high and 30 mm. in diameter; flowers about 45 mm. in diameter; leaves broadly rounded. .........2.-.----2----- 3. N. americana. Capsules about 25 mm. high and 18 mm. in diameter; flowers 22 to 26 mm. in diameter; leaves acutish......22.2...0.2.2222.-- 4. N. fraterna. Petioles subterete; leaves erect or floating. Lower surface of leaves glabrous. Leaves ofan orbicular type............ 8. N. fluviatilis. Leaves conspicuously longer than broad. Submersed leaves present; leaves and sepals very thin... .... 9. N. chartacea. Submersed leaves wanting; leaves and sepals thick. Sinus closed; stigma rays usually more or less con- fluent at the base, broad; disk usually oval... .....2....... 7. N. ludoviciana. Sinus usually open; stigma rays not confluent at the base, narrower; disk orbicular, MILLER AND STANDLEY—-NORTH AMERICAN NYMPHAEA. 71 Fruit not tinged with red when mature. Leaves 15 to 35 cm. long; rounded at the apex.... 5. N. advena. Leaves 25 to 50 cm. long,acutish... 5a. NW. advena macrophylla. Fruit conspicuously tinged with red at matu- rity. Leaves large, 25 to 40 cm. long, acut- ish; seeds very numerous...... 5b. NW. advena erythraea. Leaves smaller, 12 to 20 cm. long, broadly round- ed at the apex; seeds few (15 to 30).....------- 6. N. ozarkana. Lower surface of the leaves more or less pubescent. Leaf blades of an oval type; pubes- cence dense.................-6- 12. N. ovata. Leaf blades orbicular in outline, or nearly so. Lower surface of leaves with little pubescence; petioles nearly glabrous; leaves 16 to 21 cm. wide........-.......-.- 13. N. puberula. Lower surface of leaves and peti- oles covered with a dense, silvery, silky pubescence. Fruit small, 18 to 24 mm. in diameter; stigma rays about 12 (9 to 14); leaves 16 to 24 cm. wide...... 14. N. microcarpa. Fruit larger, 45 to 50 mm. in diameter; stigma rays much more numerous; leaves 30 to 50 cm. wide. Sinus closed; stamens in 9 or 10 rows; diam- eter of flowers about 50 mm.; seeds about 4 mm. long and 3 mm. thick........ 15. N. orbiculata. Sinus open; stamens in 6 or 7 rows; diame- ter of flowers about 70 mm.; seeds about 6 mm. long and 4.6 mm. thick........ 16. N. bombycina. 22075°—12——-2 72 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 1. Nymphaea microphylla Pers.! Nymphaea lutea L. Sp. Pl. 810. 1753, in part. Nymphaea lutea B kalmiana Michx. Fl. Bor. Amer. 1: 311. 1803. Nymphaea microphylla Pers. Syn. Pl. 63. 1807; Robins. & Fern. in Gray, Man. ed. 7. 391. 1908. Nymphaea kalmiana Sims, Curtis’s Bot. Mag. pl. 1243. 1809; Britton, Man. ed. 2. 407. 1905. Nuphar kalmiana Ait. Hort. Kew. ed. 2. 8: 295. 1811. Nuphar lutea kalmiana Torr. Fl. N. Y. 1: 40. 1843. TYPE LocaLity: Eastern Canada. DistripuTion.—Eastern Canada, south through New York to eastern Pennsyl- vania and northern New Jersey. DESCRIPTION. Floating leaves oval to broadly oblong or suborbicular, thin, broadly rounded at the apex, 35 to 100 mm. long and 35 to 75 mm. wide, the sinus 15 to 35 mm. deep, two- thirds as long as the midrib or more; sinus narrowly tri- angular, or closed, the lobes just meeting, these triangular, acutish; veins rather inconspicuous, glabrous above, gen- erally somewhat pubescent beneath, especially when young; submersed leaves orbicular, very thin, somewhat crispate, of about the same size as the floating ones, their sinuses usually more open; peduncles and petioles terete, very slender; flowers depressed-obovoid, about 18 mm. in diameter and 12 mm. high when normally open, the perianth when spread measuring about 30 mm.; sepals 5, obovate or elliptical, about 10 mm. long and 6 to 8 mm. wide, scarcely overlapping when expanded, obtuse at the apex, not contracted into a claw at the base; petals7 to 10, Fic. 2—Leaf outline of Nym- broadly spatulate, 6 mm. long and 3mm. wide; stamensin3 phaea microphylla. Scale 4. rows, about 15 to the row; filaments linear, cuneate, 5 mm. long, 1 mm. wide; anthers one-half as long as the filament; sepals gamboge yellow throughout, except those parts exposed in the bud, these apple green; petals orange, narrowly edged with gamboge yellow; anthers maize yellow, filaments gamboge; ovary pale apple green slightly variegated with yellow and much tinged with burnt carmine, especially on the ridges and at the base, the disk burnt carmine; fruit very small, ovoid, strongly constricted above into a neck 3 mm. long, the whole about 14 mm. high and 11 mm. in diameter, smooth except near the top, there faintly ribbed; stig- ge matic disk orbicular or oval, crenate, plane, 4 or 5 mm. in diam- eter; stigma rays 6 to 10, extending almost to the edge of the disk, Fic. 3.Stigmatic pattern of Nym- somewhat confluent in the center, with no trace of a median groove; phaea micro- color of fruit oil green, the disk bright red, often edged with yellow, phylla. Natural the body of the fruit usually with more or less of red; seeds oblong, “ie 3mm. long and 2 mm. in diameter, yellowish brown, shining. (PLareEs 35, A; 36, A. FIGURES 2, 3, 4, c.) * The use of the synonym microphylla in place of the original name kalmiana is in conformity with article 49 of the Vienna Code, which provides that ‘‘when a tribe becomes a family * * * a subdivision of a species becomes a species, or the reverse * * * the earliest name (or combination of names) received by the group in its new position must be regarded as valid, if it is in conformity with the rules, unless there exist any of the obstacles indicated in the articles of section 7.”’ Although ‘‘this Contr. Nat. Herb., Vol. 16. PLATE 35, A. FRUIT OF NYMPHAEA MICROPHYLLA PERS. B. FRuIt OF NYMPHAEA FRATERNA MILLER & STANDLEY. C. FRUIT OF NYMPHAEA ADVENA AIT. Contr. Nat. Herb., Vol. 16. PLATE 36. SEEDS OF SEVERAL SPECIES OF NYMPHAEA. MILLER AND STANDLEY—-NORTH AMERICAN NYMPHAEA. 73 EXPLANATION OF PLATE 35.—A. Fruit of Nymphaea microphylla. 3B. Fruit of Nymphaea fraterna, type collection. C. Fruit of Nymphaea advena. A1l natural size. EXPLANATION OF PLATE 36.—Seeds of (A) Nymphaea microphylla, (B) N. rubrodisca, (C) N. americana, (D) N. fraterna, (E) N. advena. (F) N. advena macrophylla, (G) N. ozarkana. All natural size. Specimens examined: In formalin— Mare: Birch Brook near north end of Cross Lake, Eagle Lake Chain, Aroostook County, 1903, W. C. Kendall. New Hampsuire: Cambridge River near Lake Umbagog, 1905, Kendall. New York: Piseco, 1901, W. L. Ralph; Gray, 1901, J. Perkins; Lake Champlain, 1901, Eggleston. Dried !— Canapba: Fredericton, New Brunswick, 1892, Fowler (Gray); Amqui Station, Metapedia River, 1882, Macown 97; Otter Slide Lake, Ontario, 1900, Macoun 21699; St. Johns County, New Brunswick, 1869, Fowler; near Quebec, Mrs. Percival (C.);St. Johns River, New Brunswick, 1885, G. U. Hay (C.); Saguenay River, 1890, 7. F. Allen (C.); Moose River Basin, Northern Ontario, 1903, J. M. Bell; Punk Island, Lake Winnipeg, 1884, Macoun. MaINeE: Moosehead Lake, 1875, Charles E. Smith; Cabossucontu Lake, Monmouth, 1899, W. C. Kendall; shallow pond, St. Francis, Aroostook County, 1893, Fer- nald 10; Sunkhaze Stream, Milford, 1892, Fernald (N. E.); West Baldwin, Cumberland County, 1894, Furbish (N. E.); Piscataquis River, Dover, 1894, Fernald (N. E.); Orono, 1878, Scribner (Greene); Penobscot River, Somerset County, 1882, Charles E. Smith (Phila.); near the east branch of the Penob- scot, 1847, A. Young, jr. (C.); without locality, 1847, Thurber (Gray); Green, 1878, Scribner (Mo.); Winthrop, 1862, Sturtevant (Mo.). Vermont: Shelburn, 1879, Pringle; Winooski River, Colchester, 1876, Pringle; Joes Pond, Danville, 1894, Grant & Eggleston (C.); Lake Memphremagog, 1860 (N. Y.); Little Otter Creek, Ferrisburg, 1880, Z. & C. E. Faxon (Gray); Shoreham, 1878, Brainerd (Mo.); Burlington, 1841, John Carey (Mo.). New Hampsuire: Ponds, no locality, Oakes Herbarium. action is in the highest degree arbitrary, as contravening a cardinal principle” (more specifically article 15 of the Vienna Code itself: namely, that the only valid designation of a group of plants is the earliest name applied to it within certain clearly defined limitations) article 49 is one of the portions of the Vienna Code accepted by the Nomen- clature Commission of the Botanical Club of the American Association for the Advance- ment of Science, and thus incorporated in the American Code of 1907 now used as the standard by writers in the Contributions from the United States National Herbarium. Apart from its contravention of the ‘‘cardinal principle” which lies at the base of all stability in nomenclature, article 49 is further objectionable on account of the encour- agement which it offers to slovenly and incomplete study of the literature, and to the multiplication of useless new names; while finally, though here the situation is bright- ened by a note of comedy, it rests on the tacit assumption that between tribe and family, or subgenus and genus, or subspecies and species, there is an actual, knowable difference of kind.—G. 8. M. 1 The letters in parentheses refer to the herbaria where the species are to be found. “Gray” denotes the Gray Herbarium; ‘‘N. Y.,”’ that of the New York Botanical Garden; ‘‘C,”’ the herbarium of Columbia College, deposited at the New York Bo- tanical Garden; ‘‘Mo.,’’ the Missouri Botanical Garden; ‘‘N. E.,’’ the herbarium of the New England Botanical Club, deposited with the Gray Herbarium; ‘‘Greene,”’ the private herbarium of Dr. E. L. Greene; ‘‘Phila.,’’ the herbarium of the Academy of Natural Sciences of Philadelphia. All specimens not marked thus are in the Na- tional Herbarium. 74 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Specimens examined—Continued. Dried—Continued. Massacuusetts: Holyoke, 1883, G. R. Lumsden; Concord River, Concord, 1886, H. 8S. Richardson (N. E.); Northampton, 1871, Jesup. Connecticut: Milford Pond, in river meadow, 1895, Eames; Derby, Oakes (Phila.); North Haven, 1899, C. H. Bissell 335 (N. Y.); Reynolds Bridge, 1898, E. B. Harger (N. E.) New York: Troy, Dr. W. E. A. Atkin (Phila.); near Niagara, Doctor Eddy (C.); Raquette Lake, 1896 (C.); Penn Yan, Sartwell (Gray); Albany Lake, 1879, Ward; Gray, Herkimer County, 1901, House; McDonough, 1884, A. L. Coville. - - -- -- - -—-"" -- rd ¥ aeeoer” o-- eee eer - a cee reese “eee, aodant . %, iy ‘ Fig. 4.—Map showing distribution of (a) Nymphaea microcarpa; (b) N. orbiculata; (c) N. microphylla. PENNSYLVANIA: Colliers Ferry, Lancaster County, 1863, Porter (Gray); Toby- hanna Creek, Pocono Mountains, 1858, T'raill Green (Gray); Naomi Pines, Monroe County, 1893, Dr. & Mrs. Britton; Monroe County, Traill Green, Silver Lake, Pike County, 1899, Stewardson Brown (Phila.); Bristol, 1865, C. F. Parker (Phila.); Philadelphia, Nuttall (Phila.); head of Naomi Pines Lake, 1904, Harshberger (Mo.). New Jersey: Closter, 1860, Austin (C.); Pompton Lake, Pompton, 1892, Morong (C.). This, the smallest of our Nymphaeas, is the American representative of the Palearctic N. pumila. Indeed it has often been considered identical with that species. The MILLER AND STANDLEY—-NORTH AMERICAN NYMPHAEA. 75 leaves of the two are much alike, so too the flowers, in size at least. The anthers of our plant are oblong, while in N. pumila they are cubical. In the European species the stigmatic disk is only partially tinged with orange-red, while in our plant the whole disk is a dark carmine. Michaux, in the original description of the plant, says: ‘‘Quamvis differentiam specificam verbis notare non possim; diverse tamen videtur.’’ Evidently he had only dried specimens, otherwise it would have been very easy to separate it from Nymphaea lutea, of which he considered it a subspecies. 2. Nymphaea rubrodisca (Morong) Greene.' Nuphar advena hybrida Peck, Ann. Rep. N. Y. Mus. Nat. Hist. 34: 53. 1881. Nuphar rubrodiseum Morong, Bot. Gaz. 11: 167. 1886. Nymphaea rubrodisca Greene, Bull. Torrey Club 15: 84. 1888; Robins. & Fern. in A. Gray, Man. ed. 7. 391. 1908. Nymphaea fletcheri Lawson, Proc. & Trans. Roy. Soc. Canada 64: 119. 1888. Nuphar advena minus Wats. & Coult. in A. Gray, Man. ed. 6. 56. 1889, not Morong. Nymphaea hybrida Peck, Bull. N. Y. State Mus. 6: 75. 1899; Britton, Man. ed. 2. 407. 1905. Type tocauiry: Lake Champlain, Vermont. on. ; . di DisTRIBUTION: Quebec and Ontario, through Fia. 5. narra rubrodisca, New York and the New England States to eastern Pennsylvania and northern New Jer- sey, also in eastern Minnesota and western Wisconsin. DESCRIPTION. Floating leaves 75 to 200 mm. long, 55 to 145 mm. wide, oval or ovate, rounded at the apex; sinus about half as long as the midrib or slightly longer, closed or very narrow; blades rather thin, glabrous, their lobes oblong-triangular or rounded-triangular, obtuse; submersed leaves well developed, very thin, crispate, broadly oblong or ovate, broadly rounded and retuse at the apex, of about the same size as the floating ones, their sinuses broader and more open; rootstocks comparatively slender, somewhat flattened, about 25 mm. in their longest diameter, the FG. 6.—Stigmatie leaf scars elliptical or oval, 8 to 10 mm. long; flowers 25 to 35 mm. peter “" ayn in diameter, 20 mm. high, depressed-globose; sepals usually 5, vetural sive. glabrous, rather thin, all similar, oblong or oval or almost orbicular, obtuse or truncate, only slightly narrowed at the base; petals spatulate, truncate, 8 or 9 mm. long; stamens in 4 or 5 rows, their anthers about one-half as long as the filaments, sometimes longer, but always shorter than the filaments; sepals canary yellow, with or without red on their inner surfaces, the red when present less vivid than in ameri- cana; petals clear yellow or slightly tinged with green; stamens yellow, the anthers light buff; ovary greenish yellow, lighter than the sepals and stamens, faintly marked 1 The use of the synonym rubrodisca in place of the original name hAybrida is in ac- cordance with article 49 of the Vienna Code. See footnote under Nymphaea micro- phylla, p. 72.—G. 8. M. 76 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. with carmine beneath the disk, the disk carmine; fruit ovoid, slightly constricted above, about 22 mm. high and 10 mm. in diameter, rather strongly and coarsely ribbed; disk when mature about 11 mm. in diameter, orbicular, its edge undulate; stigma rays 8 to 13, usually 10 to 12, linear, 3 mm. long, extending almost to the edge of the disk, distinct, with no trace of a median line, the center deeply depressed, smooth; fruit when mature dark purplish brown; seeds ovoid, 2.5 to 3mm. long, 1.5 to 2 mm. in diameter, light brown, shining. (Pxate 36, B, facing p. 73. Ficures 5, 6, 7, 6.) Specimens examined: In formalin— CANADA: Ottawa, 1901, Fletcher. New Hampsuire: Cambridge River near Lake Umbagog, 1905, W. C. Kendall. New York: Smiths Pond; Gray, 1901, J. Perkins; Blind Bay near Fishers Land- ing, Jefferson County, 1902, Mazon; Thousand Island Park, 1902, Maxon; Lake Champlain, 1901, Eggleston. WISCONSIN: West Superior, 1902, Charles Bullard. Dry— Canapa: Danville, Quebec, 1894, Berg (C.); near Pictou, Nova Scotia, 1901, Howe & Lang 610 (N. Y.); Red Pine Lake, 1900, Macoun 23261 and 23262 (N. Y.); Whites Lake, 1900, Macoun 21698 (N. Y.); Brigham Creek, Ottawa River, 1882, Fletcher; Ottawa, 1902, Fletcher; St. Francis River, Quebec, Eggleston 3010. Maine: Androscoggin River, Auburn, 1896, Merrill 4430 (N. E.); Birch Brook, north end of Cross Lake, Eagle Lake Chain, Aroostook County, 1903, W. C. Kendall; Hartford, 1886, Parlin (Gray); Milford, 1892, Fernald (Gray); Valley of the St. Francis River, Aroostook County, 1902, Eggleston & Fernald (N.E.); Mattawamkeag Lake, Aroostook County, 1897, Fernald (N. E.); Bradley, 1890, Fernald (N. E.). New Hampsutre: Gilmore Pond, Jaffrey, 1897, Robinson 263 (Gray). Vermont: Joes Pond, Danville, 1894 (C.); Lake Champlain, Ferrisburg, 1879, Brainerd (C.); Burlington, 1878, Pringle (N. Y.); Manchester, 1898, M. A. Day (Gray); Williamstown, 1881, L. & C. E. Faxon (Gray); Lake Champlain, 1873, H. G. Jesup (Gray); sluggish tributaries of Lake Champlain, 1879, Pringle. Massacnusetts: Woburn, 1909, Wim. Boott (Gray); Ashburnham, 1896, Sydney Harris (N. E.). New York: Wilmurt Lakes, Herkimer County, 1901, House; vicinity of Fishers Landing, Jefferson County, 1902, Robinson & Maxon 75; Carpenters Pond, Onondaga County, 1903, House; Troy, 1876, J. H. Wibbe (Greene); Raquette Lake, 1896 (C.); Troy, 1829, Doctor Atkin; Bronx River, Williams Bridge, 1890, Bicknell (C.); ponds in the Adirondacks, 1884, Morong (N. Y.); West- chester County, Williams Bridge, Dr. 7. P. Allen (N. Y.); Smiths Pond, Herkimer County, 1855 (Gray); Gorham, Doctor Torrey. New Jersey: Barrack Creek, Burlington, Solomon Conrad (Phila.); Pompton Lake, Pompton, 1892, Morong (C.); Green Pond, Morris County, 1894, Wm. Van Sickle. PENNSYLVANIA: Pocono Plateau, Monroe County, 1893, Dr. & Mrs. Britton. WIscoNSIN: West Superior, 1902, Charles Bullard. Minnesota: Daniels Lake, 1891, #. F. Wolf; Vermilion, 1891, Sandberg 500. This species is so readily distinguishable from the other American members of the genus, and particularly from those which occupy the same general region, that no special comparisons are required. By many authors the plant has been regarded as a hybrid between Nymphaea americanaand N. microphylla. Itscharactersare for the most part intermediate between those of the two better-known plants. It shows a less degree MILLER AND STANDLEY—-NORTH AMERICAN NYMPHAEA. TT -of fertility than other members of the genus, so much so that ripe fruit with well-devel- oped seeds is not easily obtained. In geographic range it appears to be a boreal form, occurring therefore in the same region as NV. microphylla, and in that part of the range of N. americana which extends north of the Transition Zone and overlaps that of the smaller plant. All of these facts might readily be construed as indicating a hybrid origin. On the other hand the characters of Nymphaea rubrodisca are quite as con- stant as in other American species, and no specimens are yet known which show any tendency to reversion toward either of the supposed parents. The plant is by no meansinvariably found locally associated with both N. americana and N. microphylla, as s Pose eee ee -_ oe - _eoorr™ a-- -- le ae o=ste -- = - ~*~ - ate woneeer7” oo Fig. 7.—Map showing distribution of (a) Nymphaea ovata; (b) N. rubrodisca; (c) N. ludoviciana; (d) N. ulvacea; (e) N. bombycina; (f) N. sagittifolia. either or both of these may be absent from the particular stream or pond in which it grows;' while in general distribution it extends decidedly beyond the western limit of the range of N. microphylla as now known. Finally, it is not unusual among plants for certain members of a genus to produce fruit less abundantly than others, particularly in groups where asexual reproduction and dispersal can readily occur. The hypothe- sis of hybrid origin seems, therefore, to present the greater number of difficulties, and until more facts can be brought to its support we prefer not to accept it. 1Tt was found thus alone by Peck at the type locality of hybrida, 783 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 3. Nymphaea americana (Provancher) Miller & Standley. Nuphar americana Provancher, Fl. Canad. 29. 1862, excluding synonyms. Nuphar variegatum Engelm.; Peck, Ann. Rep. Univ. N. Y. 19:73. 1866. Nuphar advena variegatum Engelm. in A. Gray, Man. ed. 5. 57. 1867. Nuphar advena minor Mo- rong, Bot. Gaz. 11: 167. 1886. Nymphaea variegata G. 8. Miller, Proc. Biol. Soc. Washington 15: 13. 1902. ‘‘Nymphaea advena So- land.’’ Small, Fl. South- east. U. 8, 456. 1903, in part; Britton, Man. ed. 2. 407. 1905, in part. Nymphaea advena variegata Fernald, Rhodora 10: 49. 1908. TYPE LocaALiTy: Lake St. Jean-Georgie, Quebec. DIsTRIBUTION: Eastern Canada westward to British Columbia, south to Montana, Nebraska, northern Indiana, and Ohio, eastern Pennsyl- vania, and New Jersey. DESCRIPTION. Floating leaves usually 17 to 28 cm. long and 11 to 22 cm. wide, oblong or oval, the blades averaging narrower and smaller than in advena, slightly but noticeably narrowed towards the apex, the sinus 50 to 75 mm. deep, closed or very narrow, the lobes semiorbicular or oblong; submersed leaves sometimes but not always present, similar in form to the floating ones but broader, very thin and membranous; petioles slender, con- spicuously flattened, with a conspicuous median ridge form- ing a prolongation of the midrib; peduncles terete, slender, glabrous; flowers about 45 mm. in diameter, 30 mm. high, the perianth when spread measuring about 100 mm., in other particulars similar to those of advena; petals about 16, the largest 8 mm. wide at the apex and sometimes almost spatu- late; stamens usually in 6 rows, about 26 to the row; filaments Fic. 9.—Stigmatic pat- often 4 mm. wide; color of flowers variable but usually quite tern of Nymphaea distinct from that of N. advena; outside of outer sepals oil Sue Natural green, the inner ones lemon yellow outside; lower half of , ‘inner surface of all sepals usually, although not invariably, red, this color some- times covering the whole inner side of the smaller sepals, the exact shade varying in different specimens from a mere indefinite tinge in the yellow to a bright pure maroon; petals clear yellow or sometimes bright parrot green, usually tipped with green, and Fia, 8.—Leaf outline of Nymphaea americana. Seale, }. PLATE 37. 16. Contr. Nat, Herb., Vol. "YNVLNOI ‘SNV7] GVSHLVI4 NI SASTONVLS 3 HST (YSHONVAOH) YNVOINSWY YSVHIWAN PLATE 38. at. Herb., Vol. 16 ! \ Contr. N SLLSSNHOWSSW) “OT3SISONIHdS WOUS ‘ATTONVIS I YATTIW (YSHONVAOU) VNVOINSWY VAVHdWAN 4O LIN MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 79 occasionally tinged or tipped with red; stamens clear, bright lemon yellow throughout; fruit smaller than that of advena, 40 mm. high and 30 mm. in diameter, ovoid, con- spicuously constricted above, the ribs less prominent and the crater more shallow than in that species, the crater usually about 5 mm. deep; margin of the disk definitely although not deeply crenate; stigma rays 7 to 25, usually 12, 13, or 14, extending to within 1 mm. or less of the edge of the disk; capsule oil green, usually tinged with red, and often becoming a bright red with age; in the brightest colored fruits the stigmatic region usually remaining greenish yellow in striking contrast; seeds similar to those of N. advena but smaller, about 5 mm. long, the greater diameter 3.6 mm., the lesser diameter 3.2 mm., less compressed and witha less prominent raphe. (PLATEs 36, C, facing p. 73; 37, 38. Fiacures 8, 9, 10,c.) EXPLANATION OF PLATES 37, 38.—PIl. 37, Nymphaea americana, in Flathead Lake, Montana. Photo- graphed by Prof. M. J. Elrod. PI. 38, fruit of Nymphaea americana, collected at Springfield, Massachusetts, by Mr. Frederick Knab. Natural size. Specimens examined: Formalin— Canaba: Port aux Basques, Newfoundland, 1901, F. W. True; 150-mile House, Cariboo, British Columbia, 1901, A. C. Brooks; near Toronto, 1901, J. H. Fleming; Lake Joseph, Muskoka, Ontario, 1902, Fleming; Ottawa, 1902, Fletcher; Pointe 4 Pic, 1902, D. G. Elliott; St. Clair River near Walpole, Ontario. Marne: Maneskootuck, Rangeley, 1901, F. S. Dickson; Green Lake, 1901, E. E. Race; Cape Niddick, 1901, Charles Bullard; Little Sebago Lake; Birch Brook, Eagle Lake Chain, Aroostook County, 1903, W. C. Kendall. New HampsuireE: Intervale, 1901, G. M. Allen; Mud Pond, tributary to first Connecticut Lake, Coos County, 1904. Vermont: Lake Champlain, 1901, Eggleston. Massacuusetts: Springfield, 1901, Frederick Knab; Wareham, 1901, 0. Bangs; Stockbridge, 1908, J. A. Loring. Connecticut: Pembroke Lake near Bridgeport, 1902, G. A. Meeker. RuoveE Isranp: Lymansville, 1901, Angell & Cash; without locality, 1900, Mearns. New Jersey: New Bedford, Monmouth County, 1902, M. W. Lyon; Clementon, 1902, S. N. Rhoads; Spring Lake, 6 miles south of Ocean Grove, 1902, Lyon. PENNSYLVANIA: Three and one-half miles east of Lopez, Sullivan County, 1901, Witmer Stone; pond near Lehigh River at Lehigh Gap, Lehigh County, 1901, J. A. G. Rehn; Shady Nook. New York: Chautauqua Lake, 1901, M. Schlegel; Peterboro, 1901, B. D. Miller; Piseco, 1901, W. L. Ralph; Lake Titus, 1901, EZ. W. Nelson; South Mountain Lake, Catskill Mountains, 1903, Shull; Fish Creek, Oneida Lake, 1901, Mazon; Thousand Islands, 1902, Mazon; Machias, 1901, F. E. Ferris; Sodus Bay, 1901, G. B. Turner; Smiths Pond, B. D, Gilbert. Ouro: Sandusky, 1902, W. A. Kellerman; Sandusky, 1903, M. T. Cook; Squaw Bay, Put in Bay Harbor, 1901, U. 8. Fish Commission. Micuican: Belle Isle Park, Detroit, 1907, Farwell; St. Clair Flats, 1901, U. S. Fish Commission. Wisconsin: Lake Superior, 1901, H. V. Ogden; Green Bay, 1903, J. H. Schuette; Upper Nemahin Lake, Waukesha County, 1901, H. V. Ogden; Milwaukee, 1902, Ogden; West Superior, 1902, Charles Bullard. Inp1ana: Wolf Lake, 1903, Shull. Iowa: Manchester, 1901, R. S. Johnson. Minnesota: Lake Itasca, 1902, 7. S. Robertson; St. Louis River near Duluth, 1901, L. E. Balbridge. Montana: Big Fork, 1902, M. J. Elrod; Big Fork, 1901, W. C. Barr. 80 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Specimens examined—Continued. Dry— CANADA: On an island of the Simpson Group 40 miles northeast of Fort Resolu- tion, Mackenzie Territory, 1903, Preble 242; St. Francis River, Quebec, 1902, Eggleston 3011; Canso, Nova Scotia, 1901, Fowler; Golden Lake, Renfrew County, Ontario, 1899, Umbach; Gananoque, Ontario, 1887, Fowler; Cache Lake, Algonquin Park, Ontario, 1900, Macown 21697; Port aux Basques, Newfoundland, 1901, A. W. Prentiss; North Sidney, British Columbia, 1883, Macoun 8 (Gray); pond near Whitbourne, Newfoundland, 1894, Robinson & Schrenk (Gray); Seven Islands, 1907, C. B. Robinson 889 (N. Y.); Channel, he e o n 7 ee meee eee / ' / ‘ 7? 1 - owners | bene enee eed porn --37" 1 ase, J | etait rr are . l i r ‘ wales * *s i ‘ a \ ~wceneeeerr pO | — ' ' \ b *, | weree | Me ' i 1 @ e. ' H e -en ' poooeedh eonrrr™ bd ey. ae Fia 10.—Map showing distribution of (a) Nymphaea puberula; (b) N. fluviatilis; (c) N. americana. Newfoundland, 1901, Howe & Lang 870, 944 (N. Y.); near Digby, Nova Scotia, 1901, Howe & Lang 313 (N. Y.); St. Jean ?Evangeliste, Nouvelle, Quebec, 1904, Collins & Fernald (Gray); Belleville, Ontario, 1883, Macown 95; Cross Lake, Renfrew County, Ontario, 1899, Umbach; Killarney, Mani- toba, 1896, Macoun, (Greene); Kaministigma River, 1889, Dr. & Mrs. Brit- ton (C.); Fredericton, New Brunswick, 1880, Fowler. MILLER AND STANDLEY—-NORTH AMERICAN NYMPHAEA. 81 Specimens examined—Continued, Dry—Continued., ‘ Maine: Valley of Saint Francis River, 1902, Eggueston & Fernald (Gray); Petti- quaggamas Lake, Aroostook County, 1893, Fernald 9; Orono, Penobscot County, 1897, Fernald (N. E.); Industry, Franklin County, 1894, Fernald (N. E.); Manchester, 1874, Scribner (N. E.); West Auburn, 1896, Merrill 4429 (N. E.); Westbrook, 1899, Ricker 605; Clear Pond, Pleasant Ridge Town- ship, Somerset County, 1906, Chamberlain; Winthrop, 1862, Sturtevant. New Hampsuire: Frost Pond, Jaffrey, 1898, B. L. Robinson 495 (Gray); White- field, Coos County, 1896, Deane (N. E.). Vermont: Johnson, 1893, Grout (C.); Willoughby, 1892, Rusby (C.); Th Plot River, Shelburn, 1879, Pringle; Barnumville, Manchester, 1898, Mf. A. Day (Gray). Massacuusetts: Vicinity of Cambridge, 1891, Underwood 2694 (C.); without locality, 1865, Gray; South Framingham, 1888, Sturtevant; Dartmouth, 1889, Sturtevant; Ashland, 1879, Morong; Waverly, 1895 (Gray); Melrose, 1880, R. Frohock (N. E.); Malden, 1872, H. L. Wordy (N. E.); Beaver Brook Reser- vation, west of Kame, 1896, Deane (N. E.); Eastham, 1907, F. S. Collins (N. E.); Stockbridge, Berkshire County, 1902, R. Hoffmann (N.E.); Stony Brook, 1895, W. H. Manning (N. E.); Great Barrington, 1894, Pollard. Connecticut: West Goshen, 1891, Underwood 3208, 3210, 3210a (C.); Southing- ton, 1899, L. Andrews (N. E.); Tranquility Farm, Middlebury, 1896, W. M. Shepardson (N. E.); New Haven, 1884, Safford 164. New Jersey: Swartzwood Lake, 1906, C. S. Williamson (Phila.); New Bedford, Monmouth County, 1902, M. W. Lyon; Egg Harbor City, 1882, J. H. Red- field 243; Inskip, 1894, Brinton & Keller; Mays Landing, Isaac Burk. PENNSYLVANIA: Pocono Plateau, 1904, Harshberger (Phila.); Shady Nook, Sulli- van County, 1901, Stewardson Brown (Phila.). New York: Steeres Pond, Preston, 1886, F. V. Coville; South Mountain Lake, Catskills, 1903, Shull; Smiths Pond, Litchfield, Herkimer County, 1903, House; Peterboro, Madison County, 1904, G. S. Miller; Brisben Station, Chenango County, 1887, F’. V. Coville; Geneganslet Lake, McDonough, 1886, F. V. Coville; South Bay, Wellesley Island, Jefferson County, 1902, Robinson & Mazon 74; Troy, Schweinitz Herbarium; Gorham, Torrey; Raquette Lake, 1896 (C.); Lake Ontario, 1892, Britton (C.); Great Otter Lake, Lewis County, 1884, O. R. Willis (C.); Sylvan Beach, Oneida County, 1900, House; Little York, Cortland, 1905, G. T. Hastings; South Bay, eastern end of Oneida Lake, 1910, Mazon 4616. Onto: Sandusky Bay, 1902, J. H. Schuette; Sandusky, 1902, Kellerman; Black River, Lorain County, 1892, J. W. Beach; Sandusky, 1903, Mel T. Cook; Fox Lake, Wayne County, 1899, Selby & Duvel. Wisconsin: Mirror Lake, 1903, Eggert; Waupaca, 1907, . F. Garerche; St. Croix Falls, Polk County, 1899, Baker (Gray); near Milwaukee, 1902, H. V. Ogden, De Pere, 1888, 7. S. Kellogg. MicuicaAN: Iron River, 1902, Rydberg (N. Y.); west end of Swan Lake, Allegan County, 1902, Wight 5; Bloody Run, Detroit, 1865, Bigelow (Mo.); Parkhouse Marsh, Detroit, 1866, Bigelow (Mo.); Connors Creek, Detroit, 1866, Bigelow (Mo.); Belle Isle, Detroit, 1865, Bigelow (Mo.). Minnesota: Silver Lake, 1887, Cratty; Lake Itasca, 1902, T. S. Roberts; Lake Kilpatrick, Cass County, 1893, Ballard; Minnetonka, 1891, Sandberg 631. Nepraska: Weigand, 1893, Fred Clements 2686; Lake Region of Grant County, 30 miles south of Whitman in Swan Lake, 1893, Rydberg 1650; Cherry County, 1892, Smith & Pound 168 (Mo.). 82 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Specimens ecamined—Continued. Dry—Continued. , Iowa: Iowa City, Hitchcock (Mo.); Hamilton County, 1891, P. H. Rolfs (Mo.); Armstrong, Emmet County, 1887, Cratty (Mo.); Iowa and Minnesota Line, 1897, Pammel 520 (Mo.); Winnesheik County, 1895, Fitzpatrick (Mo.). Sourn Daxora: White, 1893, Thornber; Six Mile Creek, west of White, 1902, .A. G. Johnson; White, 1893, T. A. Williams; Aurora Creek southeast of Brook- ings, 1904. . Montana: Rost Lake, 1901, MacDougal 652; Columbia Falls, 1892, R. S, Wil- liams 869. For comparison of this plant with Nymphaea advena see page 88. 4. Nymphaea fraterna Miller & Standley, sp. nov. Type in the U. 8. National Herbarium, no. 441399, collected in Toms River, New Jersey, August 6, 1903, by M. W. Lyon. DIstRIBUTION: East- central New Jersey. DESCRIPTION. Petioles flattened, with a median ridge along the inner side, glabrous; float- ing leaf blades oblong- ovate or ovate, 10 to 22 cm. long and 8 to 15 cm. wide, broadest at or below the middle, conspicuously narrowed and _ acutish towards the apex, rather thin, glabrous; sinuses open and very narrow, or closed and the lobes partly overlapping, 20 to 65 mm. deep, the lobes rounded; submersed leaves very thin and delicate, nu- merous, broader than the floating ones, broadly ovate or almost orbicular in outline, 85 to 155 mm. long and 70 to 135 mm, wide, broadly rounded or slightly narrowed at the Fic, 11.—Leaf outline of Nymphaea fraterna. Scale }. apex, the sinus 25 to 50 mm. deep, usually open, rather narrow, the lobes broadly rounded, the margins somewhat plicate; flowers depressed-globose, 22 to 26 mm. in diameter, only slightly depressed at the base; sepals 6, when spread measuring 50 to 65 mm.; outer ones thin, glabrous, rather narrowly oblong, 16 to 21 mm. long and 11 to 13 mm. wide, rounded at the apex; inner sepals deltoid-cbovate, shallowly emarginate, 18 to 23 mm. long, gradually narrowed to the base; stamens in usually 4 rows, their anthers slightly longer than the MILLER AND STANDLEY—-NORTH AMERICAN NYMPHAEA. 83 filaments; petals narrowly oblong, truncate, 6 or 7 mm. long; fruit ovoid, gradually narrowed above to a short neck, 21 to 30 mm. high and 17 to 20 mm. in diameter, inconspicuously ribbed above, almost smooth near the base; rim of the disk 3 mm. high, divergent or erect; disk 12 to 14 mm. in diameter, orbicular, entire or slightly crenate, the crater usually shallow, 2 to 4 mm. deep; rays 11 to 16, usually 14, linear, 3 or 3.5 mm. long, narrow, extending to within 1 or 2mm. of the edge of the disk, distinct, the center3 or 4 mm. in diameter, smooth; seeds 3.5 to4 mm. long, ovoid, pointed, with an acutish and rather conspicuous raphe., (Piates 35, B, facing p. 72; 36, D, facing p. 73. Fieures 11, 12, Fic. 12.—Stigmatic 13, 5.) pattern of Nym- yo.) ; h ; Described from material preserved in formalin collected by hace freterna W. M. Lyon, jr., in Toms River, New Jersey, August, 1902, July 27, 1903, August 6, 1903, and August 15, 1902. Dried specimens collected August 6, 1903, are mounted on sheets 441395 to 441399, inclusive, in the National Herbarium. === —--- J Ae Fia. 13.—Map showing distribution of (a2) Nymphaea ozarkana; (b) N. fraterna; (c) N. chartacea. Additional material seen: In formalin— New Jersey: Pemberton, June 24, 1900; Speedwell, Burlington County, June 20, 1901, Witmer Stone. 84 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Additional material seen—Continued. Dry— New Jersey: Toms River, July 26, 1903, Lyon; same locality, August 11, 1902, Lyon; New Bedford, Monmouth County, August 14, 1902, Lyon; Forked River, May 29 to June 2, 1896, collected on an excursion of the Torrey Botan- ical Club. . This species comes from a region long known to botanists as one producing many interesting plants. Although closely related to Nymphaea americana it seems amply distinct in its mych smaller flowers, smaller, greenish fruit, pointed leaves, smaller seeds, and numerous and conspicuous submersed leaves. While submersed leaves are occasionally found in N. americana they are never so numerous or conspicuous asin the New Jersey plant. 5. Nymphaea advena Ait. Nymphaea advena Ait. Hort. Kew. 2: 226. 1789; G. 8S. Miller, Proc. Biol. Soc. Wash- ington 15: 12. 1902; Small, Fl. Southeast. U.S. 456. 1903, in part; Britton, Man. ed. 2. 390. 1908, in part; Rob- ins. & Fern. in A. Gray, Man. ed. 7. 390. 1908. Nymphaea arifolia Salish. Ann. Bot. 2: 71. 1806. Nuphar advena Ait. Hort. Kew. ed. 2. 3: 295. 1811. Nuphar advena tomento- sum Torr. & Gr. Fl. N. Amer. 1: 58. 1838. TYPE LocaLity: Vicinity of Philadelphia, Pennsyl- vania. DistRIBUTION: Eastern Fic. 14,—Leaf outline of Nymphaea advena, Seale }. Wisconsin and Southern Michigan and New York to eastern Nebraska and Kansas, southern Missouri, Kentucky, and North Carolina. Northern limit coinciding with that of Upper Austral zone. DESCRIPTION, Leaves erect, usually borne above the surface of the water, occasionally floating in deep water; blades ovate to rounded oblong or oval, rather thick and firm, oil green, glabrous, 16 to 33 cm. long and 14 to 25 cm. broad, usually very broadly rounded at the apex; sinus 4.5 to 10 cm. deep, open, the lobes usually diverging at an angle of about 80°; lobes mostly triangular, often acutish; peduncles stout, glabrous; petioles stout, subterete, glabrous; flowers depressed-globose, 30 to 40 mm. in diameter, about 23 mm. high when normally spread, the perianth when spread measuring about 80 mm.; sepals usually 6, the 3 outer broadly ovate, about 35 mm. long and 25 mm. wide, obtuse, the three inner suborbicular, about 35 mm. long and 45 mm. wide, Contr. Nat. Herb., Vol. 16. PLATE 39. INDIANA. IN MONTEERS POND, KNOX COUNTY, NYMPHAEA ADVENA AIT., Contr. Nat. Herb., Vol. 16. PLATE 40, A. EARLIER VERNAL STAGE OF NYMPHAEA ADVENA AIT., Four MILE RUN. VIRGINIA. B. LATER VERNAL STAGE OF NYMPHAEA ADVENA AIT., FOUR MILE RUN, VIRGINIA. MILLER AND STANDLEY—-NORTH AMERICAN NYMPHAEA. 85 their bases abruptly narrowed to a short claw about 6 mm, wide, truncate or retuse at the apex; petals about 20, cuneate-oblong, 8 mm. long and 3 to 5 mm. wide, trun- cate or retuse at the apex; stamens usually in 7 rows, varying from 5 to 8, about 35 to the row; filaments cuneate-linear, 10 mm. long, 2 mm. broad at the tip; anthers a little more than half as long as the stamen; outer sepals rich oil green outside, lighter within and occasionally but rarely tinged with red; inner sepals clear yellow throughout except at the tips, there tinged with green, the inner side occasionally tinged with purplish red;! petals yellow like the inner sepals, usually tinged with red; filaments dull red except the ex- posed tip, this yellow; anthers and pollen yellow; stamens occasionally clear yellow throughout, all becoming dull red throughout in age; fruit subglobose, about 40 mm. high and 50 mm. in diameter (the height usually less than the diam- eter, but this character not constant), with conspicuous longitudinal ribs extending its entire length; stigmatic disk orbicular, entire, faintly undulate, strongly concave; stigma rays distinct, varying in number from 9 to 23, but usually 15, 16, 17, or 18, 7 to 8 mm. in length and about .75 mm. wide, extending to within about 2 mm. of the edge of the disk, without a median furrow; capsule oil green, the stigmatic region abruptly lighter and more yellowish; seeds elliptical or obovate, slightly flattened laterally, with a prominent raphe, 6 mm. long, greater diameter 5 mm., lesser diameter 4 mm. (Piates 35, C, facing p. 72; 36, E, facing p. 73; 39, 40. Ficurzs 14, 15, 16, b.) Fig. 15.—Stigmatic pattern of Nymphaea advena, Natural size. EXPLANATION OF PLATES 39, 40.—PI. 39, Nymphaea advena, at Monteers Pond, Indiana, photographed by Mr. Robert Ridgway. PI. 40, vernal stage of Nymphaea advena, at Four Mile Run, Alexandria County‘ Virginia; photographed by Mr. Gerrit S. Miller, jr., (A) April 20, (B) May 4, 1902. Specimens examined: In formalin— New York: Lake Grove, Long Island, 1901, A. H. Howell; Sing Sing, 1901, E. Acker; near Croton, 1901, Van Cortlandt. New JERSEY: Whale Pond Creek, Monmouth County, 1902, M. W. Lyon; Milburn, Essex County, 1902, Lyon; arm of Newton Creek, Collingswood, Camden County, 1902, S. N. Rhoads; Centerton, 1900, G. S. Miller; Haddonfield, 1907, Rhoads, PENNSYLVANIA: Erie, 1906, 8S. E. Bacon; same locality, 1902, Bacon; Philadelphia, 1901, Witmer Stone; Meadville, 1902, H. C. Kirkpatrick. MARYLAND: Pocomoke City, 1902, W. P. Hay; Havre de Grace, 1902, Shull. Vireinta: Holmes Run, Fairfax County, 1901, G. S. Miller; Luray, 1901, Lewis Willis; Clifton, 1901, Miller; Four Mile Run, Fairfax County, 1901, Miller. West Virernia: Marlinton, 1902, W. B. Kellerman. Nort Carona: Raleigh, 1901, Brimley Brothers; Hendersonville, 1901, Clayton. Texas: Fort Clark, 1901; Del Rio, 1901, Franks. Outro: Buckeye Lake, Licking County, 1902, W. A. Kellerman; Cadiz Junction, _ _Harrison County, 1902, W. A. Kellerman; Sandusky, 1903, M. T. Cook. InpDIANA: Merrillville, 1903, Shull; Winona Lake, 1902, C. A. King. InurNois: Mount Carmel, 1901, Schneck. MicuicANn: Northville, 1901, U. S. Fish Commission; Washtenaw County, J. B. Steere. 1 Among 137 flowers collected at Four Mile Run, Virginia, May 18, 1902, there were 16 with purplish-blotched sepals, 121 with red on stamens, and none with red on the disk. ; 86 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Specimens eramined—Continued., Dry—- New York: Ponds near New York, M. Ruger (Greene); Dougan Hills, Staten Island, 1890, Britton (C.); Valley Stream, Long Island, 1903, LZ. T. Hanks (N. Y.); Van Cortlandt Swamp, 1893, Pollard; Catskills, 1903, Shull; without locality, 1841, H. R. Browne. PENNSYLVANIA: Schuylkill River below South Street Bridge, Philadelphia, 1888, McElwee 301 (Phila.); Neshonning, Bucks County, 1891, McElwee (Phila.); McCalls Ferry, York County, 1904, Rose & Painter 8167; Meadville, 1902, H. C. Kirkpatrick; Mahoning, Carbon County, 1906, Krautter; Conewago Creek, eastern Pennsylvania, 1898, C. W. Eisenhower (Mo.); in fluvio Lecha Pennsylvaniae, 1832, C. J. Moser (Mo.); York, 1895, Glatfelter (Mo.). New Jersey: Newton Creek near Collingwood, Camden County, 1902, S. N. Rhoads; Milburn, Essex County, 1902, Lyon; Springfield, Essex County, near Rahway River, 1902, Lyon; Spring Lake near Ocean Grove, 1902, Lyon; Lake Hopatcoug, 1890, Nash 1004; Belleville, Essex County, 1902, Lyon; Delawanna Station below Passaic, 1902, Lyon; Clementon, 1902, Rhoads; Newark, 1878, C. H. Harding; Cedar Swamp, S. Conrad (Phila.). Onto: Cincinnati, R. Buchenau (Phila.); Chapelle Creek, Florence, 1903, Moseley; Salem, Columbiana County, 1903, H. S. Fawcett; Cadiz Junction, Harrison County, 1902, W. A. Kellerman 3882; near Canton, Nicholas Riehl (Mo.); Sandusky Bay, 1902, J. H. Schuette; near Salem, 1906, Fawcett. MicuicaNn: Detroit, 1855, Bigelow (Mo.); Coldwater, 1877, Milligan; Bloody Run, Detroit, 1866, Bigelow (Mo.); Connors Creek, Detroit, 1866, Bigelow (Mo.). Inpiana: Sanford Lake, Orange Township, Noble County, 1905, Deam 301; Lost Lake, Marshall County, 1903, Paul Bartsch; Merrillville, 1903, Shull; Winona Lake, 1902, C. A. King; Lake Maxinkuckee, 1900, Scovell & Clark 1261. ; Iuurnots: Bluffs Lake, 1876, Eggert; north end of Clover Lake, 1888, J. W. Davie (N. Y.); DuPage River near Naperville, 1897, Umbach; near Woodlawn, Washington County, 1898, Eggert (Mo.); Iuka, 1904, Jensen (Mo.); lakes in the American Bottom opposite St. Louis, 1838, Engelmann 771 (Mo.). Missouri: Near Poplar Bluff, 1893, Eggert (Mo.); St. Francois River, Dunklin County, 1897, Trelease 13 (Mo.); Carterville, 1906, Z. J. Palmer 888; Barton County, Broadhead Herbarium (Mo.). Wisconsin: Delavan, 1907, Ned Hollister; Green Bay, 1903, J. H. Schuette. NEBRASKA: Wiegand, 1893, Fred Clements 2686. Kansas: Chautauqua County, 1896, Hitchcock 607. Kentucky: Bear Creek, Edmonson County, 1901, Miss Price. District or CoLtumBIA: Eastern Branch, 1873, Ward 30; without locality, 1869, H. Brummell; without locality, 1885, McCarthy; Jackson City, 1897, Steele; Washington, 1878, Chickering (Mo.). MARYLAND: Patuxent, Anne Arundel County, 1905, House; Ardwick, Prince George County, 1903, Lyon; 9 miles southeast of Pocomoke City, 1902, W. P. Hay. Virginia: Ashland, De Chalmot; Alexandria, 1874, George Vasey; Ocean View, Norfolk County, 1898, Kearney 1464; Dismal Swamp, 1898, Kearney 1611; Great Falls, 1902, A. H. Howell; Passage Creek, Warren County, 1897, G. 8. Miller; Washington Canal near Duke, Dismal Swamp, 1893, Boettcher. Nort CAROLINA: Swamps near Hendersonville, 1897, Biltmore Herbarium 4281a. Texas: Devils River, Merican Boundary Survey; Fort Clark, Kinney County, 1893, Mearns 1365. Mexico: Tamesin River, Alta Mira, Tamaulipas, 1898, Goldman 95. 87 MILLER AND STANDLEY—NORTH AMERICAN NYMPHABFA. ‘ban.yjfiza Dusapo ‘AT (p) Snpodashjod ‘yy (2) ‘puaapn “az (q) ‘oyhydosspw Duaapo vaoydwfiyr (D) Jo VONNQLYSIp SuLmoys dep_—‘or “DIT rv seecocwwese ses 22075°—12——3 88 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, The differences between Nymphaea advena and N. americana (under the specific name of variegata) have been discussed by Miller in the Proceedings of the Biological Society of Washington, but it is worth while to repeat them here. The most striking difference existing between the two is found in the shape of the petioles. In N. advena these are subterete; in americana they are conspicuously flattened and there is a median ridge, a continuation of the midrib, running along the dorsal surface. This difference in petiole structure has an important effect upon the leaf habit of the two species. When the leaves of advena first appear in the spring the petioles are comparatively weak and the blades are often found floating upon the water; sometimes, however, when the plants are growing in mud or in very shallow water, especially when sub- jected to the action of tides, the blades are erect from the first. Almost invariably in the typical form, as the leaves become older the blades are held stiffly erect. In americana the weakness of the petioles, due to their flattening, makes it impossible for the blades to be held erect, hence they are always floating, or, if the water in which they are growing is lowered, they lie flat upon the mud. Asa consequence of this difference in habit, americana, driven away from very shallow water probably by injury following its drying up, is usually found in comparatively deep water; while advena, able to live in ground which is even moderately moist, clings to the shores, or even retreats to the moist meadows bordering streams and ponds. Habit is not the only respect in which the two plants differ. The outline of the leaf blades is so striking that it is almost always possible to separate the two plants certainly from herbarium material. In advena the lobes of the leaves are always pointed while in americana they are broadly rounded. The flowers, too, show apparently constant differences. Those of the latter species are usually somewhat larger. In addition to this the flowers of americana have the inner surface of the sepals blotched with red near the base, while in advena the corresponding region is normally shaded with green, and only in rare instances tinged with red. The fruit of americana is smaller, less strongly ribbed, and is usually tinged with red, while that of advena is uniformly green throughout. In the former the most common number of stigma rays is 12 while in the other species it is 16. The ranges of the two plants, as well, are rather clearly marked. Nymphaea ameri- cana is confined to the northern part of the United States and eastern Canada, extend- ing as far south as Pennsylvania. Nymphaea advena is confined to the southern United States, extending northward into New Jersey and southern New York. In the central States we know less of the ranges but all the data accessible indicate that the ranges there are what we would expect from conditions in the east. Thus it will be seen that americana is usually restricted to the Boreal and Transition zones while advena occurs in the Upper Austral. The range of advena in the extreme Southern States seems rather peculiar. Perhaps it would appear less so if we had fuller material from all parts of the range. The species is most common from eastern Pennsylvania south through Maryland and Virginia to northern Florida. We have seen no true advena from any of the Gulf States west of Florida. The plant has been reported from most of these States, but such material as we have seen is referable to species here described for the first time. Tn fact we are not certain that the plant appears again until we reach the Rio Grande region of western Texas. Here this same species is found in at least two localities. Fresh material which we have secured offers no means of separating this extreme western plant from the form so common in Pennsylvania and Virginia. Farther south, in Mexico, a plant is found which is apparently the same, though the driéd material that we have seen shows that the leaves of the Mexican plant are more acute than those of the typical form. Still farther west the plant has been reported again, this time in California. We have seen only dried material from that State, hence can not be sure as to its identity. It is probable that it is merely an abnormal form of Nymphaea polysepala. MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 89 In the National Herbarium there is an interesting specimen collected at Lake Ellis, North Carolina, July 3, 1908, W. H. Brown 72. The leaf blades are 11 to 17 cm. long and only 6.5 to 10 cm. wide; in outline they are lance-ovate and very acute at the apex; the sinus is very narrow or closed; flowers 35 mm. in diameter. Unfortunately the material is too scanty to show whether the form is anything more than an individual variation. Torrey and Gray’s subspecies tomentosum was based upon a specimen collected by Thomas Nuttall in the vicinity of Philadelphia and labeled by him Nuphar tomen- tosum. We have examined this specimen in the herbarium of the Philadelphia Academy of Science. The “pubescence’’ con- sists merely of algee or some similar low organ- isms which cover the lower surfaces of the leaves and the petioles. At present a part of the surface has become com- pletely glabrous, owing to the falling away of the artificial covering. This same phenomenon we have observed in other herbarium specimens. In the National Herba- rium is a specimen of some cultivated Castalia, which at first glance ap- pears to have coarsely dentate leaves. On closer inspection it js “seen that the teeth are masses of alge which have adhered to the edge of the blade. 5a. Nymphaea advena macrophylla (Small) Miller & Standley. Nymphaea macrophylla Small, Bull. Torrey Club 25: 465. 1898. Type Locauiry: The type, in the herbarium Fic. 17.—Leaf outline of Nymphaca advena macrophylla. Scale }. of Columbia College, was collected in August, 1894, in the vicinity of Eustis, Lake County, Florida, by Geo. V. Nash (no. 1751). DistrisuTion: Northeastern Florida. DESCRIPTION. Habit of leaves as in N. advena; blades 28 to 40 cm. long and 20 to 27 cm. wide, glabrous, ovate, acute or at least acutish, with a V-shaped sinus 8 to 13 cm. deep, the lobes triangular, acutish; no submersed Jeaves known; flowers depressed-globose, 32 90 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. to 38 mm. in diameter, about 22 mm. high; sepals 6, glabrous; the outer ones obovate, rounded, somewhat narrowed at the base, about 32 mm. long and 20 mm. wide; inner sepals thinner, deltoid-obovate, retuse; petals broadly cuneate, 10 mm. long, truncate or retuse; stamens in 6 rows, their anthers almost twice as long as the filaments; color of sepals as in N. advena; disk when young of the same color as the sepals; stamens lemon yellow; petals a slightly lighter yellow throughout; fruit narrowly ovoid, 35 mm. high and 25 mm. in diameter in the largest specimens, only slightly constricted above, coarsely ribbed above but almost smooth at the base; stigmatic disk orbicular, 20 mm. in diameter, depressed about 4 mm. in the center, its edge entire or slightly undulate; stigma rays linear, distinct, unequal, with slight traces of a median groove, 6 mm. long, extending to within 2 mm. of the edge of the disk, 10 to 15 in number, usually 12 or 14; body of fruit apple green, the disk chrome yellow. (Puare 36, F, facing p, 73. Fiaures 16, a, 17.) Specimens examined: In formalin— FrLoripa: Santa Fe River, southern edge of Columbia County, 1902, T. Wayland Vaughan; Dade City, H. 8. Fawcett; Whitfield, 1903, W. E. C. Todd; Jack- sonville, 1901, Curtiss; Kissimmee River, 1901, Mearns. Dry— Fiorina: Vicinity of Eustis, type; in the Everglades near the unfinished railroad grade between Cocoanut Grove and Cutter, 1903, Small & Carter 665 (N. Y.); vicinity of Eustis, 1894, Hitchcock; Kissimmee River, 1874, E. Palmer 7; Alligator Lake near Lake City, 1907, H. S. Fawcett; Jacksonville, 1901, Curtiss 6844; Jacksonville, 1894, Curtiss 4684; Southport Canal, Kissimmee Valley, Mearns; North Santee, 1837, G. Maurigault (N. Y.). Cuba: Without locality, 1860-64, Wright 1858 (Gray); Provincia de Pinar del Rio, 1904, Earle & Wilson 1656 (N. Y.); San Cristobal, 1905, M. T. Cook 130; Herradura, 1905, Jf. T. Cook 132; without locality, 1906, M. T. Cook 1, 6 (N. Y.). Dr. J. K. Small in the original description of this plant compares it with so-called Nymphaea advena, pointing out numerous differences. The plant with which it was compared was not advena but the northern Nymphaea americana. In the herbarium of the New York Botanical Garden at that time there were practically no specimens of N. advena, nearly all of those so labeled being americana. Evidently the Florida plant is amply distinct from the latter. The description of Nymphaea advena in Doctor Small’s Flora of the Southeastern United States applies to N. americana and the key separates americana (under the name advena) from advena (under the name macrophylla). It can readily be seen by examination of the key that the characters used for separat- ing the two species will not hold for separating true advena from the Florida plant. The material at our command, and it seems to be ample, does not warrant us in main- taining macrophylla as a separate species. The only difference that we can see lies in the larger size of the Florida plant and the longer, more acute, thicker leaves whose lobes are rather narrower.! It is not certain whether the Cuban specimens belong here. The leaf outline seems tobe thesame. Fresh material collected in 1910 in the vicinity of Havana by Brother Leén shows that the outer edge of the stigmatic disk is tinged with a dull purplish red. This color does not extend to the interior of the crater and is very different from the bright geranium red of the following subspecies. 1 Under the provisions of the American Code we would be justified in substituting a new name for the somewhat inappropriate macrophylia.—G. 8. M. MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 91 5b. Nymphaea advena erythraea Miller & Standley, subsp. nov. Type in the U.S. National Herbarium, in formalin, collected at Miami, Florida, by Mr. E. A. Brewer, April 17, 1902. Distrisution: Southern Florida. This appears to differ in no way from typical macrophylla except in having the disk of the fruit of a bright geranium red. Our material consists of plants preserved in formalin, collected by E. A. Brewer near Miami, April 17, 1902. Probably the same, although it is impossible to tell from material which has faded in drying, are speci- mens collected by J. H. Simpson in the Miami River, March 7, 1892 (no. 555). It is possible that some of the specimens listed under macrophylla belong here, but we are unable to tell from dried material. (FicureE 16, d.) 6. Nymphaea ozarkana Miller & Standley, sp. nov. Type in the U. S. National Herbarium, no. 615581, collected by Mr. Otto M. Smith in southern Missouri along White River, in August, 1910. The material was received fresh. Additional material is mounted on sheet no. 615582. DistRIBUTION: Ozark region of southern Missouri, probably also in northern Arkansas. DESCRIPTION. Rootstock slender; leaf blades apparently floating in most cases but sometimes erect; petioles te- rete, glabrous, 3 to 11 mm. in diam- eter; leaves orbicular to oblong, bright yellowish green, glabrous and smooth, 12 to 20 cm. long and 7 to 19 cm. wide, or even larger, those of average size measuring about 14 by 12 cm.,_ broadly rounded at the apex, the sinus about one-third the length of the blade, open, triangular, the lobes deltoid-orbicular or semiorbicular, rounded; peduncles stout, gla- brous; flowers depressed-globose, 30 mm. in diameter or less; sepals thin, glabrous, pale green, often yellowish toward the tips, the Fic, 18,—Leaf outline of Nymphaea ozarkana. Scale }, inner thinner and sulphur yellow, all red within or at least tinged with red; stamens in about 5 rows, the anthers longer than the filaments; stigma rays 10 to 12, broadly linear, distinct, extending to within 1.5 mm. of the edge of the disk; fruit subspherical, abruptly constricted above, 15 to 25mm, high and 14 to 20 mm. in diameter, smooth; disk 8 to 12 mm. in diameter, slightly depressed, the edges vertical; seeds few, 15 to 30, large, 5 mm. long and 3.5 mm. in the greatest diameter, ovoid, scarcely pointed, with a rather obtuse and con- spicuous raphe, pale brown, shining; body of fruit bright yellowish green, the stig- matic disk strongly tinged with orange red. (PiLate 36, G, facing p. 73. Ficures 13, a, 18.) 92 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Additional specimens examined: Missourt: Swan, Taney County, 1897, Trelease 14; Little Black River near Pleasant Grove, 1897, Mackenzie 370; Swan Creek near Swan, 1899, Bush 749; Shannon County, September 16, 1888, Bush; Ironton, Iron County, Septem- ber, 1897, Colton Russell; Current River, Carter County, 1897, Trelease 12; Mineral Point, Washington County, May 29, 1892, Eggert; Greene County, June 2, 1888, Blankinship. All of the specimens listed above, with the exception of the type, are in the herba- rium of the Missouri Botanical Garden. We have seen no material elsewhere, chiefly for the reason that so few plants from southern Missouri are to be found in eastern herbaria. The plant appears to be not uncommon in the Ozark region. All our specimens are Missourian, but the range of the species must extend south into Arkansas. We have seen the plant growing abundantly in the James River south of Springfield, and in the same stream farther south in Christian County. Mr. B. F. Bush has also written us concerning its occurrence in the region. Most of the specimens cited were labeled in the herbarium as N. hybrida, evidently because of the characteristic color of the fruit. There seems to be no very close relationship between the two species, that of the Ozarks being more closely connected with N. advena and N. americana. From the former it differs in the coloring of the flowers and fruit and in the shorter leaves more rounded and obtuse at the apex; from the second of these species it differs in the terete petioles and the open sinuses of the leaves. With both it disagrees decidedly in the small number of seeds, this being one of the most striking characteristics of our plant. 7. Nymphaea ludoviciana Miller & Standley, sp. nov. Type in the U. 8. National Herbarium, no. 441413; also material preserved in formalin, collected by R.S. Cocks, early in April, 1903, ‘‘in stagnant ponds, dug out in making the railroad track, two miles from a place known as North Shore on Lake Ponchartrain, about 27 miles from New Orleans,’’ Louisiana. In the National Herbarium other dried material of this collection is mounted on sheets 441410 to 441414 inclusive. Distrisution: Southern Louisiana, near the coast. DESCRIPTION. Leaf blades floating, oblong-ovate, somewhat narrowed towards the apex, rather thin, 29 to 38 cm. long and 20 to 27 cm. wide, widest about the middle; sinuses closed, 9 to 12 cm. deep; lobes rounded, slightly overlapping; blades glabrous throughout like the petioles, the lateral veins 23 to 25 on each side, about parallel for three-fifths their length, then branching dichotomously; petioles and peduncles subterete; flowers depressed-globose, 30 to 35 mm. in diameter, the sepals when spread measuring about 65 mm.; outer sepals oblong or obovate, 25 to 32 mm. long, rounded, somewhat narrowed to the base; inner sepals orbicular to obovate, narrowed at the base into a short claw, slightly longer than the outer ones and thinner; stamens in about 5 rows, the anthers 2 to 3 times as long as the filaments; no mature fruit with the type material but the immature capsules ovoid, 20 mm. high and 25 mm. in diameter, rather con- spicuously ribbed; stigmatic disk strongly depressed, oval or almost orbicular, 16 mm. in diameter; stigma rays 13 to 19, usually 17, extending to within 1.5 mm. of the edge of the disk, about 1.3 mm. wide, usually confluent at the base; edge of the disk shal- lowly crenate; color of sepals rather dark chrome yellow at the apex, becoming green below; petals deep chrome yellow growing paler towards the base; anthers maize yellow, the filaments pale cream; disk deep chrome with its rays light purplish brown. (Piate 41, B. Fraures 7, ¢, 19, 20.) EXPLANATION OF PLATE 41.—A, Fruit of Nymphaca microcarpa. B. Fruit and unopened flower of Nymphaea ludoviciana. Both natuyal size. Contr. Nat. Herb., Vol. 16. PLATE 41. A. FRUIT OF NYMPHAEA MICROCARPA MILLER & STANDLEY. B. FRUIT AND UNOPENED FLOWER OF NYMPHAEA LUDOVICIANA MILLER & STANDLEY. NORTH AMERICAN NYMPHAEA. 93 MILLER AND STANDLEY Mr. Andrew Allison, on June 28, 1904, collected in Lake Charles, Louisiana, mature fruit that undoubtedly belongs to this species; its description is as follows: Subglobose, abruptly constricted above, 35 to 40 mm. high and of the same diameter, smooth at the base, rather conspicuously ribbed above; edge of the disk 5 or 6 mm. high, diverging; disk usually oval, its center mostly umbo- nate, depressed 6 or 7 mm.; stigma rays linear, or widened at the base, usually slightly confluent at the base, with a distinct median groove; seeds about 4 mm, long and 3 mm. in diameter, the raphe only acutish and not conspicuous. Additional specimens examined: Dry— In bayou, Vermilion, Lafa- yette County, May 27, 1883, Langlois (Greene); near Lake Charles, July 10, 1893, Langlois (Greene); vicinity of Lake Charles, 1904, Andrew Allison 219 and 323. The collector writes further: ‘The plants can nearly always be found in the ponds or very slow-flowing streams of the pine barrens. The leaves are always floating on the surface with stems sometimes,a foot or two long. The plant can be found blooming from March to De- cember.”’ Not all the material from the vicinity of New Orleans belongs to this species, as noted elsewhere. In the National Herbarium there are two sheets of a Nymphaea collected at the south end of Long Pond, Lowndes County, Georgia, September 4, 1902, Harper 1611. These specimens represent a plant resembling N. ludoviciana, and possibly identical with it. The leaves are floating, the sinuses closed, and their outlines similar to those of this species. Unfortunately Mr. Harper was unable to secure fresh material. Fig. 19.—Leaf outline of Nymphaca ludoviciana. Scale }. Fic. 20.—Stigmatic pat- tern of Nymphaea lu- aevscana. Natural Nymphaea fluviatilis Harper, Bull. Torrey Club 83: 234, 1906. Type Locauiry: ‘In sloughs of the Canoochee River near Groveland, in the north- western corner of Bryan County,”’ Georgia. DistrIBUTION: Coastal plain of Georgia, perhaps also in northern Florida. 8. Nymphaea fluviatilis Harper. 94 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. DESCRIPTION. Floating leaves thin, glabrous, 18 to 25 cm. long and 16 to 22 cm. wide, almost orbicular in outline or broadly rounded-oblong, rounded and slightly emarginate at the apex; sinuses 5 to 8 cm. deep, narrow; lobes rounded; submersed leaves very thin and delicate, crisped, similar in outline to the floating ones, 9 to 19 cm. longand 10 to 12 em. wide; rootstock comparatively small and slender, about 3 cm. in diameter in the specimens examined, the leaf scars oval, about 10 mm. long and 7mm. wide; flowers depressed-globose, about 25mm. in diameter and 50 mm. when spread; sepals glabrous, very thin, the outer oblong, 16 to 18 mm. long and 15 or 16 mm. wide, obtuse, very slightly narrowed at the base; inner sepals obovate, slightly longer than the outer, rounded at the apex, narrowed at the base into a claw about 8 or 10 mm. wide and 6 or7 mm. long; stamens in usually 4 rows; Fic. 21.—Leaf outline of Nymphaea fluviatilis. Scale 3}. no mature fruit seen, but the im- mature ovoid, the stigma rays about 12, broadly linear, with a distinct median line, extending almost to the edge of the disk; center of the crater smooth. (ficurEs 10, b, 21, 22.) Specimens examined: In formalin— Grorata: Rather scanty material from the Canoochee River: near Groveland, June 22, 1903, Harper. Dry— GeoretA: In Canoochee River near Groveland, June 22, 1903, Harper 1849 (type collection); Savannah, a sheet inthe Herbarium of the Missouri Botanical Garden, labeled in Nuttall’s hand- writing, With regard to this plant Mr. Harper writes as follows: “Nymphaea fluviatilis seems to be quite common in creeks, small rivers, and the swamps of large rivers, but apparently never in ponds, in the coastal plain. I have seen it in the Savannah River swamp in the southeastern corner of Effingham County, in Rocky Comfort Creek near Louisville, in Buckhead Creek near Millen, in the Ogeechee River near Chal ker, Millen, Rocky Ford, Dover, and Meldrim, in the Canoochee at the type locality, in the Ohoopcee near Ohoopee and Reidsville, in the swamps of the Altamaha near Doctor- town and Barrington, in the Oconee swamps near Mount Vernon, in the little Oemul- gee near Lumber City, in Echeconnee Creek near its mouth (on the line between Bibb and Houston Counties), in the Ocmulgee River swamps near Abbeville, in the With- lacoochee near Nashville, and in the Flint River swamps in Crawford County near Everett.”’ FiG, 22.—Stigmatic pat- tern of Nymphaea fiu- viatilis. Natural size, 9. Nymphaea chartacea Miller & Standley, sp. nov. Type in the U. S. National Herbarium, no. 592491, collected at Mobile, Alabama, June 14, 1885, by Dr. Charles Mohr. DistriBution: Mississippi, Alabama, and western Florida, near the Gulf coast. MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 95 DESCRIPTION. Petioles and peduncles slender, glabrous; floating leaves glabrous, thin, ovate, 14 to 23 cm. long and 8 to 14 cm. wide, conspicuously narrowed at the apex, broadest at or just above the base; sinuses 3.5 to 6 cm. long, very narrow or closed, the lobes some- what unequally rounded; submersed leaves similar in outline, 15 to 23 cm. long, rounded at the apex, crisped, very thin and delicate; flowers 30 to 35 mm. in diameter; sepals 6, 12 to 20 mm. long, oblong, obtuse; fruit not seen. (Ficures 13, ¢, 23.) In Mohr’s Plant Life of Alabama this is listed as Nymphaea sagittifolia. The distri- bution of this species in Alabama is given as the “Central Pine belt to Coast plain. Still-flowing water. Most frequent in the coast region. Tuscaloosa County (E. A. Smith). Montgomery, Mobile, and Baldwin Counties. Flowers lemon yellow. June, July; not rare.”’ Just how many of the plants thus listed belong to our new species it isimpossible to tell. Probably most of them belong here. At any rate, none of the plants thus referred to are N. sagittifolia. We are re- luctant to describe a Nymphaea from dried material, but in this case there can be no doubt about the dis- tinctness of the plant. The leaves, while resembling those of N. ulvacea and N. sagittifolia in texture, are very different in outline. From our studies in this genus we may confidently expect that when fresh ma- terial of this plant is secured it will present other marks of distinction. We have seen the following additional dried mate- rial that seems to belong here. The leaves of the Florida specimens are somewhat more obtuse than those from farther west, but they are equally thin; the flowers have the very thin sepals so characteristic of this and Nymphaea fluviatilis. Additional specimens examined: Mississippi: Biloxi, March 26, 1898, Tracy 5012; Ocean Springs, April 5, 1889, Tracy; ponds and bayous near the coast, May, 1859, Hilgard; Wells Ferry, April 22, 1895, Skehan. Fioripa: Without locality, Chapman Herbarium; Apalachicola, December 6, 1898, TJrelease. FIG. 23.—Leaf outline of Nymphaea chartacea. Scale }. 10. Nymphaea sagittifolia Walt. Nymphaea sagittifolia Walt. Fl. Carol. 155, 1798. Nymphaea longifolia Michx. Fl. Bor. Amer. 1: 312. 1803. Nymphaea sagittata Pers. Syn. Pl. 2: 63. 1807. Juphar sagittaefolium Pursh, Fl. Amer. Sept. 2: 370. 1814. Nuphar longifolium Smith; Rees’s Cycl. no. 5. 1819. Ropalon sagittatum Raf. New Fl. N. Amer. 2:17. 1837. Nymphaea hastata Steud. Nom. Bot. ed. 2. 2: 199. 1841. Nymphaea sagittaefolia Britt. & Brown, Illustr. Fl. 2: 43. 1897; Small, FI. Southeast. U. 8. 456. 1903, in part; Britton, Man. ed. 2. 407, 1905, in part; Rob- ins. & Fern. in A. Gray, Man. ed. 7. 391. 1908, in part. Type Locatity: South Carolina. DistrisuTion: Eastern North and South Carolina, 96 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. DESCRIPTION. Floating leaves rather thick and leathery, narrowly oblong or usually narrowly oblong-lanceolate, 14 to 28 em. long and 5 to 10 cm. wide, not conspicuously narrowed at the apex, rounded; sinus 30 to 35 mm. deep, open, V-shaped; submersed leaves well developed, similar in outline to the floating ones but larger, sometimes 36 cm. long and only 7 cm. wide, very thin and delicate, crispate, glabrous like the floating ones; petioles slender, glabrous, cylindrical, 8 or 9 mm. in diameter; rootstocks 20 to 25 mm. in diameter, densely leafy; leaf scars narrowly oval to semioval, 6 to 8 mm. long; flowers depressed-globose, 26 to 29 mm. in diameter and 19 to 22 mm. high; sepals 6, glabrous, the outer ones 21 to 26 mm. long and 17 to 20 mm. wide, oblong, slightly narrowed towards the base, thin; the inner sepals of about the same length, orbicular, thin; stamens in 5 or 6 rows, the anthers scarcely if at all longer than the filaments; color of sepals canary yellow, tipped with green; petals and stamens yellow but paler than the sepals; fruit ovoid, 31 to 34: mm, high and 23 to 28 mm. in diameter, considera- bly constricted above, smooth or almost so below, conspicuously ribbed above; edges of the disk raised 4 or 5 mm., vertical or slightly spreading, orbicular in outline, almost or quite entire, 15 to 17 mm. in diameter, its center depressed about 3 mm., smooth, 3 to 6 mm. in diameter; stigma rays 10 to 14, linear, 4 mm. long, extending to within 2.5 mm. of the edge of the disk, distinct, with no trace of a median groove; color of fruit shining apple green, a little darker towards the top, occasionally darkening to oil green throughout; seeds ovoid, pointed but not sharply so, 4 to 5 mm. long and 3 mm. in diameter, the raphe rather obtuse and not con- spicuous. (PuLates 42, A, facing p. 96. 44, A, facing p.99. Fia- Fig. 24.—Leaf out- URES 7, f, 24, 20.) line of Nymphaea sagittifolia. Scale 1 EXPLANATION OF PLATE 42.—Seeds of Nymphaea spp. (A) Nymphaea sagittifolia, (B) N. ulvacea, (C) N. ovata, (D) N. puberula, (E) N. orbiculata, (F) N. bombycina, ‘ (G) N, polysepala. Al] natural size. Specimens examined: In formalin— Nortu CARoLina: Fayetteville, 1902, Boynton. Dried— Norru Carouina: Fayetteville, 1902, Boynton; Wilmington, 1880, Dr. Thos. F. Wood; without locality, 1884, McCarthy; without local- ity, 1885, McCarthy 9; Northeast River, 1879, [Zyams (N. Y.); Cape Fear River, Wilmington, 1855, Z/examer & Maier (N. Y.); Fayetteville, 1904, Biltmore [Herba- rium 9657d; Wilmington, C. S. Williamson (N. Y.); near Fayetteville in Little River, Major Le Conte (Mo.); in the fresh water of Cape Fear River and its bayous under the influence of the tides, 1867, Canby (Mo.); Fis. 25.—Stigmatie Wilmington, Curtiss (Mo.). pattern of Nymphaea . + sagittifolia. Natural SoutH Carona: Georgetown, 1857, LZ. R. Gibbes (N. Y.). size. In Mohr’s Plant Life of Alabama this species is reported as occurring within that State. The specimens thus referred belong, for the greater part at least, to Nymphaea chartacea. In Gray’s Manual it is reported as occurring in southern Indianaand Illinois; but the ponds in which the plant grew are now drained, and we have been unable to procure fresh material from the region. It is exceedingly improbable that the species is found outside the States of North and South Carolina, Contr. Nat. Herb., Vol. 16. PLATE 42. SEEDS OF SEVERAL SPECIES OF NYMPHAEA Contr. Nat, Herb., Vol. 16. PLATE 43. A. FRUIT AND UNOPENED FLOWER OF NYMPHAEA ULVACEA MILLER & STANDLEY. B. Fruits oF NYMPHAEA OVATA MILLER & STANDLEY. MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA, 97 We have seen no specimens from other States nor have we any information that clearly indicates the plant’s occurrence elsewhere. 11. Nymphaea ulvacea Miller & Standley, sp. nov. Type in the U. 8. National Herbarium, no. 357366, collected in the Blackwater River near Milton, Florida, May 14, 1898, by A. H. Curtiss (no. 6409). DisrrisutTion: Extreme western Florida. DESCRIPTION. Petioles slender, terete, smooth, glabrous, 7 mm. in diameter, 45 to 70 cm. long; rootstocks stout, 2 to 5 cm. in diameter, oval in cross-section, the leaf scars 11 to 13 mm. long and 8 or 9mm. wide; floating leaves lanceolate or lance-ovate, 115 to 165 mm. long and 54 to 66 mm. wide, very much narrowed at the blunt apex, glabrous; sinus closed, 20 to 28 mm. deep, the lobes over- lapping conspicuously; submersed leaves 23 to 28 cm. long and 7 to 10 cm. wide, very thin and delicate, the margins notably plicate; flowers 20 to 23 mm, in diameter, 15 to 18 mm. high, depressed-globose; sepals 6, when spread measur- ing 45 to 65 mm.; outer sepals 24 to 25mm. long and 14 to 18 mm. wide, broadly oblong, rounded, slightly narrowed at the base; inner sepals rounded-obovate, thinner, narrowed at the base into a claw 6 mm. long and 7 mm. wide; stamens in 4 or 5 rows, the anthers slightly longer than the filaments; fruit subglobose, abruptly contracted above, prominently ribbed almost to the base, 15 to 21 mm. high and 16 to 24mm. in diameter; rim of the disk 1.5 or 2mm. high; disk orbicular, 11 to 18 mm, in diameter, its edges vertical, the center de- pressed about 3mm. and smooth; stigma rays 11 to 16, usually 12 to 14, elliptical, 5 mm. long and 1.5 mm. wide, acutish at both ends, distinct, with a very faint median groove or smooth; seeds 3.5 to 4 mm. long, 2.5 mm. in diameter, pointed, with an acute and conspicuous raphe. (PLATES 42, B, facing p. 96; 43, A. Ficures 7, d, 26, 27.) EXPLANATION OF PLATE 43.—A. Fruit and unopened flower of Nymphaca ulvacea. B. Fruit and flower of Nymphaea ovata. Both natural size. Fic. 26.—Leaf outline of sys * “« Nymphaea ulvacea, Additional specimens examined: scale }. In formalin— Fiori: Milligan, September 24, 1901, Curtiss. Dried— Froripa: Milligan, May 14, 1898, Curtiss 6409; Blackwater River, Santa Rosa County, Curtiss 104. Although related to Nymphaea sagittifolia the Florida plant may be distinguished by its much wider, more pointed leaves, its smaller fruit, and its elliptical stigma rays. The submersed leaves when dried strongly suggest specimens of certain alge, especially the genus Ulva, hence the specific name. 12. Nymphaea ovata Miller & Standley, sp. nov. mG. 27 -— Stigmatic Date Type in the U. S. National Herbarium, collected at San 0) ympnaead ulvacea. ‘ NN atineal Marcos, Texas, August 6, 1901, by Mr. J. L. Leary. Preserved size. in formalin. The description was based principally upon the fresh material of this collection. Distripution: Central-eastern Texas. 98 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. DESCRIPTION. Petioles stout, terete, 13 to 15 mm. in diameter, smooth, more or less silky-pubescent; leaf blades ovate, 22 to 35 em. long and 16 to 28cm. wide, broadest just above the base, thick, smooth and shining dark green above, densely silky-pubescent beneath, the principal lateral veins about 24 on each side, distinct and parallel for two-thirds the Fic. 28,—Leaf outline of Nymphaea ovata. Seale }. distance to the margin; sinuses open but very narrow, 7 to 9cm. long; flowers depressed-globose, 30 to 40 mm. in diameter, 18 to 23 cm. high; sepals 6, when spread measuring 65 to 82 mm.; the outer ones 24 to 30mm, long, 20 to 25 mm. wide, oblong to suborbicular, often asymmetri- cal, silky-pubescent on the outer surface, strongly convex, green except sometimes at the tips, there yellowish; inner sepals 28 to 32 mm. long and about as broad, broadly deltoid-obovate, truncate or emarginate, glabrous or sparingly pubescent near the middle of the base, narrowed at the base to a short, broad claw, bright chrome yellow throughout or greenish toward the base; stamens in 5 or 6 rows, the anthers twice as long as the fila- ments; fruit globose-ovoid, slightly constricted above, 30 to 35 mm. high, 30 to 40 mm. in diameter, rather inconspicu- ously ribbed below, prominently 80 above, green becoming yellowish above; rim of disk vertical, 5 to 7 mm. high; disk depressed 5 to 8 mm., 22 to 24 mm. in diameter, almost orbicular, smooth in the center, chrome yellow; rays buff, 13 to 20, usually 15 to 17, 5 to 8 mm. long, extending to within 2 to 4 mm. of the edge of the disk, linear-lanceolate, 1.3 mm. wide at the base, narrower and acutish at the apex, usually confluent at the base, with a strong median groove; seeds 3.5 to 4 mm. long, 3 mm, in diameter, ovoid, pointed, with a prominent acute raphe. (Puates 42, C, facing p. 96; 43, B, facing p. 97. Fiaures 7, a, 28, 29.) Additional specimens examined: Dry— Texas: New Braunfels, June !7, 1906, Otto Locke; same locality, June 15, 1910, Otto Locke: San Marcos, No- vember 6, 1897, Trelease. This can not be confounded with any other species, Although its leaves are pubescent beneath they are very dif- Fig. 29.—Stigmatie pat- tern of Nymphaeaovata. Natural size, erent in outline from those of the other species whose leaves are pubescent. Material collected by Mr. Andrew Allison at Slidell, Louisiana, in July, 1904, resembles this very closely; unfortunately it consists of leaves only, One sheet in the Contr. Nat. Herb., Vol. 16. Plate 44. A. FLOWER AND FRUIT OF NYMPHAEA SAGITTIFOLIA WALT. ‘ | B. FRUIT OF NYMPHAEA PUBERULA MILLER & STANDLEY. MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 99 Mohr Herbarium belongs here so far as one can tell from dried material; it was collected by Doctor Mohrat Mobile, Alabama. The one leaf is 19cm. long and densely pubescent beneath. A specimen in the herbarium of the Missouri Botanical Garden, coilected at San Antonio, Texas, by Gustav Jermy, may also represent this plant. 13. Nymphaea puberula Miller & Standley, sp. nov. Type in the U. 8. National Herbarium, collected in Brays Bayou, about 4 miles south of Houston, Texas, September 6, 1901, by TH. P. Attwater. Preserved in formalin. Described from the fresh ma- terial of this collection. DistrisutTion: Near Houston, Texas. DESCRIPTION. Leaves floating, rather thin, orbicular- ovate, rounded at the apex, widest about the middle, 17 to 25 cm. long and 16 to 21 cm. wide, the sinus about 6 cm. deep, open but very narrow; blades smooth above, rather sparsely puberulent beneath; peti- oles and peduncles terete, almost glabrous; lateral nerves of the blades 14 to 16, par- allel and unbranched for one-half or two- thirds their length; flowers depressed- globose, 26 to 33 mm. In diameter, the Fia. 30.—Leaf outline of Nymphaea pubcrula, perianth when spread measuring about 60 Seale 3. mm.; outer sepals oblong, subtruncate at the apex, slightly narrowed towards the base, 20 to 26 mm. wide; inner sepals broader, broadly deltoid-obovate to almost orbicular, retuse, narrowed abruptly at the base into a claw 5 mm. long and 6 mm. wide; stamens in 5 or 6 rows, linear- cuneate, the anthers one and one-half times as long as the filaments; sepals in color canary yellow, the disk of the ovary slightly and the stamens decidedly paler, no trace of red anywhere present in the flowers; fruit broadly ovoid, only slightly con- stricted above, 31 to 38 mm. high, 32 to 40 mm. in diameter, the edges of the disk raised asa rim 3 to 5mm. high; stigmatic crater conspicuously depressed; stigmatic rays 9 to 20, usually 13 to 15, 4 to 5 mm. Jong, linear, with no trace of a median groove, extending to within 1.5 or 2 mm. of the edge of the orbicular disk; center of disk umbonate; fruit faintly ribbed above, almost smooth below; seeds 4 to 5 mm. long and 2.5 or 3 mm. in diameter, conspicuously pointed, with a sharply acute raphe; color of fruit pepper green, the disk yellowish. (Piares 42, D, facing p. 96; 44, Fia.31.—Stigmaticpatten B. Fiaures 10, a, 30, 31.) of Nymphaea puberula. . ' EXPLANATION OF PLATE 44.—A. Flower and fruit of Nymphaca sagitti- Natural size. folia. B. Fruit of Nymphaea pubderula. Both natural size. Here belongs a sheet in the herbarium of the Missouri Botanical Garden, collected in running water near Houston by Lindheimer. The plant suggests Nymphaea fluviatilis but the pubescent leaves distinguish it at once. Its pubescence throws it into the group with N. orbiculata, N. bombycina, and N. ovata, but the leaves are very different in outline and their pubescence is much more sparse, 100 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Concerning the locality where the specimens were gathered Mr. Attwater, the col- lector, says: ‘‘All of the specimens are from Brays Bayou, about 4 miles south of Houston. This bayou is like many others in the Gulf Region; indry weather the water stands in pools or holes. Its sides are lined with trees, bushes, and tangled weeds, sv that the water is kept shaded all day long. The particular spot where these lilies were taken would get only a few gleams of sunshine during the day, but I presume the plants grow in more open water where they are not shaded all day long. They were growing in water about knee-deep.”’ 14. Nymphaea microcarpa Miller & Standley, sp. nov. Type in the U. 8S. National Herbarium, collected in the San Antonio River near San Antonio, Texas, March 26, 1910, by Mr. Bernard Mackensen. Preserved in for- malin. Described from the fresh material of the type collection. ~ Distrisution.—Vicinity of San Antonio, Texas. DESCRIPTION. Leaves broadly rounded-ovate to orbicu- lar in outline, widest near the base, 12 to 25 cm. long and 16 to 24 cm. wide, rounded at the apex, emarginate, dull green and gla- brous above, rather densely silky-pubescent beneath; sinus broadly V-shaped; lobes rounded; petioles terete, silky-pubescent throughout; sepals thin, the outer orbicular to oblong, the inner obovate, rounded, truncate, or slightly emarginate at the apex, the outer oil green, yellowish at the apex, the inner sulphur yellow, the flowers with no tinge of red; stamens in 5 or 6 rows, the anthers slightly longer than the filaments; fruit subglobose to almost cylin- drical, 20 to 26 mm. high and 18 to 24 mm. in diameter, slightly constricted above, deeply ribbed, almost lobed, from base to top; rim of the disk somewhat spreading or erect, 1.5 to 2.5 mm. high; disk orbicular, 13 to 20 mm, in diameter, entire, slightly depressed (1 to4 mm.), the center umbonate, 3 or 4 mm. in diameter; rays 9 to 15, mostly 12, linear or slightly widened at the base, distinct, with an evident median groove, extend- ing to within 1.5 or 2 mm. of the edge of the disk; body of the fruit clear green, the edge of the disk faintly tinged with red; seeds 3.5 mm. long and 2.5 mm. in diameter, shining dark brown, ovoid, acutish. (Piare 41, A, facing p. 92. Fiaures 4, a, 32, 33.) Fig. 33.—Stigmatie pattern of Nymphaea microcarpa. Natural size. Fia. 32.—Leaf outline of Nymphaea microcarpa, Scale }. We first became acquainted with this plant through rather incomplete material collected by Mr. H. P. Attwater, October 18, 1902, in the Medina River, about 15 miles southwest of San Antonio, Texas. This consisted of leaves and fruit without flowers. Later we were fortunate enough to secure specimens from Mr. Bernard Mackensen, of San Antonio, This last sending enables us to complete our diagnosis and to be more certain of the distinctness of the species. The same collector has forwarded less complete material than the type collected in the Cibolo River at Sutherland Springs, 30 miles east of San Antonio, March 27, 1910. He states that the leaves are usually floating, rarely erect. Nymphaea microcarpa is most closely related, perhaps, to N. ovata of San Marcos, Texas, a locality not fardistant. The outline of the leaves, however, is very different, Contr. Nat. Herb., Vol. 16, PLATE 45. A. FLOWER AND FRUIT OF NYMPHAEA ORBICULATA SMALL. B. FLOWER AND FRUIT OF NYMPHAEA BOMBYCINA MILLER & STANDLEY. MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 101 and the plants lack the peculiar yellowish cast exhibited by the San Marcos specimens. The fruit, too, is much smaller, and the number of stigmatic rays decidedly less. 15. Nymphaea orbiculata Small. Nymphaea orbiculata Small, Bull. Torrey Club 23: 128. 1896. Type Locauity: ‘In a small lake near Thomasville, Thomas County, southern Georgia.”” Type collected by J. K. Small. DisTRIBUTION: Known only from Thomas County, Georgia. DESCRIPTION. Leaves floating, orbicular in outline but broadest near the base, thick and leathery, glabrous above but silky- pubescent beneath, 30 to 50 cm. in diameter, emarginate, the closed sinus 6.5 to 14 cm. deep, the broadly rounded lobes overlapping; veins par- allel almost to the margin; petioles and peduncles cyl- indrical, stout, pubescent throughout; flowers depressed - globose, about 55 mm. in diameterand 30mm. high, the perianth when spread meas- uring about 100 mm.; sepals normally 6, the 3 outer elliptical-oblong, about 45 mm. long and 30 mm. wide, gla- brous throughout, the three inner suborbicular, about 40 mm. in diameter, their bases abruptly narrowed toa claw 10 mm. wide, truncate or retuse at the apex; petals 22 to 24, varying from linear to oblong or oblong-obovate, 9.5 mm. long and 2.5 to 6 mm. wide, retuse, truncate, or rounded at the apex, with a gland- ular spot 2 mm. in diameter usually present on the outer side near the tip; stamens usually in 8 rows, about 30 to the row; anthers slightly but distinctly longer than the filaments; color of flowers not accurately known, but specimens after a few weeks’ immersion in formalin showing a pattern in no way peculiar and proving the entire absence of red; fruit sub- globose, about 45 mm. in diameter, distinctly marked with fine ribs above, smooth about the base; stigmatic crater oo about 28 mm. wide and 5 mm. deep, its outer wall usually Fi. 35.—Stigmatic pattern Vertical; center of disk smooth; stigma rays distinct, varying of Nymphaea orbiculata. , Natural size. in number from 12 to 22, but usually 15, 16, or 17, when fully developed 4 to 6 mm. in length and about 0.5 mm. wide, extending to within 1.5 mm. of the edge of the disk; surface of rays with barely indicated median line; seeds ovoid, about 4.5 mm. long and 3 mm. in diameter, with a distinct raphe. (PuarTe 42, E, facing p. 96; 45, A. Ficures 4, }, 34, 35.) EXPLANATION OF PLATE 45.—A. Flower and fruit of Nymphaea orbiculata. B. Flower and fruit of Nymphaea bombycina. Both natural size. Fia. 34.—Leaf outline of Nymphaea orbiculata. Scale 4. 102 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Specimens examined: In formalin— Georeia: Heards Pond, 1901, Harper. Dry— Grorata: Along the Ochlockonee River, near Thomasville, 1895, Small, type (N. Y.); Heards Pond, Thomas County, 1901, Harper 1178. 16. Nymphaea bombycina Miller & Standley, sp. nov. Type in the U. S. National Herbarium, no. 394797, collected at Wellborn, Florida, October 11, 1901, by A. H. Cur- tiss (no. 6900). Besides the type sheet another in the National Herbarium, no. 394798, is of this collection. DistriBpution: Northern Flor- ida and southern Georgia. DESCRIPTION. Rootstocks stout, very large and fleshy, rough, 7 or 8 cm. in diameter; leaf blades floating, glabrous on the upper surface, densely silky-pubescent be- neath, thick and leathery, very broadly ovate to almost orbicu- lar in outline, 20 to 30 em. long and 20 to 25 em. wide, broadest near the base, rounded at the apex, and emarginate; sinus open, V-shaped or U-shaped, two- fifths as long as the blade, the lobes broadly rougded; veins conspicuous, parallel almost to the margin, the midrib stout and very prominent; petioles stout, terete, about 10 mm. in diameter, silky-pubescent: peduncles stout, densely pubescent, 10 to 15 mm. in diameter; flowers large, depressed-globose, much flattened, 60 to 80 mm. in diameter, 35 to 45 mm. high, when spread measuring 12 to 14 cm.; sepals very thick and fleshy, the outer oblong, obtuse, the inner obovate, broadly rounded; outersepalsoil green, yellowish at the tips, the inner almost entirely sulphur yellow, nored present anywhere in the flower; stamens in 6 or 7 rows, the anthers about twice as long as the filaments; fruit ovoid or almost cylindrical, 50 to 60 mm. in diameter, smooth below, faintly and finely ribbed above, slightly constricted just below the spreading edges of the stigmatic disk, this 30 to 35 mm. in diameter, orbicular, the margin undulate; rays 15 to 18, usually 16, linear, distinct and distant, extending to within 2 mm. of the edge of the disk, with no trace of a median groove; central area about 12 mm. in diameter, smooth, depressed about 5 mm.; seeds ovoid, light chestnut brown, 6 mm. long and 4.5 mm. in diameter. the raphe acutish; color of fruit much as in N. orbiculata, the body oil Fic. 36.—Leaf outline of Nymphaea bombycina. Scale 3. Fig. 37.—Stigmatie pattern of Nym- phaea bombycina, Natural size. PLATE 46. rb Contr. Nat. He "OGVHONND ‘SLING YHSLVUD LV ‘SNSSYD (‘WIS9NQ) VIVvVdaSAI10d VAVHdWAN MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 108 green or lighter, the disk yellowish. (PLatres 42, F, facing p. 96; 45, B, facing p. 101. Fiaures 7, ¢, 36, 37.) Additional specimens examined: Dry— Grorata: Ii shallow water (30 to 60 cm. deep) of Ocean Pond, Lowndes County, September 4, 1902, R. M. Harper 1610. With Nymphaea orbiculata, the only closely related species, this forms a group sharply differentiated from all other North American Nym- phaeas. The plants are marked by their large, almost orbicular, silky-pubescent, thick and almost leathery leaves, and large flowers and fruit. This species may be separated from its near relative by the open sinus of the leaves, larger flowers and fruit, and larger seeds. In addition the stamens are in decidedly fewer rows. 17. Nymphaea polysepala (Engelm.) Greene. Nuphar polysepalum Engel. Trans. Acad. St. Louis. 2: 283. 1865. Nuphar polysepalum pictum Engelm. loc. cit. 285. 1865. Nymphaea polysepala Greene, Bull. Torrey Club 15: 84, 1888. Type Locauity: “In Osborn’s Lake, in the same region, at an sine terion”? de" Gia ‘s Lake Fig, 38.—Leaf outline of ane polysepala, typical form. near Long’s Peak,’’ Colorado. The type was collected by Dr. C. C. Parry. DistripuTion: Alaska south through northern California and in western Nevada, and southeastward through northern Idaho and western Montana to western South Dakota and Colorado. DESCRIPTION. Leaves usually floating, rarely held above the water, the lobes, however, often elevated, oblong or ovate, rather thick, dull green, glabrous, 21 to 40 cm. long and 16 to 26 cm. wide, sometimes slightly narrowed at the apex but usually rather broadly rounded, widest just above the base; sinus usually open and broadly V-shaped, sometimes closed, 7 to 10 cm. deep, the lobes rounded or acutish; petiolesand pedunclesglabrous, terete, 8to16mm. in diameter; submersed leaves lacking; flowers 55 to 70 mm. in diameter and 40 mm. high, depressed-globose, the perianth when spread measuring 110 to 140 mm.; sepals usually 9, sometimes 7 to 12, the outer ones small, rounded-oblong, 30 to 50 mm. long, the inner orbicular, 40 to 55 mm. long and 35 to 22075°—12——4 Fic. 39.—Stigmatic pattern of Nymphaea polysepala. Nat- ural size. 104 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 50mm. wide, truncate or retuse at the apex, gradually narrowed to the base or abruptly narrowed into a short claw; petals very thick, oblong, equaling the stamens; these in 5 to 7 rows, their anthers slightly exceeding the filaments; capsules globose-ovoid to almost cylindrical, 45 to 90 mm. high and 35 to 60 mm. in diameter, abruptly con- stricted above, rather conspicuously ribbed throughout; stigma rays 14 to 30, usually 20 to 25; crater suborbicular, 30 to 35 mm. in diameter, usually about 20 mm. deep, its margin almost entire, sometimes slightly undulate; rays linear, with no trace of a median groove, distinct, 14 to 18 mm. long, extending to within 1 mm. of the edge of the disk; center of the crater smooth; outer sepals oil green, yellowish at the tips; inner sepals chrome yellow, the tips usually shaded with green; petals greenish yellow, their inner surfaces tinged with purplish brown except at the base and tip; filaments light greenish yellow, the anthers dark prune purple; pollen straw yellow; fruit varying in color from light apple green to citron yellow; seeds 3.5 to 4 mm. long, oblong, dull brown, shining. (Piares 42, G, facing p. 96; 46, 47. Ficures 16, ¢, 38-40. ) EXPLANATION OF PLATES 46, 47.—Pl. 46, Nymphaca polysepala at Crater Butte, Colorado; photographed by Mr. E. R. Warren. PI. 47, fruit of Nymphaea polysepala (typical). Natural size. Specimens examined: In formalin— AtasKa: Iliamna, 1902, W. 1. Osgood; Kodiak, 1901, W. J. Fisher. WASHINGTON: Tacoma, 1902, Flett; same locality, 1901, Fleit. Ipano: Henrys Lake, Lake P. O., 1901, J. Sherwood. Wyominac: Beaver Lake, Yellowstone National Park, 1902, Mearns. Sours Daxkora: Spearfish, 1901, D. C. Booth. CoLorapbo: Near Boulder, 1902, R. T. Young; near Leadville, 1901, Tulian. CatirorniA: Near Albion, Mendocino County, 1901, James Mc Murphy; Stock- ton, 1902, LZ. Belding. Dry— AuasKA: Yakutat Bay, 1899, Trelease & Saunders 3758, 3758a, 3759, 3760 (Mo.); Yakutat, 1904, Piper 4362; Seldovia, 1904, Piper 4346; vicinity of Yakutat Bay, Khantaak Island, 1892, Funston 43; Copper River Region, 1902, Wm. L. Poto 114; Kodiak, 1900, Walpole 1173; Sitka, 1900, Walpole 1132; Camp Retreat, 1886, H. G. U. Stoney; Yakutat Mission, 1899, Trelease 3759; near Orca, Prince William Sound, 1899, Coville & Kearney 1317; Kodiak, 1899, Coville & Kearney 2323; Kodiak, 1897, Evans 394; Short Bay, 1895, Gorman 107; Kodiak, 1874, U. S. Coast Survey (Gray); in small mountain ponds near Yes Bay, 1895, T. Howell (Greene). British Cotumsra: Vancouver Island, 1893, Macoun (Greene); Chilliwick Val- ley, 1901, Macoun 33763; Revelstoke, 1890, Macoun; Victoria, Vancouver Island, J. R. Anderson; swamp by Goldstream, Selkirks, 1903, C. H. Shaw 936 (Phila.); San Juan Lake and River, Vancouver Island, 1902, Rosendahl 893 (N. Y.); Fort Rupert, Vancouver Island, 1904, Geo. Hunt (N. Y.); Lower Frazer River, 1859, Lyall (Gray); near Victoria, 1885, Fletcher (Gray). WasHINGTON: Falcon Valley, 1882, Suksdorf; Union Flat, Whitman County, 1892, Hull 421 (N, Y.); Chehalis County 1897, Lamb 1260 (N. Y.); Seattle, 1892, Mosier; Big Meadows, 8 miles west of Ione, 1902, Kreager 426; Spokane River, Wilkes Exploring Expedition 546; Lake Union, King County, 1898, Savage, Cameron & Lencoker (Mo.). Oregon: Sauvies Island, 1886, JT. Howell (C.); Forest Grove, Washington County, 1894, F. E. Lloyd (N. Y.); without locality, 1871, Hall (Gray); Blue Mountains, 1888, Cusick 1525; Buck Lake, Klamath County, 1897, Coville & Applegate 48; in a slough near Todds, Forest Grove, 1902, 7. E. Contr, Nat. Herb., Vol. 16. PLATE 47. FRUIT OF NYMPHAEA POLYSEPALA (ENGELM.) GREENE. MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA, 105 Specimens examined—Continued. Dry—Continued. Wilcor 1; Klamath Agency, 1902, Walpole 2221; ponds, Salem, 1871, Hall; Mount Hood, 1898, Applegate 2844; vicinity of Portland, 1905, Wm. Palmer 1474; ponds near Portland, 1900, Henderson 44 (Mo.); near Coos Bay, 1880, George Engelmann (Mo.). Montana: Spanish Basin, Gallatin County, 1898, Blankinship; Big Fork, 1901, MacDougal 566 (N. Y.); Forks of the Madison, 1897, Rydberg & Bessey 4058; Spanish Basin, 1897, Rydberg & Bessey 4059 (N. Y.); Spanish Creek, Galla- tin County, 1901, /. Vogel (Gray); Lost Horse Pass, Bitter Root Forest Reserve, 1897, Leiberg 2982; Madison Valley, Robert Adams; no locality, E. V. Wilcox 326. Wyromina: Norris Geyser Basin, Yellowstone National Park, 1899, Blankin- ship (Mo.); Shell Creek, Big Horn Mountains, 1899, Tweedy 2284 (N.Y.); shallow water near Leighs Lake, 1901, Merrill d& Wilcor 1116; Beaver Lake, Yellowstone Na- tional Park, 1902, Mearns 2626; in a small lake, Norris Geyser Basin, 1899, A. & E. Nelson 6152; Yel- lowstone National Park, 1902, Mearns 2526, 2624; head of Green River, 1894, A. Nelson 898; pond one mile east of the Falls of the Firehole River, Yellowstone Na- tional Park, 1887, Ward. Ipano: Valley of Lake Tesemini, Koo- tenai County, 1892, Sandberg, Mac- Dougal & Heller 701; Lake Coeur d’Alene, 1892, Ation 71 (N. Y.); Farmington Landing, Lake Coeur d’Alene, 1892, Sandberg, Mac- Dougal & Heller 621; in the South Fork of the Coeur d’Alene River, near Old Mission, 1895, Leiberg 1412; Clarks Fork, 1889, Greene (Greene); North Fork of Columbia River near Lake Pend Oreille, Fic. 40.—Leaf outline of Nymphaea polysepala, Leiberg. southern form. Scale 4. Uran: Uintas, northern Utah, 1869, Watson. CoLorapo: Small lake in Animas Valley, 1875, T. S. Brandegee 1166 (Phila.); Osborne Lake, 1864, Parry (Gray, Mo.). Nevapa: Marlette Lake, Washoe County, 1902, Baker 1479. CaLirorNiA: Stockton, 1902, L. Belding; 10 to 15 miles west from Amedee, alt. 1,200 meters, 1897, Jones; Susie Lake Trail, Lake Tahoe Region, 1909, McGregor 179; without locality, Bridges 13a (C.); Lagunitas, Marin County, 1877, Edwards (N. Y.); Pudding Creek, Mendocino County, 1894, Eastwood (Gray); Plumas County, Mrs. Ames (Gray); Blue Lakes, Lake County, 1893, Blankinship (Gray); Stockton, 1890, Greene (Greene); Stockton, 1890, J. A. Sanford (Greene); Siskiyou County, 1876, Greene (Greene); Eureka, Hum- boldt County, 1890, Greene (Greene); Big Lagoon, Big River, 1903, McMur- phy 194; without locality, 1875, G. R. Vasey; near Lassen Buttes, Plumas County, 1897, Brown 656; without locality, Hartweg 1637 (Gray); sloughs 106 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Specimens examined—Continued, Dry—Continued. near Stockton, 1896, Jepson (Gray); Mad River near Vances, Humboldt Bay, 1901, Chandler 1236; Olema Lake, Marin County, 1862, State Survey 1481; Truckee, 1893, Miss Michener; Long Valley, 1866, Bolander (Mo.). The species seems to be fairly constant in all its essential characters. As in other species, the leaves are somewhat variable but within definite limits. Figure 38 rep- resents the typical leaf form exhibited by the greater number of specimens. Figure 39 shows a leaf outline characteristic of the southern representatives of the species. There seem to be no concomitant characters which, combined with the difference in leaf form, might separate the northern and southern plants. BIBLIOGRAPHY. The following list includes most of the titles dealing with the genus Nymphaea in North America, No attempt has been made to include mere isolated data regarding local distribution. The articles listed are only those that have been referred to in the prepara- tion of the present paper. Aiton, Wiuuiam. Hortus Kewensis 2: 226-228. 1789. Contains the original description of Nymphaea advena, Aiton, Wmuam T. Hortus Kewensis ed. 2. 3: 295, 1811. Makes the new combinations Nuphar advena and N, kalmianum. Bailey, L. H. Cyclopedia of American Horticulture 3: 1096-97. 1901. Enumerates 7 species from North America, Brewer, W. I1., Watson, Sereno, and Gray, Asa. Botany [of California] 1: 17. 1876, Description of Nymphaea polysepala with notes upon distribution. Britten, James. The nomenclature of Nymphaea, Journ. Bot. Brit. & For. 26: 6-11. 1888. A discussion of the application of generic names. Britton, N. L., and Brown, Addison. Illustrated Flora 2: 42-3. 1897, Descriptions and illustrations of 4 species. Bubani, P. Flora pyrenaea 3: 259-261. 1901. Publication of the generic name Nymphona. ‘Caspary, R. Nymphaeaceae. In Eng]. & Prantl, Nat. Pflanzenfam. 3°: 1-10. 1888. Chesnut, V. K. Plants used by the Indians of Mendocino County, California. Contr. Nat. Herb. 7: 295-408. 1902. Notes on Nymphaea polysepala on p. 347. Chifflot, J. B. J. Contributions a l’étude de la classe des Nymphéinées, Ann. Univ. Lyon 10: 1-294. 1902. A morphological and histological study of various members of the family, including Nymphaea advena and N. kalmiana. Cook, M. T. Development of the embryo-sac and embryo of Castulia odorata and Nymphaea advena. Bull. Torrey Club 24: 211-220. pl. 12, 15. 1902. Coulter, J. M. The internal hairs of Nymphaea and Nuphar. Bot. Gaz, 6: 250- 255, 1831. Review ofan article by C. F. Cox in American Monthly Microscopical Jourpal for June and July, 1881. Coville, F. V. Notes on the plants used by the Klamath Indians of Oregon. Contr. Nat. Herb. 5: 87-108. .1897. Contains notes on the uses of Nymphaea polysepala. Wokas, a primitive food of the Klamath Indians. Rep, U.S, Nat. Mus, 1902: 725-739. pl. 1-18. 1904. An extensive account of the economic uses of Nymphaea polysepala. De Candolle, A. P. Prodromus naturalis regni vegetabilis 1: 113-116. 1824. On p. 116 the section Nuphar of the genus Nymphaea is treated, with 6 species listed, 3 of them from North America. MILLER AND STANDLEY—NORTH AMERICAN NYMPHAEA. 107 Du Mortier, B. Note sur deux Nymphéacées du Luxembourg. Bull. Soc. Bot. Belg. 3: 4-8. 1864. Eaton, Amos. Manual of Botany for North America 305. 1836. Cites 3 species of Nuphar from North America. Elliott, SrepHeN. Sketch of the Botany of South Carolina and Georgia 2: 8-9. 1824. Two species described, with notes. Engelmann, Georce. In Parry’s Physiography of the Rocky Mountains. Appen- dix Trans. Acad. St. Louis 2: 283-285. 1865. Describes Nuphar polysepalum and subspecies pictum, Fernald, M. L. Notes on some plants of northeastern America. Rhodora 10: 46-55. 1908. Nymphaea advena variegata, a new combination. Gandoger, M. Flora Europae terrarumque adjacentium 1: 308-313. 1883. Publication of a large number of new names in Nymphaea and Nuphar. Gorman, M. W. Economic botany of Alaska. Pittonia 3: 64-85, 1896. Notes upon Nymphaea polysepala included. Gray, Asa. Manual of the botany of the northern United States 24. 1848, Two forms of Nuphar admitted. Greene, E. L. Bibliographical notes on well-known plants. II. Nymphaea and Nuphar. Bull. Torrey Club 14: 177-179. 1887. Notes on the application of the two generic names. Bibliographical notes on well-known plants. VII. Castalia and Nymphaea, Bull. Torrey Club 15: 84-85. 1888. Two new combinations made in the genus Nymphaea. Harper, R. M. Some new or otherwise noteworthy plants from the coastal plain of Georgia. Bull. Torrey Club 38: 229-245. 1906. Nymphaea fluviatilis described. Havard, V. Food plants of the North American Indians. Bull. Torrey Club 22: 120. 1895. Notes on the economic uses of Nymphaea polysepala. Hooker, Wm. J. Flora Boreali-Americana 1: 32-3. 1829. Contains notes on the habits of several species of Nuphar. Lawson, Georce. On the Nymphaeaceae. Proc. & Trans. Roy, Soc. Canada 6: 97-125, 1889. Extended notes on the family. Linné, Cart von. Species Plantarum 510. 1753. Publishes the genus Nymphaea with four species. Under one, WN, lutea, there is a reference to North American material. Michaux, ANDREAS. Flora Boreali-Americana 1: 311-12. 1803. Recognizes, under Nymphaea, three forms of the genus from North America, two of them new, Miller, G. S., jr. The large yellow pond lilies of the northeastern United States. Proc. Biol. Soc. Washington 15: 11-13. pl. 2. 1902. Nymphaea variegata distinguished from N. advena. Mohr, Cuarves. Plant life of Alabama. Contr. Nat. Herb. 6: 504. 1901. Notes on N. advena and N. sagittifolia in Alabama. Morong, THomas. Revision of the North American species of Nuphar. Bot. Gaz. 11: 164-169. pl. 6. 1886. Two new forms published. Peck, ©. H. Report of the Botanist. Ann. Rep. N. Y. Mus. Nat. Hist. 34: 24-58. 1881. A new subspecies described on p, 53. Plants of North Elba, Essex County, New York. Bull. N. Y. State Mus. 6: 67-266. 1899. A new combination made on p. 75. Piper, ©. V. Flora of the State of Washington. Contr. Nat. Herb. 11: 264-5. 1906. Notes on the distribution and habits of Nymphaea polysepala. Pringle, ©. G. Northeastern notes, 1879. Bull. Torrey Club 6: 365, 1879. Remarks upon two species of Nuphar. 108 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Provancher, ABBE L. Flore Canadienne. 1862. _On p, 28 the new name Nuphar americanum published. Pursh, Freperick. Flora Americae septentrionalis 2: 369-70. 1814. Four species recognized under Nuphar and a new combination made. Rafinesque-Schmaltz, C.S. New flora and botany of North America 2: 17. 1836, The genus Ropalon published. Robertson, CHARLES, Flowers and insects. Bot. Gaz. 14: 120-126. 1889, On p. 122 an extended discussion of Nuphar advena., Robinson, B. L. In A. Gray, Syn. Fl. 1: 77-79. 1895. The genus Nuphar with four species and three subspecies recognized, Salisbury, R. A. Descriptions of the natural order of Nymphaeeae. Ann. Bot. 2: 69-76. 1806. The new name Nymphaea arifolia published. Schuster, Jutrus. Zur Systematik von Castalia und Nymphaea. Bull. Herb. Boiss. 7: 853-868, 901-916, 981-996. 1907. 8: 65-74, pl. 1. 1907. An exhaustive account of the European Nymphaeas with references to American ones. Sibthorp, Jon. Florae Graecae prodromus 1: 361, 1808. Publication of the generic name Nuphar. Small, J. K. Studies in the botany of the southeastern United States. V. Bull. Torrey Club 28: 125-130. 1896. Describes Nymphaea orbiculata. Studies in the botany of the southeastern United States. XIV. Bull. Torrey Club 25: 465-484. 1898. Describes Nymphaea macrophylla, Torrey, JoHN. Flora of the State of New York. Nat. Hist. N. Y. pt. 2. 1: 39-40, 1843. Three species of Nuphar recognized. and Gray, Asa. Flora of North America 1: 57-8. 1838. Three species and two subspecies recognized under Nuphar, one subspecies new. Walter, THomas. FI. Carol. 154-155. 1798. Recognizes two species of Nymphaea of the section Nuphar. DESCRIPTIONS OF NEW PLANTS PRELIMINARY TO A REPORT UPON THE FLORA OF NEW MEXICO. By E. O. Wooton and PavL C. STANDLEY. INTRODUCTION. For several years the writers have had in preparation a flora of New Mexico. The manuscript of the grasses was the first part to be completed, the material of this family in the herbarium of the New Mexico Agricultural College having been carefully worked over in the winter of 1906-7. There has been published recently as a bulletin of the New Mexico Agricultural Experiment Station a list of the grasses and grass-like plants of the State in which are enumerated all the species of the Poaceae, Juncaceae, and Cyperaceae, which we have seen from within our limits, the paper being accompanied by keys to the genera and species. While the material in the Agricul- tural College Herbarium was the basis of that report, all of that in the National Herbarium has now been gone over carefully, resulting in many additions to our first list. The Cactaceae of the State have been treated in a similar bulletin published by the same institution. The flora of New Mexico is par- ticularly rich in representatives of this family, no less than 67 being known to occur within the State, while additional ones are doubtless to be discovered. Other briefer papers dealing with special groups of New Mexican plants have been published from time to time, notably one treating the genus Androsace,’ one upon the genus Delphinium,’ and several describing miscellaneous new species from the State.’ 1 Wooton, E. O., and Standley, Paul C. The Genus Androsace in New Mexico. Bull. Torrey Club 34: 517-520. 1907. 2 Wooton, E.O. The Larkspurs of New Mexico. Bull. Torrey Club 37:31-41. 1910. ® Wooton, E. O. A new Southwestern Rose. Bull. Torrey Club 25:152-154. pl. 335. 1898. Wooton, E. O. New Plants from New Mexico. Bull. Torrey Club 25: 257-264, 804-310, 451-459. 1898. Wooton, BE. O., and Standley, Paul C. Some Hitherto Undescribed Plants from New Mexico. Bull. Torrey Club 86: 105-112. 1909. Wooton, E. O., and Standley, Paul C. A new Lathyrus from New Mexico. Muhlenbergia 5: 87, 1909. Standley, Paul C. More Southwestern Castillejas. Muhlenbergia 5: 81-87. 1909. 109 110 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, ° During the winter of 1910-11 both writers gave most of their time to the completion of the report upon the flora of New Mexico, the work being carried on at the National Herbarium. Here are found nearly all the larger collections made in the State from the earliest times, the only important exceptions being those of the James, Wis- lizenus, and Emory expeditions. The National Herbarium includes the larger sets of Fendler, Wright, Bigelow, Newberry, Heller, Earle, Skehan, and Metcalfe, besides many smaller ones, prominent among the last being one by Dr. E. A. Mearns and another of several hun. dred numbers gathered by members of the Biological Survey in con- nection with their field work in the State. More important than these collections, at least for the purposes of determining distribu- tion, are those secured by the writers themselves, embracing several thousands of specimens from almost every part of the State. During the summer of 1911, besides, a careful examination was made of all the New Mexican material in the herbarium of the Agricultural College, resulting in the addition of many other species to our lists. The latter herbarium contains all the more recent sets of New Mexi- can plants, besides specimens of nearly all collections made by the writers. In addition, there are hundreds of smaller collections to be found nowhere else, not having been made in duplicate. Such are those of Prof. T. D. A. Cockerell, Dr. C. L. Herrick, Mrs. W. T. Bartlett, Miss Charlotte Ellis, Messrs. Maltby and Coghill, and many others who have forwarded their specimens to the Agricultural Col- lege for identification. As a result of our study of this rather ample material we have com- piled a list of the plants of the State, which shows that the flora of New Mexico will compare favorably in number of species with that of any of the Western States. It is to be remembered that the plant life of the State is still imperfectly known, except in certain limited localities. Even in those areas which have been fully investigated unknown plants are often turning up; and there are extensive ranges of mountains and hills, as well as stretches of plains, where little or no collecting has been done. For example, the Jemez Range, one of the largest in the State, has never been visited by a botanist. Fewer things of interest are to be expected there, however, than in some of the regions near the boundaries, particularly on the eastern and southern sides. One of the writers in the summer of 1911 collected in the northwestern corner of New Mexico and found more than a hundred species that had not been known previously from the State. Equally productive would be collections made along the southern edge of New Mexico, especially in the Guagalupe and San Luis Moun- tains and about the south end of the Sacramentos. Along the western border there may be expected many Arizona species which have not yet been collected in New Mexico. When it is realized that the area * WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 111 of New Mexico is above one hundred and twenty thousand square miles, which is considerably more than the combined areas of New York and the New England States, and that the number of those who have collected extensively in the region is less than a dozen, it is clear that there remains a fertile field for exploration by those inter- ested in taxonomic botany. When new plants are stil] being found in New England, where for the past century or more hundreds of botanists and botanical collectors have been at work, it is evident that it will be many years before any botanist working in almost any part of New Mexico will fail to find plants that have not before been reported from the State. There has been published but a single flora covering any consider- able part of the great southwestern region, Coulter’s Botany of Western Texas, although Doctor Gray published several more or less extended papers dealing with various collections from New Mexico. The latter, however, were reports upon the collections of the earliest botanical explorers, who passed through the country hurriedly and were unable to visit the most interesting collecting grounds. Certain groups of southwestern plants have been treated in monographs, but the material from New Mexico examined by students heretofore has been scant and often imperfect. Hence, as might be expected, an attempt to write a complete flora of New Mexico in the light of abundant material has found the taxonomy of our plants in an alinost chaotic condition. This is particularly true of those groups which have not been monographed recently. The number of plants here described as new is thus rather large. The diagnoses published in the present paper, however, include practically all undescribed species found by us while working upon the flora of the State. The manuscript for the proposed New Mexican flora is nearly com- pleted and, it is expected, will be published shortly. It is deemed advisable for several reasons to issue the descriptions of the new spe- cies in advance of the complete work. A state flora in its usual form is bulky enough with the material that it must contain without being burdened with pages of descriptions of new plants. Moreover, the amateur in botanical work, for whose use a flora is chiefly intended, is likely, unless all the species are described therein, to have his atten- tion attracted especially to those plants of which he finds descriptions and to strive unduly to associate his specimens with those species. The insertion of occasional diagnoses destroys the uniformity of a work also and seems to us in every way undesirable. The descriptions published here are arranged by families in their natural order, the genera and species in most cases being in alpha- betical sequence. Most of the species discussed are from the southern part of the State, where less botanical work has been done than else- where. Not a few, however, are from the northern part. The latter 112 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. might have been expected in Doctor Rydberg’s Flora of Colorado, and their omission may have resulted from their being overlooked within that State; or they may be plants which do not range so far north. While the flora of those mountains which are the southern extension of the Sangre de Cristo Range, and form the great mass of peaks lying between Santa Fe and Las Vegas, is similar to that of the mountains of southern Colorado, there are found in their canyons and on their slopes more than a few well-known plants which appar- ently do not occur in Colorado. The types of the following new species are all in the National Herbarium. With a few exceptions, which are plainly indicated, all are from New Mexico. In nearly every case we have had access to an abundance of material, consisting either of several collections from adjacent or distant localities or of additional individuals of the type collection. In every instance in which the material was scanty the plant was one so strongly marked that its specific distinctness could not be questioned. Nearly all the new species, as well as new names, are published jointly by the two authors. Exceptions are clearly indicated. We have included descriptions of several plants determined as new spe- cles by Dr. E. L. Greene, but never described. These are principally from the collections made by Mr. O. B. Metcalfe. Many other plants of Mr. Metcalfe’s collections, distributed under new names, we have associated with published species. The present paper includes also descriptions of several new species detected by Prof. J. J. Thornber, of the University of Arizona, in connection with his work upon the flora of that State. DESCRIPTIONS AND NEW NAMES. POACEAE. Aristida pansa Wooton & Standley, sp. nov. Erect, cespitose perennial, 25 to 40 em. high, forming tufts 10 cm. or more in diameter; culms simple, rigid, though slender, minutely puberulent, glabrous in age, somewhat striate; leaves mostly basal, narrowly linear, involute, striate, puberulent throughout, more or less curled; sheaths of the basal leaves overlapping, 1 to 2 cm. exposed, those of the culms 4 to 5 cm. long, closely in- vesting the culms; ligule a tuft of very fine, white hairs encircling the sheath both inside and out; blades 5 to 15 cm. long, those of the upper culms shortest; panicle at first strict, 10 to 20 cm. long, bearing many more or less crowded spikelets, at last spreading, the branches rigidly ascending, mostly in pairs; rachille slender, almost capillary, bearing several crowded small spikelets; glumes slightly unequal, the first shorter, glabrous, narrowly lanceolate, acumi- nate, 1-nerved, the nerve sometimes slightly produced, purple when young, yel- lowish in age, the second glume about the length of the lemma; this 8 to 10 mm. long, attenuate upwards, slightly twisted at maturity, scabrous above, callous and bearing a tuft of white hairs; awns short, 10 to 20 mm. long, at WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 1138 first merely spreading, in age twisted at the base and bent at right angles to the glume. Type in the U. S. National Herbarium, no. 690259, collected on Tortugas Mountain, Dona Ana County, October 6, 1904, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Tortugas Mountain, October 22, 1892, Wooton 1087. Lloyd’s 195, from foothills near Hacienda de Cedros, Mazapil, Zacatecas, Mexico, collected in 1908, is probably the same species. " In the type locality the species is associated with several others of the genus. It somewhat resembles A. vaseyi, with which it is found, but may be recognized by the spreading panicle with its numerous spikelets and by the widely diverg- ent awns. Aristida vaseyi Wooton & Standley, sp. nov. Aristida reverchoni augusta [angusta] Vasey, Contr. U. S. Nat. Herb. 3: 46. 1892. TYPE LOCALITY: “ Comanche Peak,” Texas. SPECIMENS EXAMINED: Socorro, 1895, Plank 56; mountains west of San An- tonio, 1908, Wooton 8860; Tortugas Mountain, October 6, 1904, Wooton; Pena Blanca, October 21, 1906, Wooton & Standley; Mangas Springs, September 2, 1897, Metcalfe; Alamogordo, 1908, Hitchcock 2542. Our New Mexican material exactly matches the type (Reverchon’s plant), and some of it has been referred to A. reverchoni by various students of the genus. It is probably a distinct species. There seems to be a typographical error in the original publication, the name being printed augusta, not angusta, as Doctor Vasey doubtless intended. In order to avoid the use of a name about which there is some uncertainty, and at the same time to give the plant the specific rank it certainly deserves, we dedicate it to Dr. George Vasey, who was for years a careful student of the grasses of the southwestern region and first recognized this plant as distinct. This may prove to be A. fasciculata Torr., described from material collected by Doctor James in the “forests of the Canadian,” a locality somewhere in northeastern New Mexico. We have been unable to compare our material with the type of that species or with authentic specimens, CONVALLARIACEAE. Salomonia cobrensis Wooton & Standley, sp. nov. Rootstocks slender; stems slender, somewhaj flexuous, 20 to 40 cm. high, glabrous; leaf blades elliptic, 50 to 95 mm. long, 10 to 82 mm. wide, acute, nar- rowed at the base into a broad petiole 3 or 4 mm. long, glabrous, slightly paler beneath, very faintly nerved, none of the nerves prominent except the midrib; peduncles strongly and sharply deflexed, 10 to 15 mm. long, each dividing into 2 or 3 slightly shorter pedicels, these stout and strongly flattened laterally, glabrous; perianth 12 to 19 mm. long, tubular, somewhat expanded toward the mouth, the lobes oblong, obtuse, twice as long as the tube; anthers 6 mm. long, acute, slightly exceeding the almost filiform, slightly roughened filaments; no mature fruit seen, but that present about 6 mm. in diameter. Type in the U. 8. National Herbarium, no. 36170, collected in June at the Copper Mines (Santa Rita) by Dr. J. M. Bigelow (Mexican Boundary Survey no. 1478). ADDITIONAL SPECIMENS EXAMINED: Copper Mines, 1851, Wright 1917; near Kingston, 1904, Metcalfe 1036. 114 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Our plant is related to S. biflora and 8. commutata. From the former it differs in its narrower and glabrous leaves, strongly deflexed peduncles, and larger flowers; it agrees with it, on the other hand, in the form of the stamens. From 8. commutata it differs in the form of the stamens, the strongly de- flexed peduncles, and the narrower and less prominently nerved leaves. ALLIACEAE. Allium deserticola (Jones) Wooton & Standley. Allium reticulatum deserticola Jones, Contr. West. Bot. 10: 30. 1902. Allium reticulatum as applied to plants of southern New Mexico in various reports by Watson, not Fraser. This is the largest flowered wild onion we have in the State. The flowers are pale pinkish to white with a darker midrib, fading to a dry papery envelope in fruit. The plant is found with us in the foothills of the hotter and drier moun- tains. It is said to extend into eastern Utah and southern Colorado. We have it from the mountains of the northwestern corner of the State and from the Organ Range. Allium rhizomatum Wooton & Standley, sp. nov. Plants about 20 em. high, arising from small, single, scaly-coated, ovate bulbs about 1 cm. in diameter, the outer coats grayish and opaque, the inner white and hyaline, having a few indistinct longitudinal nerves but not reticulate, the bulbs arising from slender, scaly rhizomes 2 or 8 em. long; leaves 2 or 3, gen- erally longer than the scape, flat, 2 to 3 mm. wide, much broader and clasping at the base, very finely serrulate; scape terete; spathe 2-valved, the valves scarious, broadly ovate, acute, at first pinkish-veined, becoming reflexed and white; umbels erect, few-flowered ; pedicels (in young umbels) 1 em. long or less; flowers small for the genus; perianth segments oblong to oblanceolate, acute, 6 to 8 mm. long, pale with purplish or pinkish midvein, slightly carinate at the base; stamens about equaling the perianth, included, the filaments dilated at the base and coalescent; stigma simple; ovary slightly crested; fruit not seen. Type in the U. 8. National Herbarium, no. 690251, collected at the Gila Hot Springs, August 20, 1900, by E. O. Wooton. Transition Zone. We hesitate to name another Allium, but our material is like nothing else we have been able to discover, being the only species with long and slender rhizomes except A. glundulosum Link & Otto, which was named from central Mexico. All material of that species which we have seen comes from the central States of Mexico, not far from Mexico City, a thousand miles or more from the habitat of our plant, and the two are very conspicuously different. DRACAENACEAE. Yucca baileyi Wooton & Standley, sp. nov. Acaulescent; plants solitary; leaves very numerous, rigid, short, 25 to 30 em. long, 4 or 5 mm. wide or even narrower, smooth, glabrous, yellowish green, thick, convex on both sides near the base, toward the apex flat or shallowly concave on the upper surface, often triangular in cross section, armed with a short, stout tip, the edges of the young leaves with faint white margins, abundantly filiferous, the filaments soft and very slender; inflorescence a simple raceme 50 to 80 cm. long, stout, glabrous; lower bracts subtending the flowers elongated, 10 cm. long, scarious and white or purple at the base, with flat, green tips 4 to 8 em. long; upper bracts broad, all or nearly all with green herbaceous tips; pedicels stout, 2 to 3 cm. long, erect in flower and in fruit; WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 115 perianth segments pure creamy white, not greenish, 60 to 65 mm. long, broadly lanceolate or lance-elliptic, acute; style narrowly oblong, about 1 cm. long, white; filaments slender, papillose; fruit oblong, dehiscent, 5 cm. long, 2.5 cm. thick, rough, costate on the back, erect, short-beaked; seeds dull black, semi- orbicular, somewhat oblique, rounded on the angles, 10 to 12 mm. long, 8 or 9 mm broad, smooth. Type in the U. 8S. National Herbarium, no. 686602, collected on a dry slope in pine woods in the Tunitcha Mountains, August 8, 1911, by Paul C. Standley (no. 7638). The description of the fruit is drawn from specimens collected in a canyon of the Carrizo Mountains, July 30, 1911, Standley 7448. The plant is named for Mr. Vernon Bailey of the Biological Survey, who first collected it in the Chusca Mountains at an elevation of about 2,700 meters. This was at first believed to be Y. angustissima Engelm., but when complete material was collected in the summer of 1911 it was seen to differ essentially from that species. Yucca angustissima has much smaller flowers, strongly constricted capsules, a branched inflorescence, and much smaller seeds of a different shape, and lacks the foliaceous bracts of the inflorescence. Yucca baileyi reaches a higher elevation than any other of our species of the genus, being the only one that extends into the Transition Zone. It reaches much farther up into the mountains than Y. baecata. So far as we know the species is confined to the chain of mountains including the Chuscas at one end and the Carrizos at the other. It is not improbable, however, that it may be found in some of the not far distant ranges of Arizona and Utah. Yucca neomexicana Wooton & Standley, sp. nov. Acaulescent from a thick root, propagating by means of numerous radiat- ing rootstocks, the young plants at the ends of these forming a more or less reg- ular circle about the old plant; leaves short, 25 to 30 cm. long, 8 to 10 mm. wide, constricted near the base and there only about 5 mm. wide, acute and tipped with a slender, straw-colored, very sharp point, glabrous, rather thin, smooth, yellowish, the margins white, cartilaginous, abundantly filiferous below the middle with coarse, white, straight but finally curled filaments; inflorescence a simple raceme, 60 to 90 cm. high, stout, glabrous, bearing at the base 1 or 2 short, reduced leaves. above furnished with numerous narrowly triangular, scarious, white to purple bracts; pedicels stout, 12 to 20 mm. long, recurved. in flower; perianth nearly white, 4 cm. long, the segments elliptic-oblong or oblong, obtuse or acutish; style short, greenish, 5 to 7 mm. long, swollen at the base; filaments densely papillose; fruit not seen. Type in the U. S. National Herbarium, no. 685238, collected on a voleanic hill about half a mile vorth of Des Moines, Union County, June 20, 1911, by Paul C. Standley (no. 6208). Altitude about 2150 meters. Additional ma- terial is mounted on sheets 685259 and 685240. The description is drawn chiefly from material*preserved in formalin. Young plants are growing in the greenhouse at Washington. The plant was very abundant in this one locality, growing al! over the top of a low hill, in loose, rocky soil. It was not observed elsewhere. Yucca glauca was common on the sides of the hill but the two were readily dis- tinguished at a glance. The principal difference between the two is in the leaves, but this is so striking that the species can scarcely be confused. AMARYLLIDACEAE. Agave neomexicana Wooton & Standley, sp. nov. PLATE 48. Leaves numerous, crowded together and forming a compact, almost globose rosette 45 to 60 cm. in diameter when mature; leaves 10 to 30 cm. long, oblong 116 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. to ovate-lanceolate, bluish green, glaucous, tipped with a very sharp spine, this brownish black when young, grayish in age, decurrent into a horny, toothed margin bearing 3 or 4 brown or gray, retrorsely hooked teeth on the upper half of the leaf, the lower part armed with fewer and smaller spines, the leaves thick and very rigid; panicle 3 to 5 meters high, with 10 to 15 divergent lateral branches bearing subumbellate clusters of flowers; fresh flowers dull brownish red outside, deep yellow to orange within, 5 to 6 em. long including the exserted style and stamens; filaments attached by an expanded base to the base of the perianth segments in a saceate depression, 2 cm. long, yellow; anthers versatile, 15 mm. long; style slightly exceeding the stamens; fruit an oblong-elliptic, light brown capsule, 25 to 35 mm. long and about 20 mm. in diameter. Type in the U. 8S. National Herbarium, no. 498333, collected in the Organ Mountains, in June, 1906, by Paul C. Standley (no. 541). ADDITIONAL SPECIMENS EXAMINED: ‘Tortugas Mountain, 1911, Standley 6481; San Andreas Mountains, December 10, 1902, J. H. Gaut. This is a part of Agave applanata Mulford, but seems to be very distinct from Agave applanata Lem. EXPLANATION OF PLATE 48.—From a field photograph by K. OQ. Wooton. ORCHIDACEAE. Achroanthes porphyrea (Ridley) Wooton & Standley. Microstylis purpurea 8. Wats. Proc. Amer, Acad. 18: 195. 1888, not Lindl. Microstylis porphyrea Ridley, Journ. Linn. Soc. Bot. 24: 320, 1888. Achroanthes purpurea Greene, Pittonia 2: 184, 1891. FAGACEAE. Quercus confusa Wooton & Standley, sp. nov. A moderately large, spreading tree, 5 to 7 meters high, with short trunk, grayish, checked bark, and dull greenish gray, pubescent, stout twigs; leaves oblong, 6 to 8 cm. long, 2 to 3 cm. wide, bluish green, glabrous above, almost velvety beneath with yellowish stellate hairs, acute, sinuate-dentate with about 4 coarse, mucronate, lobe-like teeth on each side, only slightly crispate; petioles 1 cm. long or less; buds on new growth small, 2 to 3 mm. long, ovoid, with dark brown scales; acorn 20 to 23 mm. long, barrel-shaped, obtuse; cup hemispheric, 15 to 18 mm. in diameter, covering the lower third of the acorn; scales much thickened on the back. Type in the U. S. National Herbarium, no. 691250, collected on Ruidoso Creek 5 miles east of Ruidoso Post-office, August 5, 1901, by E. O. Wooton. Altitude about 1650 meters. ADDITIONAL MATERIAL EXAMINED; Gilmores Ranch on Eagle Creek, alt, 2220 meters, July 25, 1901, Wooton. This species is most nearly related to Q. fendleri, from which it differs in being a tree and in having still larger leaves (persistent?) and a larger acorn. It occurs at a lower level than is common for Q. fendleri, being at home in the Upper Sonoran instead of the Transition Zone, although the latter sometimes comes into the Upper Sonoran. Quercus media Wooton & Standley, sp. nov. Low shrub, 1 to 3 meters high, with gray, smooth bark on the older stems, the young branches brown, sparingly pubescent; buds about 3 mm. long, with Contr. Nat. Herb., Vol. 16. PLATE 48. AGAVE NEOMEXICANA WOOTON & STANDLEY, IN THE ORGAN MOUNTAINS. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 117 rich reddish brown, glabrous scales; leaves green (not bluish green) and glabrate above, paler beneath and velvety-pubescent with yellowish, stellate hairs, oblong, elliptic, or obovate in outline, acute or obtuse, sinuate-dentate with 2 to 4 coarse, lobe-like teeth on each side, 4 to 7 em. long, 2 to 3 cm. wide; petioles 1 cm. long or less; teeth mostly mucronate, not spinulose; acorns small, 10 to 18 mm. long, ellipsoidal, acute, the cup hemispheric, 10 to 12 mm. in diameter, the reddish brown scales little or not at all thickened at the base. Type in the U. 8S. National Herbarium, no. 690255, collected at Glorieta, August 24, 1910, by EK. O. Wooton. Transition Zone. Another specimen of what seems to be the same is from Oak Canyon near Folsom, collected in 1903 by A. H. Howell (no. 178), in leaf only. There would seem to be enough species of Rocky Mountain oaks already described, especially of the type of Q. undulata, which is at best of doubtful standing. Assuming that Q. undulata is a species with bluish green, persistent leaves, the species here described resembles it in nearly all particulars except that its leaves are bright chlorophyll green and probably deciduous. This would make it intermediate between the two groups of the region—the blue green leaved species, which it resembles in habit and shape of leaf, and the green-leaved species, which it resembles in color and texture of leaves and time of dropping its leaves. It might be a hybrid, but the plant was very common about Glorieta, forming numerous clumps of bushes a rod or so in diameter, and Mr. Howell’s plant is almost a perfect match from a similar region farther east. ARISTOLOCHIACEAE. Aristolochia watsoni Wooton & Standley. Aristolochia brevipes acuminata S. Wats. Proc. Amer. Acad. 18: 148. 1883, not A. acuminata Tam. A not uncommon species of southern Arizona and northern Mexico, which barely reaches the southwestern corner of New Mexico. POLYGONACEAE. Eriogonum ainsliei Standley, sp. nov. Perennial, 15 to 25 em. high, from a stout, woody root; stems somewhat cespi- tose, well developed, slender, decumbent at the base and leafy, arachnoid- pubescent; leaves elliptic or linear-oblong, 3 to 4 em. long, acutish, glabrate above, tomentose beneath, extending about half way up the stem, all on petioles one-third to one-half the length of the blades, attenuate at the base; inflor- escence corymbose, the primary branches subtended by linear-subulate bracts 3 to 5 mm. long, the other branches furnished with smaller bracts; involucres short-pedunculate, 3 mm. long, 5-angled, viscid-tomentulose; perianths white tinged with purplish pink, glabrous, the segments obovate; fruit glabrous. Type in the U. S. National Herbarium, no. 592284, collected at Cimarron, September 20, 1909, by Mr. C. N. Ainslie of the Bureau of Entomology. Addi- tional material of the same collection is mounted on two other sheets. ADDITIONAL MATERIAL EXAMINED: Cimarron, September 10, 1909, Ainslie; Raton Mountains, 1903, Griffiths 5097; Colfax, August 13, 1910, Wooton. From the most closely related species, H. nudicaule and EH. tristichum, this plant may be distinguished by its pubescent inflorescence and by the acute lobes of the involucre, 118 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Eriogonum gypsophilum Wooton & Standley, sp. nov. PLATE 49, Perennial from a thick, woody, cespitose base, the short branches covered with the villous, scale-like bases of old leaves; leaves all basal, thickly clus- tered, broadly ovate to rotund or reniform, entire, abruptly mucronate, the blades 1 to 2 cm. long, 2 to 3 cm. wide, yellowish green, thick and succulent, glabrous except for a few hairs on the veins beneath and sometimes on the margins; petioles 2 to 3 cm. long, slender, villous, especially at the base; inflo- rescence a trichotomous cyme 10 to 15 em. high; bracts small, not leaf-like, the lowest sparingly villous, the upper glabrous; involucres broadly campanulate, 4 or 5-toothed, glabrous, with 6 flowers; pedicels slender, articulated at the base of the perianth, 1 mm. long or less; perianth broadly companulate, becoming urceolate, the segments ovate, acute or obtuse, the midrib greenish, otherwise bright yellow, sparingly white-pubescent on the middle or glabrous. Type in the U. S. National Herbarium, no 564576, collected on a hill south- west of Lakewood, growing in pure gypsum, August 6, 1909, by E. O. Wooton. The hill is capped by 50 to 100 feet of limestone, the gypsum appearing in several layers in the lower two-thirds. Our plant did not grow on the lime- stone soil but was restricted to the outcroppings of the gypsum strata. The species belongs to the section Corymbosa as used by Doctor Rydberg, but is not at all closely related to any of our western species. EXPLANATION OF PLATE 49.—Part of the type specimen. Natural size. Eriogonum leptophyllum (Torr.) Wooton & Standley. Eriogonum effusum leptophyllum- Torr. in Sitgreaves, Rep. Zuni & Colo. 168, 1854. The plant is similar to H. effusum, with which it was at first associated, but differs in its linear and revolute instead of oblong and flat leaves, and in the low, sparingly branched inflorescence not more than 5 cm. high. In #. effusum the inflorescence is densely branched and often 15 to 20 cm. dong. Eriogonum leucophyllum Wooton & Standley, sp. nov. Perennial from a thick, woody root, cespitose; leaves all basal, elliptic, 13 to 20 mm. long, about 7 mm. wide, narrowed at the base into a broad petiole 7 to 11 mm. long, densely tomentose beneath, sericeous on the upper surface, white in general appearance; stems scapiform, simple below, about 30 em. high, slender, tomentose below, tomentulose about the inflorescence, loosely corymbose above, the corymb being 10 to 15 ecm. high, its slender branches ascending; involucres in the forks of the branches on slender peduncles 20 to 35 mm. long, the others on peduncles 7 mm. long or more, broadly campanulate, 2 mm. high, with triangular teeth almost equaling the tube, finely sericeous; perianth yellow, densely silvery-pubescent, some of the flowers reflexed in age; ovary densely pubescent. Type in the U. S. National Herbarium, no, 564577, collected at Lakewood August 6, 1909, by E. O. Wooton. While related to H, lachnogynum, our plant is evidently distinct in its broader, more densely pubescent leaves, taller stems, openly branched in- florescence, and much smaller and more numerous involucres. . Eriogonum pannosum Wooton & Standley, sp. nov. Perennial from a thick, woody caudex covered with the persistent bases of the dead leaves; stems numerous, stout, 20 to 45 cm. high, corymbosely branched above, densely white silky pubescent; leaves mostly basal, those of the stem few, scattered, reduced, the basal ones oblanceolate or spatulate, obtuse, abruptly short-acuminate, narrowed at the base into a long, margined petiole, 40 to 65 mm. long, densely and finely tomentose beneath, sericeous Contr. Nat. Herb., Vol. 16. PLATE 49. ERIOGONUM GYPSOPHILUM WOOTON & STANDLEY. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 119 above with a white pubescence, this often denser along the veins so as to produce the appearance of longitudinal silvery lines; involucres 2 or 3 mm. long, campanulate, sericeous, the teeth low, triangular, acute, all on stout pedicels 5 to 10 mm. long; perianth 2 mm, long, yellow, conspicuously sericeous, the segments oblong, on slender, glabrous pedicels reflexed at maturity ; achenes glabrous, 3 mm. long, spherical and turgid at the base, narrowly winged above the middle. Type in the U. 8. National Herbarium, no, 45775, collected in the Organ Mountains, August, 1881, by G. R. Vasey. ADDITIONAL SPECIMENS EXAMINED: Organ Mountains, June 25, 1894, Wooton. Our proposed species is nearest HL. hieracifolium. That species differs, how- ever, in the loose and longer, coarser pubescence of the leaves and stem, greater size, larger and narrower, more acute leaves, larger involucres 3 to 5 mm. long, and the larger perianths. CHENOPODIACEAE. Atriplex flagellaris Wooton & Standley, sp. nov. Perennial; stems trailing, slender, 30 to 40 cm. long, weakly ascending at the tips; cortex at first densely white-scurfy, becoming glabrous, shining, shreddy near the base of the stems; leaves numerous, small, 1 to 3 cm. long, scarcely half as broad, oblong to narrowly obovate, tapering into a short petiole, glabrate above, white-scurfy beneath, obtuse or acute, the margins entire or with a few coarse teeth on each side; flowers few, axillary; pistillate flowers 2 to 5 in the axil, usually only one producing fruit; staminate flowers in small, spherical heads 2 mm. in diameter in the same axils; fruiting bracts cuneate-obovate, 5 or 6 mm. Jung, united to above the middle, prominently 3 to 5-nerved, not appendaged on the back, the upper part of each bract herbaceous with one large central tooth and one or two small Iuteral ones on each side; seed lenticular. Type in the U. S. National Herbarium, no. 562291, collected in the Mesilla Valley, June, 1906, by Paul C, Standley (no. 490). ADDITIONAL SPECIMENS EXAMINED: Mesilla Valley, July 4, 1906, Wooton; Mesilla Valley, May 1, 1907, Wooton & Standley. Judging from the description alone and from the character of the fruit, our plant is related to A. barclayana D. Dietr., but it differs in having much larger fruit with fewer teeth, while the plant is much smaller and slenderer than that species of the western coast of Mexico. This is a dooryard and wayside weed commonly found in locations preferred by A. elegans, with which it was confused for a long time. Herbarium speci- mens look somewhat like that species, but the habit of the two is very different, as are their fruits. Atriplex collina Wooton & Standley, sp. nov. Low, densely branched shrub, 25 cm. high or less, forming broad, rounded clumps; branches ascending or spreading, ending in sharp, spinose tips, stout, loosely but copiously lepidote; leaves very numerous, small, 2 em. long or shorter, elliptic-oval to elliptic or nearly lanceolate, thick and fleshy, densely lepidote, obtuse or acutish, acute or cuneate at the base, on very short, broad petioles; plants diecious, apparently, only the pistillate collected, the fertile flowers axillary, sessile; bracts united only at the base, rather thin, very broadly ovate or quadrilateral, 8 mm. long or smaller, densely lepidote, broadly cuneate at the base, acutish, all or nearly all obtusely dentate on the margins, smooth on the backs. 60541°—18——2 120 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Type in the U. S. National Herbarium, no. 686447, collected on dry hills near the north end of the Carrizo Mountains, July 31, 1911, by Paul C. Standley (no. 7481). Related to A. confertifolia, but with dentate bracts not rounded at the apex, much smaller leaves, and sessile fertile flowers. AMARANTHACEAE. Gomphrena viridis Wooton & Standley, sp. nov. Low, cespitose perennial from a long, woody root; basal leaves bright green, obovate to elliptic-oblong. the blades 3 to 7 em. long, obtuse or acutish, gradually narrowed at the base ‘into a long, slender petiole, sparingly strigose with fine short hairs or nearly glabrous on the upper surface; stems slender, prostrate or spreading, tortuous, 3 to 10 cm. long, cinereous-puberulent to thinly sericeous ; cauline leaves a single pair, in outline like the basal ones, or broader, often orbicular, on slender petioles 5 to 10 mm. long; peduncles terminal, slender, 35 to 70 mm. long, loosely sericeous but not densely so; heads subglobose, 8 to 20 mm. high; bracts scarious, white, ovate, acute; calyx lobes ljinear-oblong or oblanceolate, obtuse, entire, with a broad green midvein and scarious white margins, densely long-hairy. Type in the U. 8. National Herbarium, no. 660403, collected on Hanover Mountain, Grant County, July 31, 1911, by J. M. Holzinger. ADDITIONAL SPECIMENS EXAMINED: 1851-2, Wright 1753: base of San Luis — Mountains, up to 1800 meters, September 5, 1893, Mearns 2133. The species also occurs in southeastern Arizona. Closely related to Gomphrena caespitosa, but with, green, narrower leaves, Sparse pubescence, more conspicuously petioled cauline leaves, and promi- nently green calyx lobes. ALLIONIACEAE. Allionia linearifolia filifolia Standley. Allionia gracillima filifolia Standley, Contr. U. 8S. Nat. Herb. 12: 340, 1909. An examination of the type of Orybaphus linearifolius S. Wats. shows that Allionia graciilima Standley is a synonym of that species. Allionia subhispida (Heimerl) Standley. Mirabilis linearis subhispida Heimerl, Ann. Cons. Jard. Genéve 5: 186. 1901. Allionia linearis subhispida Standley, Contr. U. S. Nat. Herb. 12: 342. 1909. This was well described by Doctor Heimerl. It may be distinguished from A. linearis by its abundant hirsute pubescence present on all parts of the branches. It has been collected in New Mexico several times recently. PORTULACACEAE. Talinum angustissimum (A. Gray) Wooton & Standley. Talinum aurantiacun angustissimum A, Gray, Pl. Wright. 1: 14. 1852. This has long been confused with T. aurantiacum, a larger, stouter, more succulent plant with larger, orange-colored flowers. It is difficult to distinguish the two by herbarium specimens, but no one can confuse them in the field. Both species have been confused with 7. lineare H. B. K., a plant known only from central Mexico. Talinum longipes Wooton & Standley, sp. nov. Root slender, very long and tortuous, the crown covered with the persistent bases of dead leaves; leaves humerous and crowded, appearing basal, 12 to WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 121 20 mm. long, terete, acutish, slender; scapes very slender, 10 to 12 cm. high, corymbosely branched above; flowers few (3 to 5), all on slender pedicels 3 to 6 mm. long; bracts lanceolate or triangular, scarious; sepals nearly orbicular, 2 mm. long, very thin; petals pinkish, 4 or 5 mm. long; stamens 10; capsule nearly spherical, 3.5 mm. high. Type in the U. S. National Herbarium, no. 690249, collected on Tortugas Moun- tain, August 27, 1894, by E. O. Wooton. Apparently the same is part of Wright’s 875 in the National Herbarium. Our plant is unlike the related species, such as 7. parviflorum, in the form of the sepals and the number of stamens, and in the well-developed pedicels. Talinum pulchellum Wooton & Standley, sp. nov. Roots stout, dark brown, woody; stems stout and fleshy, 10 cm. high or less, simple below, coryimbosely branched above; leaves apparently terete, perhaps slightly flattened, 12 to 20 mm. long, 1.5 to 2 mm. in diameter, not narrowed at the base, blunt, scattered along the stems; flowers axillary, solitary; pedun- cles stout, 2 or 3 mm. long; pedicels slender, 6 to 20 mm. long; sepals elliptic- lanceolate, about 7 mm. long, acute, smooth, greenish below, scarious-margined ; petals about 15 mm. long, purplish red; stamens about 20. . Type in the U. S. National Herbarium, no. 617671, collected near Queen, August 2, 1909, by E. O. Wooton. Altitude about 1,770 meters. The proposed species is nearest 7. brevifoliwm, but differs in habit, size of flowers, form of sepals, and length of pedicels. The two species form a sec- tion very unlike the other members of the genus, ALSINACEAE, Arenaria mearnsii Wooton & Standley, sp. nov. A slender, diffusely branched perennial; stems reclining at the base, minutely cinereous-puberulent; leaves linear or linear-elliptic, bright green, numerous, glabrous, somewhat pungently pointed, 8 to 12 mm. long, 2 mm. wide or less; flowers numerous, on slender, ascending, almost glabrous pedicels 10 to 15 mm. long; sepals lanceolate to lance-ovate, attenuate to long, subulate tips, glabrous, bright green with scarious margins, 4 to 5 mm. long; petals 1 or 2 mm, longer than the sepals; capsules 1 to 2 mm. shorter than the sepals. Type in the U. 8S. National Herbarium, no. 233375, collected in a canyon of the San Luis Mountains on the New Mexico-Mexico boundary, September 11, 1893, by Dr. E. A. Mearns (no. 2216). Our specimens seem not to agree with any described species of the United States or Mexico. They are nearest A. sarosa and A. confusa, but differ from both in the narrow, glabrous, more or less pungently pointed leaves and sparser pubescence. From the former they also differ in the longer pedicels, and from A. confusa they are distinguished by the longer sepals and ascending pedicels. Drymaria pachyphylla Wooton & Standley, sp. nov. Annual;,stems slender, prostrate, glabrous, sparingly branched; basal leaves spatulate; cauline leaves ovate, obtuse, glabrous, glaucous, thick, 10 mm. long or less, 6 to 8 mm. wide, narrowed at the base into a slender petiole one-half as long as the blade or more; flowers solitary on the pedicels, clustered in the axils, on rather stout, glabrous pedicels 3 or 4 mm. long; sepals oblong, 3 mm. long, glaucous, with thin, scarious, white margins; capsule slightly exceeding the sepals. Type in the U. S. National Herbarium, no. 830629, collected on the dry plains south of the White Sands, August 20, 1897, by E. O. Wooton (no. 405). Alti- 122 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. tude about 1,230 meters. Also gathered by the same collector at the same place, August 5, 1899, The type collection was distributed as D. holosteoides Benth., a plant of Lower California. Our species is cited from western Texas under this name in the Synoptical Flora. That species, however, has much narrower, acutish leaves, and puberulent pedicels, Another related species is D. crassifolia Benth., also of Lower California, but that has much thicker, fleshier leaves, and is a very densely branched perennial. Besides the New Mexican specimens cited above we have a sheet collected by Havard on the Tarlinga River in western Texas. FUMARIACEAE. Capnoides euchlamydeum Wooton & Standley, sp. nov. Annual or biennial with very slender, ascending or decumbent, glabrous stems; leaves twice pinnate, glabrous, the ultimate segments cuneate or oblong, acute or obtuse, bright green above, glaucous beneath; primary pinne mostly divaricate, a pair inserted usually almost at the base of the rachis; flowers in slender, few-flowered racemes; bracts 12 to 25 mm. long, broadly oblanceolate, 4 to 9 mm. wide, acute or acuminate, thin; corolla bright yellow, about 15 mm. long; spur horizoutal, thick, half as long as the body; fruit 20 to 30 mm. long, rather slender, not very conspicuously torulose, strongly curved, not angled, on a stout, deflexed pedicel; seeds black and shining, almost smooth, with very obtuse margins. Type in the U. 8. National Herbarium, no. 690256, collected at Cloudcroft in the Sacramento Mountains, August §, 1890, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: James Canyon, June £6, 1899, Wooton; Ruidoso Creek, alt. 1,980 meters, June 29, 1895, Wooton; Cloudcroft, 1912, Stearns 348, This Capnoides is related to C. aurewm, but may be distinguished by its very large bracts and by the presence of pinni at the base of most of the petioles. It is, besides, a rather more slender plant, with more dissected and delicate leaves. RANUNCULACEAE, Clematis neomexicana Wooton & Standley, sp. nov. A woody climber, one or two meters high; stems striate, finely pubescent ; leaves pinnately 5-foliolate, on petioles 4 to 6 cm. long: leaflets ovate in outline, 35 to 60 mm. long, 25 to 45 mm. wide, shallowly 3-lobed, the terminal lobe acute to obtuse, never long-attenuate, the lobes entire or coarsely crenate with obtuse teeth; leaflets bright green, slightly paler beneath, finely and loosely pubescent on both surfaces; flowers in a loose, few-flowered panicle, on a peduncle about 5 cm. long; pedicels about 25 mm. long; sepals oblong-spatulate, obtuse, finely pubescent, 12 mm. long, much exceeding the stamens: carpels densely hairy, tapering gradually into the plumose tail 15 to 35 mm. long. Type in the U. 8. National Herbarium, no. 233000, collected in the San Luis Mountains, September 5, 1898, by Dr. E. A. Mearns (no. 2136). The collector states that it occurs “from the base up to 6000 feet.” ADDITIONAL SPECIMENS EXAMINED: San Luis Mountains, 1898, Mearns 2455; Organ Mountains, alt. 2,100 meters, September 23, 1906, Wooton & Standley; Organ Mountains, alt. 1,410 meters, 1897, Wooton 150; Organ Mountains, alt. 1,950 meters, 1897, Wooton; east side Organ Mountains, alt. 1,350 meters, August 17, 1895, Wooton. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 128 The plant is near C. ligusticifolia, but differs in its pubescent leaflets and stems, the different form of the leaflets, the shorter tails of the carpels, and the shape of the ecarpels. It may be Clematis ligusticifolia californica 8S. Wats., but the Californian material seems to be of a different species. Myosurus egglestonii Wooton & Standley, sp. nov. Plant glabrous; leaves linear or linear-oblanceolate, obtuse, thick and some- what fleshy, 8 to 14 mm. long; scapes solitary or numerous, very short, 2 to 5 mm. long, slender; sepals scarious, linear-oblong, obtuse, 38 mm. long, with a slender spur 1 mm. long; petals not seen, apparently wanting; heads of achenes elongated-oblong, 8 to 18 mm. long, 2.5 mm. in diameter; achenes small, the backs suborbicular, with a low, obtuse border, strongly keeled and with a beak from once to twice as long as the diameter of the back, the beak ascending. Type in the U. S. National Herbarium, no. 660739, collected on a mesa on the road between Tierra Amarilla and Park View, Rio Arriba County, altitude 2,250 meters, April 18 to May 25, 1911, by W. W. Eggleston (no. 6472). Evidently related to M. cupulatus, but readily distinguished by the very short fruiting spikes, the short scapes, and the elongated beaks of the achenes. In size the plant suggests V/. alopecuroides Greene, of California, but the achenes of the two are very dissimilar. Viorna filifera (Benth.) Wooton & Standley. Clematis filifera Benth. Pl. Hartw. 285. 1848, Viorna palmeri (Rose) Wooton & Standley. Mlematis palmeri Rose, Contr. U. S. Nat. Herb. 1: 118. 1891. BRASSICACEAE. Arabis angulata Greene, sp. noy. in herb. Perennial from a rather thick, woody root; stems simple, clustered, slender, erect, 25 to 40 em. high, pubescent below with few branched hairs, glabrous above; basal leaves oblanceolate to spatulate, 25 to 85 mm. long, 11 mm. wide or less, obtuse, with a few coarse teeth, rather bright green, pubestent with branched hairs; cauline leaves rather remote, smaller, oblong-lanceolate to almost linear, sessile, auricled, the auricles mostly acute, the lower leaves pubes- cent, the upper glabrous, obtuse or acute; racemes elongated, slender; pedicels slender, divergent, 12 mm, long or less; sepals purplish, oblong, obtuse, 2.5 mm. long, with scarious margins, usually with a few hairs; petals twice as long, purple; pods slender, 45 to 60 mm, long, 1 mm, wide, curved upward, glabrous; seeds in a single row. Type in the U. S. National Herbarium, no, 495141, collected at Mangas Springs, April 9, 1908, by O. B. Metcalfe (no, 12). Altitude 1,430 meters. A species of the group to which belongs A. fendleri, distinguished especially by its very long, slender pods conspicuously curved upward. Doctor Greene recognized this as an undescribed species when Mr. Metcalfe’s plants were being named, and the plants were dist ributed under the name here given to them; a description, however, has not been published until now. Arabis porphyrea Wooton & Standley, sp. nov. Perennial from a long, slender, woody root; stems slender, erect, pubescent below with branched hairs, glabrous above, purplish; basal leaves 40 mm, long or less, spatulate or oblanceolate, stellate-pubescent, long-petioled ; lower cauline leaves petioled, oblanceolate, obtuse, stellate-pubescent, the upper ones linear or linear-oblong, acute, glabrous, sessile and auricled, reduced; pedicels divaricate or reflexed, 10 or 12 mm. long, slender; sepals oblong, obtuse or acutish, 8 mm. 124 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. long, with a few branched hairs, purple; petals purple, twice as long as the sepals; pods about 85 mm. long and 1.5 mm. wide, purplish, glabrous, stout, straight or curved downward; seeds in 2 rows, winged. Type in the U. 8. National Herbarium, no. 561991, collected on the dry hills near the Cueva on the west side of the Organ Mountains, April 25, 1907, by BE. O. Wooton and Paul C. Standley. ADDITIONAL SPECIMENS EXAMINED: Modoc, March 19, 1905, Wooton; Magda- lena Mountains, June, 1880, Vasey. This somewhat resembles A. angulata, but has much shorter, broader, pur- plish pods curved downward instead of upward. Dithyraea griffithsii Wooton & Standley, sp. nov. Stems erect, stout, branched, densely stellate-pubescent; cauline leaves en- tire, narrowly lanceolate, 25 to 40 mm. long, acute, narrowed to the base, sessile, finely stellate-pubescent on both surfaces; pedicels about 15 mm. long, slender, divergent; flowers numerous, at first congested, becoming more dis- tant after anthesis; sepals narrowly oblong, 8 mm. long; petals obovate, clawed, the whole petal 6 or 7 mm, long; fruit 10 mm. wide, the segments 5 mm. high, truncate at the apex, cordate at the base, glabrous, conspicuously reticulate-veined, the border not well developed. Type in the U. S. National Herbarium, no. 486760, collected by Dr. David Griffiths on the Arroyo Ranch, near Roswell, September 1 to 4, 1903 (no. 5687). A specimen collected by Mrs. Matilda Coxe Stevenson near Zuni in 1902 (no. 74) appears to be the same. From our other species, D. wislizeni, this may be distinguished at once by its glabrous, conspicuously veined fruit. Its cauline leaves, too, are quite entire, while in D. wislizeni almost all have a few sinuate teeth, at least near the base. Draba gilgiana Wooton & Standley, sp. nov. A densely cespitose perennial from a very thick, woody root covered at the crown with the persistent bases of old leaves: leaves mostly basal, very numer- ous, densély clustered, linear-oblanceolate, acutish, 35 mm. long and 3 mm, wide or less, tapering gradually into a slender, flat, yellowish petiole; blades bright green, glabrous, or with a very few scattered hairs; stems long and slender, ascending or reclining, 12 ecm. long or less, glabrous; cauline leaves few and remote, linear to oblanceolate, 4 to 10 mm. long, acute; flowers few, clustered at the end of the stem, the racemes elongating in fruit; pedicels Slender, glabrous, nearly erect, 5 or 6 mm. long; sepals rounded-oblong, glabrous, obtuse, 2.5 mm. long; petals bright yellow, about twice as long; fruit almost 6 mm. long, oblong-lanceolate, acute, glabrous, ending in a slender style slightly more than 1 mm. long. Type in the U. 8. National Herbarium, no. 561295, collected on Organ Peak in the Organ Mountains, September 23, 1906, at an altitude of about 2,550 meters, by E. O. Wooton and Paul C. Standley. The plant was growing high up near the summit of the peak, in the Transition Zone, in rich, deep soil in the shade of oak chaparral. ADDITIONAL SPECIMENS EXAMINED: Van Pattens, August 29, 1894, Wooton; Organ Mountains, alt. 2040 meters, 1897, Wooton 675; Tortugas Mountain, September, 1898, Wooton. The specimen from Tortugas Mountain has even narrower leaves than those from the Organs. It seems improbable that a plant of the dry Lower Sonoran Zone can be the same as one found high up in the moist canyons. Our mate- rial from Tortugas Mountain, however, is insufficient for satisfactory deter- mination, WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 125 Our species is nearest D. petrophila Greene, but it is much more slender and nearly glabrous, while the cauline leaves are fewer and more reduced; the basal leaves, too, are different in outline. The plant was marked as a new species in the National Herbarium by Dr. E. Gilg, but apparently was never published by him, probably because of the scantiness of the material he had examined. Draba tonsa Wooton & Standley, sp. nov. A low perennial, 6 cm. high or less, densely cespitose from a long, thickened root; leaves in a dense cluster at the base of the scape, linear-oblanceolate, obtuse, 6 to 18 mm. long, glabrous except for the long-ciliate margins; stems scapose, with a few scattered leaves very unlike the basal ones, glabrous; cauline leaves oblong to ovate, thick, 5 to 16 mm. long, 2 to 6 mm. wide, obtuse or acutish, glabrous, or with a few long hairs on the margins; flowers congested at the ends of the simple stems, few, on glabrous pedicels 3 mm. long or less; sepals broadly rounded-oblong, obtuse, glabrous, 2.55 mm. long; petals bright yellow, about twice as long as the sepals; ovary glabrous, with a long, slender style; mature fruit not seen. Type in the U. S. National Herbarium, no. 225083, collected on Hermits Peak, in the Las Vegas Mountains, in August (year not given), by F, H. Snow. Also collected near Beulah, at an altitude of 2,400 meters, by T. D. A. Cockerell. Although our specimens have no fruit, we feel safe in describing them as new, for they seem amply distinct from D. streptocarpa, the most nearly related spe- cies, in their glabrous stems and nearly glabrous leaves, the sepals, too, being glabrous, and the cauline leaves broader. Cheirinia desertorum Wooton & Standley, sp. noy. Stout, herbaceous biennial or short-lived perennial, 30 to 50 cm. high, with one or several rigid, more or less angled stems rising from a dense cluster of basal leaves, the whole plant cinereous throughout with the appressed, 2-parted hairs common in the genus; basal leaves very numerous, 10 to 20 cm. long, narrowly lanceolate or oblanceolate, tapering into a slender petiole, acute, mostly entire, or some with a few coarse teeth, persisting until the plant is in fruit; cauline leaves linear, 3 to 5 cm. long, strongly ascending or erect; flowers small, 6 to 8 mm. long, pale yellow, the claws of the petals little if at all longer than the sepals; pods almost terete, 3 to 7 cm. long, erect or strongly ascending, on short, thick, ascending pedicels. Type in the U. S. National Herbarium, no. 563775, collected near Hachita, June 16, 1906, by E. O. Wooton. From the description this would appear to be most closely related to C. parvi- flora, and upon using Doctor Rydberg’s key to the genus in the Flora of Colorado one would run to that species at once. But our plant is affiliated with C. bakeri and C. argillosa in habit and other characters, and is found in dry rocky soil. It is easily distinguishable from either of these species by its small flowers. It is somewhat anomalous in a genus of plants that lose their basal leaves usually before the flowers appear, and especially so in that it inhabits perhaps the driest and hottest region from which species of the genus have been reported. Euklisia valida (Greene) Wooton & Standley. Disaccanthus validus Greene, Leaflets 1: 225. 1906. Disaccanthus mogollonicus Greene, loc. cit. Disaccanthus luteus Greene, loc. cit. All these may be Streptanthus carinatus Wright, but that is described as having a purple calyx and petals. In our plants they are always yellow. 126 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. In the type of D. luteus the flowers are of a deeper yellow than in the plant of the Rio Grande region. Following his description of D. mogollonicus, Doctor Greene says: “All white-flowered material from New Mexico, from Las Cruces to the upper Gila, belongs here.” Asa matter of fact, none of our plants have white flowers, but in all both the calyx and corolla are a pale, clear yellow. This fades on drying, so that the flowers often appear white in old specimens. Lesquerella lata Wooton & Standley, sp. nov. A cespitose perennial, 10 cm. high or less, from a stout, woody root; stems Slender, ascending or prostrate, rather densely lepidote-stellate, with but few, rather distant leaves; these spatulate or oblanceolate, obtuse, thick, tapering at the base into a Slender petiole, lepidote-stellate on both surfaces, 138 to 30 mm. long, 4 to 9 mm. wide, the basal leaves longer and on longer petioles; racemes 3 to 4 em. long, rather densely many-flowered ; pedicels slender, spread- ing, or reflexed in age, 7 mm. long; sepals oblong, 3.5 mm. long; petals 6 or 7 mm. long, bright yellow; capsules oblong to obovate, thinly lepidote-stellate, 3 mm. high, short-stipitate, much surpassed by the very slender style, Type. in the U. S. National Herbarium, no. 563020, collected somewhere in the Lincoln National Forest in 1903, by Mr. Fred G. Plummer. Here, too, belong young specimens collected by E. O. Wooton on White Mountain Peak, July 6, 1895, at an altitude of about 2,880 meters. From our other species with pubescent capsules this is easily separated by its broad Jeaves, small capsules, and very long styles. Lesquerella pinetorum Wooton & Standley, sp. nov. Perennial from a rather slender root; stems clustered, slender, ascending, 10 to 20 cm. high, lepidote-stellate but not very densely so, leafy, the leaves not crowded ; basal leaves spatulate, obtuse, long-petioled: cauline leaves spatulate or oblanceolate, obtuse or acutish, entire or slightly undulate, tapering gradu- ally at the base into a winged petiole, the whole leaf 20 to 30 mm. long, 3 to 7 mm. wide; racemes long, many-flowered, dense ; pedicels about 8 mm. long, erect, stout; petals bright yellow, obovate, clawed, 6 or 7 mm. long; sepals oblong, 4 mm. long, densely lepidote-stellate; capsules nearly spherical, 3.5 mm. long, not compressed, glabrous, not stipitate; style slender, slightly longer than the cap- sule; seeds about 5 in each capsule. Type in the U. S. National Herbarium, no. 561847, collected on a dry hillside under pine trees at Gilmores Ranch on Eagle Creek in the White Mountains, August 25, 1907, Wooton & Standley (no. 3460). Transition Zone; altitude about 2220 meters. The plant was also collected at Gilmores Ranch, July 29, 1901, by E. O. Wooton, and apparently the same is a specimen from the top of White Mountain Peak, collected by E. O. Wooton, August 16, 1897. The species is nearest L. fendleri, but is not nearly so densely pubescent—in- deed, the plants appear green rather than whitish; the leaves, too, are much broader and thinner, and the capsules are smaller. Lesquerella praecox Wooton & Standley, sp. nov. A dense, cespitose, branched perennial, 4 to 6 cm. high, from a thick, woody root ; leaves numerous, much crowded upon the short, stout stems; leaves linear- oblanceolate, 25 to 80 mm. long, 2.5 mm. wide or less, acutish, thick, white with a dense, lepidote-stellate pubescence, attenuate to the base; pedicels few, axillary or all arising from the tips of the branches, never racemose, 20 mm. long, erect, usually shorter than the leaves; sepals linear, 5 mm. long, sometimes persistent ; petals bright yellow, obovate, 7 or 8 mm. long; capsules spherical or nearly so, not compressed, slightly channeled along the edge of the septum, glabrous, not Stipitate, 5 mm. high, WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 127 Type in the U. 8S. National Herbarium, no. 4869, collected in New Mexico in 1853 by J. M. Bigelow. ADDITIONAL SPECIMENS EXAMINED: Gallinas Mountains, August 27, 1904, Wooton; Cabra Springs, 1878, W. B. Pease. From L. fendleri, its nearest relative, this plant is at once distinguished by its lower, densely cespitose habit and its few pedicels which are surpassed by the leaves. The general appearance of the two is very different. Lesquerella rectipes Wooton & Standley, sp. nov. Perennial or biennial from a rather slender, woody root; stems slender, clustered, ascending or spreading, 12 to 18 cm. long, sparsely leafy, lepidote- stellate but not densely so; basal leaves narrowly oblanceolate or spatulate, obtuse; cauline leaves linear-oblanceolate, obtuse, 18 to 25 mm, long, lepidote- stellate on both surfaces; racemes loosely few-flowered, 4 cm. long or less; pedicels ascending, stout, 6 or 7 mm. long; sepals 4.5 mm. long, oblong-lanceolate, only sparingly lepidote; petals pale yellow, 7 mm. long, oblanceolate, obtuse; capsules broadly oblong to almost spherical, not compressed, 4.5 mm. long, not stipitate, very lightly lepidote-stellate; styles much longer than the capsules. Type in the U. S. National Herbarium, no. 4799, collected in New Mexico June 6, 1883, by C. C. Marsh (no. 81). The exact locality is not given on the label, but it was somewhere in the northwestern part of the State. ADDITIONAL SPECIMENS EXAMINED: Thirteen miles south of Atarque de Garcia, July 19, 1906, Wooton; along the banks of the Rio Grande 19 miles west of Santa Fe, alt. 1630 meters, May 31, 1897, Heller 3634. Our plant has been confused with Z. argentea, but that species has smaller, very densely pubescent capsules on strongly recurved pedicels. Sophia adenophora Wooton & Standley, sp. nov. A coarse, canescent annual, generally with a single erect stem 80 to 120 cm. high, branching rather freely above but not at the base; leaves pinnately or bipinnately divided into rather coarse, obtuse or acute, oblong-lanceolate seg- ments, the upper cauline leaves once pinnate with oblong, abruptly acute seg- ments, the lower leaves once or twice pinnate, at least some of the segments broader; inflorescence an elongated, terminal raceme, 20 cm. long or more in fruit, glandular-pubescent throughout; flowers rather pale yellow; sepals 2 to 3 mm. long, oblong, obtuse, yellow, glandular; petals oblanceolate, spatulate, yellow, slightly longer than the sepals; siliques linear, 12 to 20 mm. long, acute, spreading or ascending; seeds crowded as if in one row; pedicels as long as the pods or longer, slightly ascending or divergent. Type in the U. S. National Herbarium, no. 562504, collected by E. O. Wooton July 13, 1900, at the Head and Wilson Ranch south of Mule Creek, in north- western Grant County. ADDITIONAL SPECIMENS EXAMINED: Rio Frisco, July 25, 1900, Wooton; Tula- rosa Creek, Socorro County, July 14, 1906, Wooton; Reserve, July 9, 1906, Wooton. This species is most closely related to S. obtusa Greene, from which it differs in having longer, more divergent pedicels, larger petals, and strongly glandular inflorescence. It is fourd in the Upper Sonoran Zone. Sophia glabra Wooton & Standley, sp. nov. A slender, erect, sparingly branched, canescent annual, 30 to 60 cm. high; stems more or less purplish throughout; leaves and stems covered with a thick coat of short, branched hairs, not at all glandular; leaves all bipinnately divided into small, oblong, obtuse or acute segments 1 to 3 mm. long, some of these again lobed; racemes elongated in fruit; flowers small, 1 to 2 mm. long; 128 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. sepals purplish, stellate-pubescent ; petals yellow, as long as the sepals, narrowly oblanceolate or spatulate; fruiting racemes, rachis, pedicels, and siliques perfectly glabrous; pedicels about 10 mm, long, divergent or slightly ascending; siliques 5 or 6 mm. long, oblong, acute at both ends; seeds in two rows. Type in the U. 8. National Herbarium, no. 564228, collected in the Organ Mountains, March 21, 1907, by E. O. Wooton and Paul C. Standley. The speci- mens were collected in an arroyo at the foot of the mountains not far from Van Pattens Camp. They grew among the rocks and about the edges of cliffs. ADDITIONAL SPECIMENS EXAMINED: Van Pattens, April 25, 1895, Wooton; Fil- more Canyon, April 18, 1903, Wooton; Bishops Cap, March 30, 1895, Wooton; Bishops Cap, 1908, Wooton 3815. The species is most closely related to 8. halictorum, but is distinguished by its habit, its strictly glabrous racemes, and the slightly smaller fruit. So far as our material shows, it is restricted to the Organ Mountains, where it is the common and almost the only Sophia, though it is not nearly as abundant as the nearly related species of the adjacent Mesilla Valley. It occurs in the Upper Sonoran Zone. Thelypodium vernale Wooton & Standley, sp. nov. Probably biennial, glabrous, about 40 em. high; stems slender, branched throughout, glaucous, purplish near the base, the branches strongly ascending; cauline leaves triangular-lanceolate, attenuate, 385 to 50 mm. long, slightly glaucous, entire, somewhat undulate, auriculate-clasping at the base, the lobes obtuse, 5 to 7 mm. long; pedicels ascending, slender, about 5 mm. long; sepals narrowly oblong, obtuse, 2.5 mm. long, green or tinged with purple; petals white, slightly tinged with purple, narrowly oblong, tapering gradually toward the base, the whole 5 mm. long or less; pods slender, 40 to 60 mm. long, some- what divergent, arcuate; septum without a midrib; style truncate, not bilobate. Type in the U. S. National Herbarium, no. 690257, collected in the low moun- tains west of San Antonio, Socorro County, April 14, 1908, by E. O, Wooton (no. 3847). . This slightly resembles 7, sagittatum, but the flowers are much smaller, the leaves more acute, and the whole plant much smaller and more slender. CAPPARIDACEAE. Peritoma breviflorum Wooton & Standley, sp. nov. Slender annual, 40 to 50 cm. high, simple at the base, above with numerous ascending or spreading branches; stems green, glabrous; leaflets 3, elliptic or narrowly elliptic-oblanceolate, green, glabrous, abruptly acute; terminal racemes very short, 4 to 6 cm. long, slender, villous: bracts small, 3 mm. long or less, nearly linear, attenuate; pedicels slender, 4 to 7 mm. long; calyx united at the base, persistent, the lobes narrowly triangular, acute, yellowish green: petals deep yellow, small, 2.5 to 4 mm. long, oblong-obovate, obtuse, entire, abruptly contracted at the base into a very short claw; stamens 6, only slightly exceeding the petals; capsules oblong, 15 to 20 mm. long, acutish, glabrous, torulose, on a slender stipe 4 mm. long; seeds 6 or fewer, ovoid, 3.5 mm. long, brownish, irreg- ularly tuberculate; style very short, about 0.5 mm. long. Type in the U. S. National Herbarium, no, 686249, collected on the dry, stony hills about Shiprock, July 25, 1911, by Paul C. Standley (no. 7282). Upper Sonoran Zone; altitude about 1425 meters. The plant is fairly abundant in the region along the low mesas bordering the valley of the San Juan River. It is associated with various species of Atriplex and other plants characteristic of alkaline situations. It is similar to P. luteum, but the flowers are only half as large, the petals relatively broader and with WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 129 shorter claws, as well as of a deeper color, the inflorescence villous instead of glabrous, and the capsules smaller and on much shorter stipes. HYDRANGEACEAE. Fendlera falcata Thornber, sp. nov. An erect shrub, 1.5 to 2 meters tall, with grayish, furrowed bark; twigs of one year’s growth glabrous or nearly so, somewhat shining, reddish; leaves 15 to 40 mm. long, 5 to 7 mm. wide, nearly sessile, tapering at the base and apex, lanceolate to narrowly lanceolate, more or less falcate, shining above and below, glabrous, or sparsely strigose beneath, the margins revolute; sepals, pedicels, and hypanthium glabrous or slightly pubescent, glabrate at maturity; sepals lanceo- late to ovate-lanceolate, 8 to 10 mm. long, about three-fifths the length of the capsule; petals 17 to 22 mm, long, 11 to 18 mm. wide, tapering into a long claw, the margins erose; anthers about 4 mm. long; capsules 11 to 16 mm. long, 7 to 8S mm. in diameter, conical, very gradually tapering upward. Type in the U. S. National Herbarium, no. 686760, collected in the Tunitcha Mountains on the Navajo Reservation in August, 1911, by Paul C. Standley (no. 7806). ADDITIONAL SPECIMENS EXAMINED: CoLorapo—Dolores, alt. 2,200 meters, June 15, 1902, Crandall; Cerro Summit, alt. 2,420 meters, 1901, Baker 165; Bayfield, 1907, Cary 174; Mancos, 1898, Baker 393. Aritzona—Carrizo Mountains, 1911, Standley 7406; Fort Apache, July 28, 1905, Thornber. New Mexico—Without locality, 1869, Palmer; Cedar Hill, alt. 1,900 meters, 1911, Standley 7971. Fendlera tomentella Thornber, sp. nov. Shrub 1 to 1.5 meters high, with dark gray, furrowed branches, and some- what shining, reddish or straw colored, puberulent twigs, the short flowering ones with 2 to 5 pairs of leaves; leaves lanceolate to narrowly lanceolate, 15 to 80 mm. long, 4 to 7 mm. wide, dull green and hispid or hispidulous above, strongly 8-nerved beneath and strigose and tomentose, appearing hoary; calyx, hypan- thium, and pedicels permanently hirsute, the lanceolate sepals extending to beyond the middle of the capsule; flowers not seen; capsules 10 to 12 mm. long, 6 to 6.5 mm. thick, gradually tapering above. Type in the U. S., National Herbarium, no. 497677, collected in the canyon of the Blue River near Coopers Ranch, Graham County, Arizona, in 1905, by Walter Hough (no. 470). ADDITIONAL SPECIMENS EXAMINED: New Mexico-——-Mangas Springs, alt. 1,450 meters, 1903, Metcalfe 89; Hurrah Creek, September 25, 1853; Bigelow ; Embudo, alt. 1,760 meters, 1897, Heller 8513; Animas Mountains, alt. 2,000 meters, 1908, Goldman 1385. CoLtorapo—lLos Pinos, 1899, Baker 367. Arizona—Santa Cata- lina Mountains, alt. 900 meters, August 20, 1903, Jones. Fendlerella cymosa Greene, sp. nov. A much branched, rather erect, low shrub, 40 to 60 em. high, with scaly root- stocks; young stems grayish or nearly white; whole plant more or less strigil- lose with nearly colorless hairs, some of those on the lower surfaces of the leaves with white, papilliform bases; leaves numerous, nearly sessile, narrowly lanceolate to oblanceolate, 10 to 25 mm, long, 1.5 to 5 mm. wide, 3-nerved, acute, with ciliate, slightly revolute margins; cymes several to many-flowered, termi- nating the leafy branches; hypanthium turbinate, pubescent like the lanceolate or oblong-lanceolate sepals, these 4 mm. long; petals white, oblong-elliptic, 3 to 4 mm. long; filaments of the shorter stamens dilated to near the base, those of the longer ones with a well-defined, constricted neck above; capsules narrowly elongated. 130 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. : Type in the U. 8S. National Herbarium, no. 45788, collected in the Huachuca Mountains of southern Arizona, July 7, 1884, by C. G. Pringle. ADDITIONAL SPECIMENS EXAMINED: TEXAS—Guadalupe Mountains, 1901, Bailey 699. New Mexico—Canyon on the east side of San Luis Mountains, 1898, Mearns 2253; San Luis Mountains, 1892, Mearns 530; Organ Mountains, June 10, 1906, Standley; Organ Mountains, alt. 2,000 meters, 1897, Wooton 459. ARIzONA—Chiricahua Mountains, alt. 1,666 meters, 1907, Blumer 1725; Hua- chuca Mountains, September, 1882, Lemmon, SAXTIFRAGACEAE, Heuchera pulchella Wooton & Standley, sp. nov. Acaulescent, with a very thick rootstock; flowering branches scapiform, naked, 7 to 10 em. high, minutely glandular; leaf blades cordate, 13 mm. wide or less, rather deeply lobed, the teeth broadly ovate, acuminate, bristle-tipped ; blades glabrous above, glandular beneath, bright green, ciliate; inflorescence secund, dense, most of the flowers solitary but occasionally two together ; hypan- thium densely glandular, slightly villous, campanulate, together with the oblong, obtuse sepals 4 mm. long; sepals purplish; petals linear-filiform, not surpassing the sepals; stamens slightly exceeding the sepals, Type in the U. 8. National Herbarium, no. 690250, collected from crevices of rocks on the summit of the Sandia Mountains, August 4, 1910, by E. O. Wooton. Another sheet in the National Herbarium was collected in New Mexico in 1869 by Dr. Edward Palmer. This is nearest H, nana, but has a larger hypanthium and calyx, a more densely glandular hypanthium, a brighter colored calyx, shorter petals, and a denser inflorescence; the ciliw of the leaf margins, too, are much shorter and fewer. ROSACEAE. Oreobatus rubicundus Wooton & Standley, sp. nov. Shrub about 1 meter high, with spreading branches; stems slender, the bark soon exfoliating in thin layers; young branches reddish, very finely and closely pubescent ; leaves 30 to 40 mm. long and about as wide, round-ovate in outline, cordate at the base or truncate, the sinus always broad and open, conspicuously 3-lobed, the lobes obtuse and often again lobed, crenate-dentate, the teeth ovate, usually obtuse and abruptly acuminate, glabrous above or nearly so, sparingly pubescent beneath along the veins, thin, bright green, the veins inconspicuous and scarcely reticulate; petioles slender, as long as the blades or considerably shorter, finely pubescent, reddish; peduncles solitary, short, 5 to 15 mm. long, stout, finely pubescent and glandular; sepals at anthesis about 10 mm. long, lanceolate, acuminate, in age becoming 15 mm. long and ov ate, with long, linear tips, finely pubescent on both surfaces, with numerous reddish glands on the outer face; petals white, fugacious, 10 to 15 mm. long, broadly oblong; fruit small, with few, small, nearly dry drupelets. Type in the U. S. National Herbarium, no, 560937, collected at Van Pattens Camp in the Organ Mountains, June 9, 1906, by Paul C. Standley. The shrubs are found in the Upper Sonoran Zone on the faces of cliffs and in deep rocky canyons. ADDITIONAL SPECIMENS EXAMINED: Organ Mountains, September 17, 1898, May 15, 1892, Wooton; Van Pattens, August 29, 1894, July 27, 1902, Wooton. This plant near O. neomewxicanus, but has smaller leaves, flowers, and fruit, and nearly glabrous leaves, and the pubescence of the branches and petioles is ~ hs WOOTON AND STANDLEY—NEW PLANTS FROM NEW’ MEXICO. 1381 fine and appressed instead of loosely spreading. We have seen it only from the Qrgan Mountains. Rosa adenosepala Wooton & Standley, sp. nov. Branches smooth, reddish brown, armed with few, slender, straight spines; stipules broad, acute, densely glandular and soft-pubescent; petioles soft- pubescent, with numerous stalked glands; leaflets cn short petiolules, obovate, obtuse, broadly cuneate at the base, sharply and deeply incised-serrate two- thirds of the way to the base or more, dull green, glabrous above or nearly so, beneath densely glandular-puberulent, of about the same color on both sur- faces; flowers in clusters of 3 to 5 at the ends of the branches, on glandular peduncles about 1 em. long; hypanthium glabrous; sepals 15 to 20 mm. long, linear-lanceolate, with very long narrow tips scarcely dilated at the apex, densely glandular-bristly with red glands 1 or 2 mm. long, villous along the margins; petals about 15 mm, long; sepals reflexed after anthesis; fruit not seen, Type in the U. S. National Herbarium, no. 306499, collected along the Pecos River 8 miles east of Glorieta, San Miguel County, June 9, 1897, by A. A. and BE. Gertrude Heller (no. 3674). Altitude 1,950 meters. Nearest &. fendleri, perhaps, but readily distinguished by the densely glandu- lar-bristly calyx lobes. Rosa hypoleuca Wooton & Standley, sp. nov. Stems bright reddish brown, more or less glaucous when young, densely armed with very slender, straight, short spines; stipules large, narrow, acute, with glandular margins; petioles with rather few stalked glands, otherwise glabrous; leaflets usually 9, elliptic-oblong or oblong-ovate, obtuse, rounded at the base, coarsely but not deeply incised-serrate, glabrous, strongly glaucous beneath, dull green above, 13 to 20 mm. long; flowers in clusters at the ends of the ascending branches on short, glabrous peduncles; hypanthium glabrous; sepals narrowly lanceolate with long, linear tips, about 15 mm. long, glabrous below, glandular toward the tips, entire or with a few subulate lobes, tomentu- lose along the margins; petals obovate, deeply notched, bright pink, 15 to 20 min. long; fruit not seen. Type in the U. 8. National Herbarium, no. 497843, collected near Kingston, Sierra County, June 1, 1904, by O. B. Metcalfe (no. 940). Altitude 1,980 meters. This is as closely related to R. fendleri as to any species. It differs, how- ever, in the perfectly glabrous leaflets, strongly glaucous beneath, Here we have placed Standley’s 4031, collected along Winsor Creek, June 29, 1908, altitude 2,520 meters. This has larger flowers and leaflets than the type but otherwise seems indistinguishable. MALACEAE. Amelanchier goldmanii Wooton & Standley, sp. nov. A shrub 1.5 meters high or less, with spreading branches; bark on the older stems grayish, on the younger ones dark reddish brown; bud scales reddish, sparingly villous; leaves almost perfectly rotund, on petioles 8 to 15 mm. long, 30 to 40 mm. in diameter, rounded or cordate at the base, with rounded serrate teeth reaching almost to the base; petioles and blades glabrate, probably more or less tomentulose when young; flowers not seen; fruit bluish black, spherical, 7 mm. in diameter, in terminal racemose clusters of several fruits. Type in the U. S. National Herbarium, no. 562614, collected by E. A. Goldman in Copper Canyon in the Magdalena Mountains, September 3, 1909, at an alti- tude of 2,550 meters. 132 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. ADDITIONAL SPECIMENS EXAMINED: Mogollon-Magdalena Road, in the Mogollon Mountains, about 15 miles east of Mogollon, August 8, 1900, Wooton. AMYGDALACEAE, PADUS. Since all but one of our seven species of Padus appear to be new it seems desirable to publish a key to them along with the descrip- tions. KEY TO THE NEW MEXICAN SPECIES. Calyx persistent in fruit. Young branches densely tawny-pubescent; young fruit pubescent ~---------------------_~__----_--- eee P.rufula, Young branches and fruit glabrous____~_-~-_------ P, virens. Calyx deciduous soon after anthesis. Plants glabrous throughout___--_--_--__-____-__--- P, meianocarpa? Plants pubescent on the peduncles, petioles, and lower sur- face of the leaves. Leaves not glaucous beneath at maturity, of about the same color on both surfaces___-----.----____ P. punricea, Leaves whitish beneath at maturity. Pedicels longer than the fruit, slender; seeds 8 to 10 mm. in diameter___._-----_-__________ P. mescaleria. Pedicels shorter than the fruit, stout; seeds 7 mm. in diameter or less. Pedicels glabrous; racemes slender: leaves elliptic, narrowed at the base; buds narrowly lanceolate in outline________ P. calophylla, Pedicels pubescent; racemes stout; leaves ob- long’ to ovate or obovate, rounded to subcordate at the base; buds ovoid__ P. valida. Padus rufula Wooton & Standley, sp. nov. Branches slender, grayish brown, with numerous small, gray lenticels; young branches densely soft-pubuescent with reddish brown hairs; petioles pubescent, slender, one-fourth as long as the blades or less; leaf blades thin, elliptic, acute, 40 or 50 mm. long, acute at the base, finely and very sharply serrate, dull green and glabrous above, pale beneath, glabrous except for an abundant persistent, tawny tomentum along the midvein; racemes slender, many-flowered, pubescent at the base, glabrous above; pedicels stout, 4 mm. long or less; flowers not seen; hypanthium and sepals persistent; fruit sessile in the hypanthium, globose, 8 to 10 mm. in diameter, pubescent when young, glabrate at maturity. Type in the U. 8. National Herbarium, no. 563898, collected on the West Fork of the Rio Gila, August 6, 1900, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: NEw Mexico—Mogollon Mountains, August, 1881, Rusby 119. Arizona—Santa Rita Mountains, 1881, C. G. Pringle. This is one of the most distinct species of the genus. It is related to P. salicifolia of Mexico, but is distinguished at once by the abundant tomentum of the leaves, the densely pubescent branchlets, and the pubescent fruit. * Padus melanocarpa (A. Nels.) Shafer in Britt. & Shaf. N. Amer. Trees 504. 1908. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 183 Padus virens Wooton & Standley, sp. nov. A tree 7 meters high or less, with a smooth, reddish brown trunk and slender, slightly drooping branches; young branchlets and racemes glabrous; leaves ovate, obovate, or oval, 5 or 6 cm. long, acutish or slightly acuminate, truncate or usually rounded at the base, thin, bright green, shining, of about the same color on both surfaces, finely crenulate-serrate with appressed teeth; petioles slender, glabrous, one-third to one-half as long as the blades; leaves glabrate in age, when young with a rather abundant tawny tomentum along the midvein, some of this persistent at maturity; flowers in slender, loose racemes; pedicels slender, 5 or 6 mm. long; hypanthium glabrous, saucer- shaped, 5 mm. broad, the lobes triangular, acute; petals 4 mm. long, obovate; fruit black, glabrous, globose, 7 or 8 mm. in diameter, sessile in the hypan- thium. Type in the U. 8S. National Herbarium, no. 560917, collected at Van Pattens Camp in the Organ Mountains, June 9, 1906, by Paul C. Standley. ADDITIONAL SPECIMENS EXAMINED: Organ Mountains, alt. 1,710 meters, 1897, Wooton 123; Van Pattens, May 14, 1895, April 29, 1899, Wooton; Organ Moun- tains, 1893, Wooton 1108; Rio Frisco, August 9, 1900, Wooton; Sycamore Creek, August 13, 1902, Wooton; Pine Cienaga, July 17, 19900, Wooton; San Fran- cisco Mountains, July 27, 1900, Wooton; Alizo near Kellys, July 24, 1900, Wooton; Holts Ranch, July 20, 1900, Wooton; near Joseph, July 10, 1906, Wooton; Tularosa Creek, August 6, 1901, Wooton; Cloudcroft, July 18, 1899, Wooton; Burro Mountains, 7 miles southeast of Leopold, alt. 2,250 meters, 1908, Goldman 1521; Burro Mountains, 5 miles southeast of Leopold, alt. 1,950 meters, 1908, Goldman 1510; Florida Mountains, alt. 1,800 meters, 1908, Gold- man 1500. This has passed as P. capuli, a Mexican species, from which our plant differs in its broader leaves with the midvein tomentose instead of glabrous. In the Organ Mountains this species occurs in abundance in the opening of the canyon in which Van Pattens Camp is located. It is known from only this one locality in the range, where it grows along with Quercus grisea and Q. ari- zonica. Attempts have been made to use the trees as stocks for grafting va- rious fruits but they have been unsuccessful. The material from the western part of the State may represent a different species. It is usually much smaller, only a tall shrub, and its leaves are narrower, thicker, not so bright a green, and on shorter petioles. Padus pumicea Wooton & Standley, sp. nov. Branches stout, numerous, ascending, dark reddish brown, the younger ones finely pubescent; petioles mostly one-fifth as long as the blades, sometimes longer, finely pubescent, each bearing two large glands; leaf blades obovate or oblong, averaging about 45 mm. long, rounded or truncate at the base, acute or acuminate, finely serrulate to the base with incurved teeth, bright green and glabrous above, finely pubescent and of the same color beneath, often becoming glabrate; flowers not seen; racemes stout, few-fruited, finely pubescent near the base, glabrous above; pedicels stout, glabrous, much shorter than the fruit; seeds globose, 6 or 7 mm. in diameter. Type in the U. S. National Herbarium, no. 563903, collected at the “ Craters,” Valencia County, July 28, 1906, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Mountains south of Canjilon, August 17, 1904, Wooton 2706. From al our pubescent species this differs in having the leaves of about the same color on both surfaces. The fruits, too, are very few and the pedicels remarkably short. The branches are very densely furnished with leaves, so that in general appearance this is unlike any of our other chokecherries. 134 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. The type specimen is without fruit, but one of the same collection in the herbarium of E. O. Wooton is well fruited, as is the specimen from Canjilon. Padus mescaleria Wooton & Standley, sp. nov. Branches slender, grayish brown, the younger ones finely and sparing!y pubescent; buds ovoid, small; petioles about one-third as long as the blades, finely pubescent, with several glands above the middle; blades oblong to nar- rowly obovoid, acute or slightly acuminate, usually rounded at the base, glabrous and dark green above, strongly glaucous and finely pubescent be- neath, rather thick, sharply and evenly serrate to the base; flowers not seen; racemes slender, loosely few-fruited, the rachis glabrous; pedicels slender, noticeably longer than the fruit, glabrous, straight; seeds 7 to 9 mm. in diameter, scarcely at all flattened. Type in the U. S. National Herbarium, no. 690233, collected on Tularosa Creek near the Mescalero Agency, August 6, 1901, by E. O. Wooton. The most distinctive features of this are the long pedicels, glabrous racemes, large seeds, and rather narrow, deep green leaves strongly glaucous beneath. Padus calophylla Wooton & Standley, sp. nov. Branches stout, dark gray, the younger ones densely and finely pubescent; buds lanceolate to narrowly elliptic in outline; petioles slender, reddish, one- fourth as long as the blades, very finely pubescent, usually with a pair of glands just below the blade; blades elliptic, acute, acutish at the base, 45 to 60 mm. long, about 30 mm. wide, dull pale green above and glabrous, decidedly paler and finely pubescent beneath, especially on the veins, rather thick and coriaceous, very finely and inconspicuously serrulate; veins prominent, the midvein reddish; flowers not seen; racemes slender, the rachis slender, dark reddish purple; pedicels rather slender, shorter than the fruit, often curved, glabrous; fruit about 9 mm. in diameter, abundant and persistent, the seed flattened, 5 mm. in diameter. Type in the U. S. National Herbarium, no. 562677, collected 5 miles west of Chloride, October 12, 1909, by E. A. Goldman (no. 1768). The foliage is different from that of any other species, the leaves being especially handsome because of their thick texture and of their pale green coloring, which contrasts with the red of the veins and petioles. The seeds are remarkably small; the fruits are densely clustered and persistent, instead of scattered and soon falling as in most of the related species. Padus valida Wooton & Standley, sp. nov. Branches very stout, dark reddish brown, all the younger ones densely but very finely pubescent; buds large, ovoid; petioles less than one-fourth as long as the blades, stout, densely pubescent; blades obovate, oval, or oblong, usually 50 to 70 mm. long, rounded or subcordate at the base, acute or more often abruptly acuminate, thick, dull green and glabrous above, glaucescent and pubescent beneath, inconspicuously serrulate with sharp, nearly subulate, inflexed teeth; racemes very numerous, stout, densely pubescent, often re- eurved, bearing only a few fruits; pedicels stout, pubescent, much shorter than the fruit; flowers not seen; fruit 10 or 12 mm. in diameter; seeds flattened, about 7 mm. in diameter. Type in the U. S. National Herbarium, no. 498048, collected in canyons near Kingston, Sierra County, August 24, 1904, by O. B. Metcalfe (no. 1248). Altitude 1980 meters. ADDITIONAL SPECIMENS EXAMINED: Copper Canyon, alt. 2700 meters, Septem- ber 8, 1909, Goldman 1675; Copper Canyon, alt. 2250 meters, September 3, 1909, Goldman 1676; Hop Canyon, May 13, 1895, Herrick 535. A most striking species, characterized by its stout branches and racemes, large fruits, and densely pubescent pedicels and racemes. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 135 MIMOSACEAE. Morongia occidentalis Wooton & Standley, sp. nov. Stems prostrate, stout, striate, densely and finely puberulent, armed with very few, distant, recurved prickles; leaves bipinnate, with 5 or 6 pairs of pinne; petioles slender, about 4 cm. long, striate, finely puberulent, armed with few stout prickles; leaflets oblong, about 4 mm. long, acute or acutish, thick, smooth, not nerved, glabrous or nearly so; flowers pink, in dense heads; calyx glabrous or nearly so, the thin lobes ovate, acute; peduncles 2 to 6 cm. long, puberulent, with few weak prickles or none; pods slender, 7 to 9 cm. long, 2 or 3 mm. wide, flattened, armed on the sides with long, slender, sparingly puberulent prickles, usually naked on the margins or with prickles of the same kind as on the sides, narrowed at the base, bearing at the apex a stout beak 5 to 7 mm. long. ° Type in the U. 8. National Herbarium, no. 660612, collected near Nara Visa, July 4, 1911, by Mr. Geo. L. Fisher (no, 190). Also collected near Nara Visa, August 17, 1910, by Mr. Fisher (no. 58). Most closely allied, perhaps, to M. angustata, but distinguished by the flat, puberulent pods armed with but few prickles, and by the few prickles of the abundantly pubescent stems. CAESALPINIACEAE. Chamaecrista rostrata Wooton & Standley, sp. nov. A slender annual, 20 cm. high or less, simple at the base, sparingly branched above; stems herbaceous, reddish, puberulent; leaflets 10 or 12, narrowly ob- long, rounded at the apex, very shortly mucronulate, 8 or 10 mm. long, glabrous ; petiolar gland oblong, small, short-stipitate ; stipules linear-lanceolate, long- attenuate; peduncles few, supra-axillary, arcuate, 1-flowered; petals bright yellow, 12 mm. long; sepals one-half to two-thirds as long as the petals, lanceolate, membranaceous; pods 25 to 35 mm. long, 5 mm. wide, appressed- pubescent, ending in a beak 2 to’3 mm. long. Type in the U. S. National Herbarium, no. 660032, collected in sandy soil at Logan, October 5, 1910, by Mr. Geo. L. Fisher (no. 93). While related to C. fasciculata, this may be readily distinguished by the long beaks of the pods, the fewer leaflets neither acute nor conspicuously mucronate, the 1-flowered peduncles, and the shorter sepals. FABACEAE. Anisolotus greenei Wooton & Standley. Hosackia mollis Greene, Bull. Calif. Acad. 1: 185, 1885, not Nutt. Lotus mollis Greene, Pittonia 2: 143. 1890, not Balf. Anisolotus neomexicanus (Greene) Wooton & Standley. Lotus neomexicanus Greene, Pittonia 2: 141. 1890. Anisolotus nummularius (Jones) Wooton & Standley. Hosackia rigida nummularia Jones, Bull. Calif. Acad. II. 5: 683. 1895. Anisolotus puberulus (Benth.) Wooton & Standley. Hosackia puberula Benth. Pl. Hartw. 305. 1848. Lotus puberulus Greene, Pittonia 2: 142. 1890. Anisolotus trispermus (Greene) Wooton & Standley. Lotus trispermus Greene, Erythea 1: 258. 1893. 60541°—13——3 136 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Aragallus veganus (Cockerell) Wooton & Standley. Aragailus pinetorum veganus Cockerell, Torreya 2: 155. 1902. Apparently this is a most distinct species, found only on the tops of the highest mountains in the ranges lying between Santa Fe and Las Vegas. The assignment of the plant to rank as a subspecies under A. pinetorum was most unfortunate, since the plant is not at all closely related to that species, but finds its affinities among some of the alpine species of Colorado and Wyoming, Astragalus albulus Wooton & Standley, sp. nov. Plant 40 to 60 cm. high, much branched from a woody root, hirsutulous- anescent throughout; stipules whitish, membranaceous, large, connate opposite the leaves, the free tips broadly triangular; leaves 5 to 8 em. long, with very short petioles; leaflets 8 to 10 pairs, linear or narrowly oblong, 10 to 15 mm. long, acute or obtuse, glabrous above; flowers in loose, axillary racemes 10 to 15 em. long; peduncles 3 to 5 cm, long, the pedicels very short, subtended by white, membranous, ovate bracts half the length of the calyx; flowers dull yellow, tinged with purple; calyx cylindric, about 7 mm. long, the subulate teeth one- fourth as long, appressed-pubescent with both black and white hairs; corolla about 12 mm, long, the wings and banner only a little longer than the keel; mature fruit not seen, the very young pods 1-celled, several-seeded, with a very short stipe, slightly flattened dorso-ventrally at the base, acute, appressed- pubescent. . Type in the U. S. National Herbarium, no. 690252, collected in a canyon on the road to Zuni some distance south of Gallup, August 1, 1904, by E. O. Wooton. Upper Sonoran Zone. Without mature fruit the relationship of this plant can not be stated defi- nitely, but it is unlike any species known to us. Astragalus altus Wooton & Standley, sp. nov. Perennial, 40 to 60 em, high, with several slender, erect, sparingly branched stems, these indistinctly striate, sparsely pubescent with very short, appressed hairs; stipules triangular-lanceolate, acuminate, neither connate nor adnate to the expanded bases of the petioles; leaves 8 to 10 cm. long; petioles 1 cm. long or less; leaflets 10 to 14 pairs, elliptic to elliptic-oblong, when mature 8 to 10 mm. long, entire, acute or obtuse, appressed-pubescent beneath; flowers in axillary racemes 6 to 10 cm. long, the peduncles elongating slightly in fruit, the flowers crowded in a short raceme, not capitate; calyx yellowish, sparsely pubescent with appressed, black hairs, 5 mm. long including the small, acute teeth scarcely 1 mm. long; corolla bright yellow, hardly 10 mm. long, the wings and banner considerably surpassing the obtuse keel; pods (about half mature) 15 to 20 mm. long including the stipe (8 to 10 mm, long), oblong, acute, mem- branous, the lower suture somewhat sulecate, several-seeded, sparsely appressed- pubescent with minute black hairs. Type in the U. 8. National Herbarium, no. 690253, collected at Toboggan in the Sacramento Mountains, Otero County, July 31, 1899, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Upper Tularosa Creek, Sacramento Moun- tains, August 6, 1901, Wooton. At first glance this might be confused with A. rusbyi, which it resembles in a general way. It differs in having shorter and broader leaflets, crowded flowers on much shorter peduncles, and pods with longer stipes and much less inflexed sutures. Astragalus neomexicanus Wooton & Standley, sp. nov. Herbaceous perennial about 380 cm. high, with somewhat the aspect of Aragallus deflexus ; stems ascending, branched from the base, pubescent ; stipules WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 1387 narrowly triangular-lanceolate, about 10 mm. long, adnate to the petiole, not connate ; leaves 10 to 18 cm. long, with about 20 pairs of leaflets, villous through- out with weak, spreading hairs; leaflets elliptic to ovate-lanceolate, 10 to 12 mm. long, narrowed at the apex, acute, entire, glabrate above; flowers in elongated, several to many-flowered racemes 15 to 25 cm. long (including the peduncles) ; bracts linear-lanceolate, twice as long as the ascending pedicels (these 2 or 3 mm. long); calyx campanulate, slightly gibbous above, 7 to 9 mm. long, the subulate teeth about one-third as long, pubescent with black and white, loosely appressed hairs, the black ones more noticeable on the buds; corolla purple, about 20 mm. long, the banner and wings considerably surpassing the rounded keel; mature fruit not seen; very young pods 1-celled, several- seeded, thick-walled, terete, neither suture intruded, densely appressed-pubescent. Type in the U. S. National Herbarium, no. 690254, collected in James Canyon in the Sacramento Mountains near Cloudcroft, Otero County, July 23, 1899, by E. O. Wooton. Altitude about 2,550 meters. We hesitate to describe this species for which the mature fruit is lacking, but it is so distinct from anything else we have seen that the description is given and a name applied merely as a means of recording the plant. Lupinus argillaceus Wooton & Standley, sp. noy. Low annual of the Platyecarpos group; stem about 10 cm. high, with numerous lateral divaricnte branches as long or longer; whole plant softly silky villous, the pubescence being somewhat appressed and thus not conspicuous ; petioles 3 to 5 em. long: leaflets 5 to 7, 1 to 2 em. long, oblanceolate, obtuse or acute, about equally pubescent on both sides; flowers few, in contracted, head-like clusters at the ends of peduncles mostly shorter than the petioles of the adjacent leaves, with broadly ovate-lanceolate bracts 3 or 4 mm, long; calyx similar to that of L. kingii, the upper lobe deeply 2-cleft, the divisions lanceolate, acute, the lower lobe minutely 8-toothed at the apex, the toothing more or less obscured by the pubescence in dried material, the upper lobe about two-thirds as long as the lower, with minute bracts at the sinuses between the two lobes; corolla blue or white, the banner narrowly ovate, reflexed, shorter than the wings and the elongated, rather straight keel; stamens monadelphous, the lower five an- thers linear, almost as long as their filaments, the upper anthers minute; ovary with 2 ovules: fruit a short, rhombic-ovate, 2-seeded legume; seeds almost rotund, lenticular, whitish, without markings. Type in the U. S. National Herbarium, no. 498990, collected near Pecos, San Miguel County, at an altitude of 2,010 meters, August 17, 1908, by Paul C. Standley (no. 4974). The plants grew on the low hills about Pecos, especially in the deep gullies, in a heavy red clay soil. ADDITIONAL SPECIMENS EXAMINED: Near Pecos, alt. 2,010 meters, 1908, Stand- ley 4975; El Rito, August 17, 1904, Wooton. Standley’s specimens were distributed as ZL. kingii and are listed as that species in his report upon his 1908 collections in Mublenbergia.” Of the two numbers collected in the same locality one had white flowers and the other blue. Lupinus laetus Wooton & Standley, sp. nov. A slender perennial with few spreading or ascending branches; stems finely and sparingly appressed-pubescent ; petioles slender, about as long as the leaf- lets, these elliptic-oblanceolate, 45 to 60 mm. long, obtuse, mucronate, bright green and glabrous above, with a few scattered, appressed hairs beneath; raceme 15 to 20 cm. long, loosely few-flowered, «short-pedunculate; pedicels slender, ascending, 8 mm. long, sparingly pubescent ; calyx sericeous, the lobes about equal, broad, rather obtuse; corolla 12 mm. long, bright blue; fruit not seen. *5: 17-80. 1909. 1388 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Type in the U. 8S. National Herbarium, no. 562219, collected at Winter Folly in the Sacramento Mountains north of Clouderoft, August 18, 1899, by E. O. Wooton. Altitude about 2,700 meters. The plant is of the group of L. plattensis, but its bright blue flowers at once distinguish it. Lupinus aquilinus Wooton & Standley, sp. nov. A much branched, suffrutescent perennial, 60 to 150 cm. high; stems slender, spreading, sparingly and finely sericeous; petioles slender, about equaling the leaflets, these elliptic-oblanceolate, 35 to 45 mm. long, obtuse, mucronate, gray- ish green, glabrous above, finely sericeous beneath; racemes on short peduncles, dense, 5 cm. long or less, few-flowered; pedicels ascending, stout, sericeous, 4 min. long; calyx densely sericeous, scarcely gibbous, the two lobes almost equal, rather broad, acute, entire; corolla 12 mm. long, pale bluish and yellowish, the banner with a dark spot; pods 80 mm. long and 8 mm. wide or less, 3 or 4-seeded ; seeds 4 mm. long, greenish gray, finely splashed with dark green. Type in the U. S. National Herbarium, no. 562095, collected at Gilmores Ranch, on Eagle Creek, in the White Mountains, August 25, 1907, by E. O. Wooton and Paul C. Standley (no, 3613). Altitude 2,220 meters. ADDITIONAL SPECIMENS EXAMINED: Gilmores Ranch, August 15, 1897, Wooton 537; Sierra Grande, August, 1903, Howell 229, This is a grayish plant with numerous slender, spreading branches. It is rather abundant about Gilmores Ranch, growing on slopes in the shade of pines. It is of the L. plattensis group and is related to L. sierra-blancae, which grows not faraway. It is a lower plant, however, than L, sierrae-blancae, with more branches, fewer flowers, and shorter, obtuse leaflets. Lupinus sierrae-blancae Wooton & Standley, sp. nov. A tall perennial, 1 meter high or more, much branched: stems stout and somewhat succulent, finely white-pubescent, the pubescence loose; petioles equaling the leaflets, these 7 to 9, 50 to 70 mm. long, linear-elliptic, very acute, attenuate to the base, yellowish green, glabrous above, finely strigillose beneath or almost glabrous; racemes long (25 to 30 cm.), many-flowered, rather loose; pedicels divergent, stout, 10 mm. long, densely pubescent -with short, spreading hairs; calyx slightly gibbous, loosely pubescent, the upper lip longer than the lower, both narrow, entire, attenuate; corolla 12 mm. long, dull bluish tinged with yellow, the banner with a large darker spot; pods ascending, stout, 85 mm. long and 35 mm. wide or less, densely hirtellous, 5 to 7-seeded; mature seeds not seen. Type in the U, S. National Herbarium, no. 562220, collected on the lower part of White Mountain Peak, July 6, 1895, by E. O. Wooton. Altitude 2,340 meters, ADDITIONAL SPECIMENS EXAMINED: Gilmores Ranch, on Eagle Creek, July 14, 1895, Wooton; Gilmores Ranch, July 29, 1901, Wooton; Ruidoso Creek, July 5 1895, Wooton. The plant is known only from the open meadows or parks which are so numerous on the timbered slopes of White Mountain Peak. The largest of our New Mexican lupines, it is a conspicuous feature of the vegetation where it occurs. While related to L. plattensis, it may be distinguished by its larger, acute leaves and the much larger size of the plants. Petalostemum prostratums Wooton & Standley, sp. nov. Stems slender, prostrate, 60 cm. long or less, glabrous, with numerous orbicu- lar, reddish brown glands; leaves long-petioled, the leaflets mostly 7, glabrous, glaucous, glandular-dotted, cuneate-oblanceolate, nearly sessile, rounded at the WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 139 apex; racemes rather densely many-flowered, long-peduncled, about 5 cm. long; flowers sessile; calyx angled, glabrous, straw colored, with a few pellucid, yel- lowish glands, the lobes lanceolate, green, attenuate, silky-pubescent along the margins; corolla rose purple, about 8 mm. long; stamens 6; fruit glabrous, 3.5 mm. long, semiorbicular in outline. Type in the U. 8. National Herbarium, no. 370697, collected near Albuquerque in 1909 by Winnie Harward (no. 17). ADDITIONAL SPECIMENS EXAMINED: Near Belen, August 1, 1906, Wooton. This is so unlike all other species of the genus that it can scarcely be confused with any. The habit alone is sufficient to distinguish it. In general appearance it much more closely resembles certain species of Parosela, but the structure of the flower and number of stamens makes it impossible to place it with them. Phaseolus dilatatus Wooton & Standley, sp. nov. Perennial from a thickened root; stems long, slender, twining, glabrous or sparingly puberulent; stipules small, lanceolate; petioles slender, equaling or longer than the leaflets, these linear-lanceolate to triangular-ovate, entire, or dilated and with 2 small rounded lobes at the base, obtuse, bright green, scaberu- lous, ciliolate; peduncles slender, somewhat exceeding the leaves, 5 to 12 em. long; pedicels 4 mm. long or less, sparingly puberulent; calyx puberulent, with rounded lobes; bracts minute, elliptic-oblong; corolla 10 to 12 mm. long; pods about 20 mm. long and 5 mm. broad, stout, slightly curved, nearly glabrous, with a slender style 1.5 mm. long. Type in the U. S. National Herbarium, no. 138616, collected in the Mogollon Mountains August 30, 1881, by Dr. H. H. Rusby. Another specimen is in the herbarium of Dr. E. L. Greene, collected in the Burro Mountains in June, 1881, by Doctor Rusby. This suggests P. grayanus, but the leaflets are not lobed as in that species, the peduncles are shorter, the pod is nearly glabrous ana smaller, and the style is long and slender. Phaseolus grayanus Wooton & Standley, sp. nov. Phaseolus wrightit A. Gray, Pl. Wright. 2: 33. 1853, not A. Gray, op. cit. 1: 438. 1852. Perennial with long, slender, climbing stems, these sparingly puberulent; stipules small, triangular-lanceolate; petioles one-half to two-thirds as long as the leaflets; leaflets deeply 3-lobed, at least the terminal one, the lobes narrowly oblong or rhombic, blunt, bright green, thin, nearly glabrous, but puberulent along the veins; peduncles much longer than the leaves, 10 to 25 cm. long, slender, glabrous or puberulent, pubescent about the flowers; flowers few, dis- tant, on pedicels 5 mm. long or less; calyx puberulent, the lobes broadly rounded, ciliate; corolla purplish, 12 to 15 mm. long; pods broad, 25 to 30 mm. long and 8 or 9 mm. wide, curved, densely soft-pubescent, acute, with a very short, stout style. Type in the U. 8. National Herbarium, no. 232982, collected in the San Luis Mountains, September 5, 1893, by Dr. E, A. Mearns (no. 2124). ADDITIONAL SPECIMENS EXAMINED: San Luis Mountains, alt. 1,720 meters, 1893, Mearns 2584; Horsethief Canyon, near Fort Bayard, alt. 2,100 meters, Novem- ber 9, 1905, Blumer 162; Mogollon Creek, alt. 2,400 meters, July 18, 1903, Met- calfe 259; Mangas Springs, August, 1901, Metcalfe; 1851, Wright 952. The species also occurs in Arizona. Our plant has always been referred to P. wrightii. Doctor Gray, in Plantae Wrightianae, speaks of Wright’s specimen, stating that it differs from the Texas plant in certain particulars and that the description of that plant must be modified. The Texan plant, true P. wrightii, differs from ours in having 140 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. smaller, usually entire, more pubescent leaflets, much shorter peduncles, glabrate pods, and a long, slender style. Phaseolus metcalfei Wooton & Standley, nom. nov. Phaseolus retusus Benth. Pl. Hartw. 11. 1839, not Moench. We have assigned the specific name in honor of Mr. J. K. Metcalfe, late of Mangas Springs, New Mexico. Mr. Metcalfe was the first to introduce this plant into cultivation, and it has been popularly known as the Metcalfe bean. It has proved of some value as a forage plant in the Southwest and has been treated of in some of thie Department of Agriculture publications under this name. Phaseolus tenuifolius (A. Gray) Wooton & Standley. Phaseolus acutifolius tenwifolius A. Gray, Pl. Wright. 2: 83. 1853. In P. acutifolius the leaflets are triangular-lanceolate or ovate, not much longer than the peduncles, while in P. tenuifolius they are elongated-linear to linear-oblong, and about twice as long as the peduncles. The latter, too, is usually a much larger, taller plant. Psoralea megalantha Wooton & Standley, sp. nov. A low perennial, 10 cm. high or less; stems very short; petioles equaling or twice as long as the leaflets, sericeous, also with a few spreading hairs; leaflets usually 6, obovate, 27 mm. long or less, cuneate at the base, rounded at the apex, densely sericeous beneath, sparingly sericeous above, dull green; peduncles stout, 20 mm. long or shorter, sericeous; bracts lanceolate or lance‘ovate, 11 mm. long or less, acute or somewhat acuminate, present only at the base of the in- florescence; flowers rather few, nearly capitate, on pedicels 8 mm. long; calyy about 18 mm. long, hirsute, the lobes nearly equal, linear, acute, equaling or shorter than the tube; corolla 20 mm. long. Type in the U. 8. National Herbarium, no. 368979, collected at Aztec, May 18, 1899, by C. F. Baker (no. 440). ''The collection was distributed as P. mephitica S. Wats. and it is related to that species. The flowers, however, are twice as large in our plant, the intlorescence subcapitate instead of elongated, and the pubescence mostly ap- pressed instead of spreading or retrorse. Robinia rusbyi Wooton & Standley, sp. nov. Shrub with stout, nearly glabrous, reddish brown branches; spines stout, straight, 15 mm. long or less; rachis of the leayes slender, minutely puberulent or glabrate; leaflets oval or broadly oblong, rounded and mucronate at the apex, rounded or slightly narrowed at the base, grayish green, glabrous above, minutely strigillose beneath; racemes many-flowered, short-peduncled, pubescent ; pedicels stout, densely glandular-pubescent; corolla 20 mm. long or more; calyx lobes ovate, acute; pods 45 to 85 mm. long, 18 mm. broad or less, glabrous, purplish. Type in the U. S. National Herbarium, no. 690288, collected on the Mogollon Road 15 miles east of Mogollon, August 8, 1900, by IE. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Eagle Peak, August 2, 1900, Wooton; Burro Mountains, alt. 2,250 meters, 1903, Metcalfe 189; Deep Creek, August 9, 1900, Wooton; head of Carrizo Creek, Mescalero Reservation, alt. 2,220 meters, 1903, Plummer. The last specimen cited may not belong here, for the leaflets are narrower, longer, and acute. It certainly is not Robinia neomexicana. R. rusbyi dif- fers from that species conspicuously in its glabrous fruit and merely glandular- pubescent peduncles and pedicels. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 141 Dr. H. H. Rusby seems to have been the first to observe this shrub and called our attention to it. He collected it somewhere about the Mogollon Mountains in 1880 or 1881. Trifolium longicaule Wooton & Standley, sp. nov. Perennial; stems slender, reclining, glabrous, much branched, 60 cm. long or less; petioles slender, several times as long as the leaflets; stipules narrow, attenuate, entire or nearly so; leaflets obovate to elliptic-oblong, rounded or obtuse at the apex, cuneate at the base, bright green, glabrous, prominently veined, the veins slightly prolonged beyond the low teeth; peduncles slender, 40 to 60 mm. long, nearly glabrous but with a few long crinkled hairs, often tomentulose just below the head; involucre short, one-third to one-half as long as the flowers, of linear-lanceolate, subulate bracts distinct almost to their bases; calyx one-half or two-thirds as long as the corolla, the teeth linear-subu- late, almost twice as long as the tube; corolla pale purplish, 11 mm, long or less, the banner emarginate. Type in the U. S. National Herbarium, no, 562146, collected along Hagle Creek at Gilmores Ranch in the White Mountains, August 25, 1907, by E. O. Wooton and Paul C. Standley. Altitude 2,220 meters. The plants grew in gravelly soil at the very edge of the water. ADDITIONAL SPECIMENS EXAMINED: White Mountains, alt. 1,890 meters, 1897, Wooton 235; Cold Spring Canyon, August 17, 1899, Wooton, This is related to 7’. lacerum, the flowers being of the same size, but the involucre is more deeply cleft and the segments much narrower, while the peduncles are more or less pubescent instead of glabrous. Vicia melilotoides Wooton & Standley, sp. nov. Perennial from a long, slender root; stems slender, angled, soft-pubescent, 80 cm. long or less, ascending or prostrate; leaflets 10 to 16, linear-oblong or linear-lanceolate, 25 mm. long or less, obtuse and mucronate or acute, bright green, rather thick, finely veined, loosely pubescent on both surfaces, becoming nearly glabrous in age; racemes many-flowered (15 to 20 or more), on peduncles as long as or often much shorter than the rachis; pedicels 1.5 mm. long or less; calyx 1.5 mm. long, at first loosely pubescent, glabrate in age; corolla 7 mm. long, creamy white; pods 25 to 80 mm. long, glabrous, about 8-seeded. Type in the U. S. National Herbarium, no. 498706, collected at Winsors Ranch in the Pecos River National Forest, July 16, 1908, by Paul C. Standley (no. 4364). Altitude 2,520 meters. The plants were abundant all through this region on open, stony hillsides in the Transition Zone. ADDITIONAL SPECIMENS EXAMINED: Mountains west of Las Vegas, 1881, Vasey; Upper Pecos River, July 27, 1898, Maltby & Coghill 108; Coolidge, June 16, 1887, Tracy 255; Hillsboro Peak, alt, 3,000 meters, 1904, Metcalfe 1245; Mogollon Creek, alt. 2,400 meters, 1903, Metcalfe 266; 1851, Wright 943; Middle Fork of the Gila, August 5, 1900, Wooton; McClures Ranch, August .2, 1900, Wooton; near Cloudcroft, July 31, 1899, Wooton; White Mourtains, alt. 2,100 meters, 1897, Wooton 288. This has long been confused with V. pulchella, which it closely resembles in general appearance. The flowers, however, are white, instead of blue as in that species, and much more numerous, while the peduncles are shorter, and the calyx less pubescent. Both species are found in the same region in the White Mountains, where they are easily distinguished in the field. 142 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. GERANIACEAE. Geranium eremophilum Wooton & Standley, sp. nov. Perennial from a slender, branched caudex; stems erect or ascending, very slender, 30 cm. long, much branched, minutely retrorse-pubescent, not glandular; leaf blades somewhat pentagonal in outline, 85 to 45 mm. broad, in age gla- brate, when young sparingly pubescent, 3 or 5-lobed, the divisions with abruptly acuminate, ovate teeth; pedicels slender, glandular; sepals oblong-lanceolate, about 10 mm. long, ending in an awn 2 mm. long, more or less pubescent, not glandular; petals purplish pink, obovate, retuse, 15 to 20 mm. long; style column canescent or rarely glandular; carpel bodies sparingly hirsute; seeds reticulate. Type in the U. 8S. National Herbarium, no. 233003, collected in the San Luis Mountains, September 5, 1893, by Dr. E. A. Mearns (no. 2142), ADDITIONAL SPECIMENS EXAMINED: San Luis Mountains, September 26, 1898, Mearns 2443; San Luis Mountains, September 11, 1898, Mearns 2194; Organ Mountains, alt. 1,800 meters, September 23, 1906, Wooton & Standley. It seems undesirable to describe any additional species of Geranium in this group where the species are so closely related and so poorly known, These plants, however, could not be placed under any name in Dr. Rydberg’s recent revision of the genus without stretching specific limits more than seems reason- able. This species is near G. fremontii and G. caespitosum, but is more slender, has paler, rather larger flowers, scarcely any glandular pubescence, and nearly glabrous leaves with broader, blunter segments. Geranium lentum Wooton & Standley, sp. nov. Perennial with a stout caudex ; stems weak, slender, ascending or spreading, 30 to 60 em. long, branched, densely glandular-villous; leaf blades reniform- pentagonal, 35 to 45 mm. wide, dull green, thin, glandular-hirsute on both surfaces, 5-parted, the divisions cuneate and with rounded obtuse teeth ; petioles very long and slender, densely glandular; pedicels long and slender, glandular; sepals elliptic-oblong, tipped with an awn less than 1 mm. long, glandular-pilose; petals white, 7 to 10 mm. long; style column densely glandu- lar; seeds finely reticulate. Type in the U. S. National Herbarium, no. 561079, collected on the West Fork of the Gila, August 7, 1900, by E. O. Wooton. ADDITIONAL SFECIMENS EXAMINED: Craters, Valencia County, July 28, 1906, Wooton; Rio Zuni, July 28, 1892, Wooton; West Fork of the Gila, August 6, 1900, Wooton; Middle Fork of the Gila, August 5, 1900, Wooton. The specimens cited here have passed as G. wislizeni, a Mexican species, Apparently none were examined in the preparation of the Geraniaceae for the North American Flora, for no description in that work agrees with our plant. Geranium wislizeni is the plant to which this is most closely related, but that lacks the dense glandular pubescence characteristic of our plant, being glandu- lar only on the style column. LINACEAE. Cathartolinum vestitum Wooton & Standley, sp. nov. Annual, 15 to 25 cm. high; stems slender, striate, densely puberulent, simple below, branched above the middle. the branches strongly ascending; leaves small, linear-lanceolate, glandular-denticulate, acute, thick, glaucous; bracts linear-lanceolate, 3 to 4 mm. long, aristate-tipped:; floral branches slender, 6 to 35 mm. long; sepals lanceolate, 5 or 6 mm. long, acute, bristle-tipped, with many glandular teeth on the margins, puberulent on the nerves; petals bright WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 148 yellow, 13 to 15 mm. long; capsules oblong-ovoid, 4 to 5 mm. long, shorter than the sepals. Type in the U. S. National Herbarium, no. 562282, collected at Mangas Springs, August, 1901, by O. B. Metcalfe. The only species with which this can be confused is C. puberulum, but that is a short, stout plant with short, thick pedicels and much shorter petals. Our plant, too, is simple below, while C. puberulum is invariably branched. Evidently related to this and probably the same is Doctor Mearns’s no. 47, collected in the Carrizalillo Mountains in 1892. ZYGOPHYLLACEAE. Kallstroemia laetevirens Thornber, sp. nov. Plants semierect, 30 to 50 cm. tall, branched from the base; stems 30 to 60 em. long, straight, sparingly pubescent throughout, beset above with scattered stiff hairs; stipules 5 to 6 mm. long, subulate to linear-lanceolate, hispid; leaves 30 to 55 mm. long, the petioles mostly equal to the lowest leaflets; leaf- lets 4 to 6 pairs, 9 to 20 mm. long, acutish, oblong to elliptical, glabrous above, pubescent with appressed hairs beneath; margins of leaflets prominently ciliate ; peduncles in fruit 15 to 40 mm. long, equaling or exceeding the subtending leaves, thickened above, appressed-pubescent and with scattered stiff hairs on the upper half, in age spreading or reflexed ; sepals subulate to linear-lanceolate, 6 to 8 mm. long, densely hispid with long, yellowish or tawny hairs; petals bright yellow, 7 to 12 mm. long, rather showy, exceeding the sepals; fruit finely canescent, the beak 5 to 8 mm. long, columnar, grooved, pubescent, often twice the length of the fruit body; nutlets 8 to 10, 3.5 to 4 mm, long, sharply tuberculate on the back, faintly reticulate on the faces. Type in the U. S. National Herbarium, no. 660420, collected on Hanover Mountain, New Mexico, July 31, 1911, by J. M. Holzinger. This well marked species differs from K, parviflora in its more erect habit, and also in having larger leaflets, leaves, and flowers. It differs from K. intermedia in its habit of growth, in its longer peduncles, and in the stems being very pilose. ADDITIONAL SPECIMENS EXAMINED: ARIZONA—-Fort Huachuca, 1894, Wilcor 295: without locality, 1869, Palmer; Empire Ranch, 1902, Griffiths & Thornber 284, New Mexico—Lincoln County, 1898, Skehan 52; Organ Mountains, alt. 1,360 meters, 1897, Wooton 423; Fort Bayard, 1905, Blumer 23; Kingston, alt. 2,000 meters, 1904, Metcalfe 1197; San Luis Mountains, 1893, Mearns 2202. RUTACEAE. Rutosma purpureum Wooton & Standley sp. nov. Perennial from a long, thick tap-root; stems very numerous, erect, sparingly branched, 30 cm. high or less, densely punctate with slightly raised glands; leaves linear, 15 mm. long or less, obtuse, thick, abundantly punctate, sessile; pedicels 1.5 mm. long or less; calyx lobes rounded-ovate, obtuse, less than 1 mm. long, purple; petals ovate, obtuse, dull reddish purple except near the base, there yellowish; fruit nearly sessile, 5 mm. high and 6 mm. wide at the top, the carpels strongly diverging, glabrous, conspicuously punctate. Type in the U. S. National Herbarium, no. 560630, collected on an arid, rocky slope at Bishops Cap at the south end of the Organ Mountains, April 4, 1903, by E. O. Wooton. 144 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. ADDITIONAL SPECIMENS EXAMINED: Bishops Cap, March 30, 1905, Wooton; Organ Mountains, September 23, 1906, Wooton & Standley; Carrizalillo Moun- tains, April 20, 1892, Mearns 118; 3 miles south of Hillsboro, alt. 1,650 meters, June 6, 1904, Metcalfe 1298; Mangas Springs, alt. 1,380 meters, 1903, Metcalfe 784. Other specimens have been seen from western Texas and southern Arizona. Doctor Gray, in the second part of Plantae Wrightianae, mentions this as “Rutosma teranum, var. corolla purpurea.” His specimens came from “ hills near El Paso.” Rutosma teranum, the other species of the Southwest, occurs in southwestern Texas and adjoining Mexico. It has yellow petals much larger than those of R. purpureum, and the carpels are erect instead of spreading. POLYGALACEAE. Polygala neomexicana Wooton & Standley, sp. nov. Stems slender, erect or ascending, 20 to 35 em. high, flexuous, simple or branched above, cinereous-puberulent; leaves numerous, thin, bright green, nearly sessile, sparingly puberulent or glabrate, lanceolate or elliptic, 15 to 30 mm. long, acute; inflorescence of few-flowered racemes; flowers soon pendulous, 6 mm. long, on pedicels 2 mm. long; perianth early deciduous: wings obovate, ciliolate, the keel naked; mature fruit oval or broadly oblong, about 10 mm. long, emarginate, the sinus closed, the faces puberulent, the margins ciliate, the seeds narrowly obovoid, hairy, the caruncle lobes forming processes one-third to one-half the length of the seed. Type in the U. S. National Herbarium, no. 497803, collected on Miller Hill, Grant County, September 8, 1897, by O. B. Metcalfe. ADDITIONAL SPECIMENS EXAMINED: Guadalupe Canyon, August 16, 1892, Mearns 692; San Luis Mountains, September 5, 1893, Mearns 2139. We have also seen specimens from southeastern Arizona. While closely related to P. puberula, our plant differs decidedly in its taller, more slender stems, larger, broader, thinner, nearly glabrous leaves, larger flowers, larger, puberulent fruit, and different seeds. The flowers appear to be yellow or white, but possibly they have faded in the herbarium. EUPHORBIACEAE. Chamaesyce chaetocalyx (Boiss.) Wooton & Standley. Euphorbia fendleri chaetocalyx Boiss. in DC. Prodr. 15*: 39, 1862. In his original description Boissier says: “Ab E. Fendleri habitu diversa videtur, sed auctoritate cl. Engelm. qui formas intermedias observasse monuit cum ea junxi.” Continued observation in the field compels us to regard this as a distinct species. It differs decidedly from C. fendleri in its erect rather than prostrate stems, its narrow, acute leaves, its elongated internodes, and its narrow appendages. Chamaesyce micromera (Boiss.) Wooton & Standley. Huphorbia micromera Boiss. in DC. Prodr. 157: 44. 1862, Chamaesyce serrula (Engeim.) Wooton & Standley. Euphorbia serrula Engelm. in Torr. U. 8. & Mex. Bound. Bot. 188. 1859. Croton eremophilus Wooton & Standley, sp. nov. Perennial, suffrutescent at the base, 35 em. high or less; stems slender, erect, densely stellate, corymbosely branched above; petioles 25 to 30 mm. long, slender; lower leaf blades oblong or oval, obtuse, 35 to 45 mm. long, grayish beneath with dense, stellate pubescence, green above and with a fine, rather WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 145 sparse, or at least not dense, stellate pubescence; upper leaves lanceolate or narrowly oblong, acute; flowers densely clustered, on stout pedicels 5 mm. long; sepals 5 or 6 mm. long, oblong, acute, densely stellate-pubescent ; staminate flowers with oblong, ciliate, acute petals; pistillate flowers apetalous; styles 38, bipartite; seeds oblong, brownish gray, 4.5 mm. long, with a stipitate caruncle. Type in the U. S. National Herbarium, no. 234163, collected by Dr. E. A. Mearns at Dog Spring in the Dog Mountains, September 16, 1893 (no. 2336). ADDITIONAL SPECIMENS EXAMINED: Parkers Well, July 19, 1901, Wooton. Similar to C. corymbulosus, but the two surfaces of the leaves are dissimilar in color, the upper being greener and much less densely pubescent; the upper leaves, too, are acute instead of obtuse, the petals are acute, and the sepals longer. Croton luteovirens Wooton & Standley, sp. nov. A slender, branched annual, 80 cm. high or less; stems glabrous, yellowish ; petioles slender, glabrous, one-third to one-half as long as the blades, these oblong-lanceolate, glabrous, yellowish green, acute; fiowers few, scattered, only a few in each raceme; sepals lanceolate, with a few stellate hairs, acute; petals wanting; capsules sparingly stellate-pubescent when young, soon glabrous; seeds broadly oval, 3 mm. long, light and dark brown striped, with an incon- spicuous caruncle. Type in the U. S. National Herbarium, no. 690282, collected on the Rio Gila, August 15, 1902, by E. O. Wooton. The plant is very abundant in this region, growing with the related C. tezr- ensis. Patches of the two are distinguishable at a distance because of their different color. Croton texensis is stellate-pubescent throughout, while our plant is glabrous; thus the two may be separated at a glance. Tithymalus altus (Norton) Wooton & Standley. Euphorbia alta Norton, Rep. Mo. Bot. Gard. 11: 108. 1899. Tithymalus chamaesula (Boiss.) Wooton & Standley. Euphorbia chamaesula Boiss, Cent. Euphorb, 38. 1860. Tithymalus luridus (Engelm.) Wooton & Standley. Euphorbia lurida Engelm. Proc. Amer. Acad. 5: 178. 1861. Tithymalus mexicanus (Engelm.) Wooton & Standley. Euphorbia dictyosperma mewicana Engelm. in Torr. U. 8. & Mex. Bound. Bot. 191. 1859. Euphorbia mexicana Norton, Rep. Mo. Bot. Gard. 11: 105. 1899, Zygophyllidium delicatulum Wooton & Standley, sp. nov. A slender annual, sparingly branched; stems erect or ascending, green, glabrous, sometimes purplish; leaves mostly alternate, the floral ones opposite or ternate, ovate to oblong, acute or obtuse, 15 to 30 mm. long, thin, bright green, finely serrulate, rounded to cuneate at the base; petioles slender, about as long as the blades; stipules mostly obsolete; flowers solitary, or clustered in the axils; pedicels 2 to 4 mm. long; involucres glabrous, with 5 glands; appendages obovate, greenish or purplish white; capsules glabrous, 4 mm. in diameter; seeds ovoid, terete, papillose, not carunculate. Type in the U. 8. National Herbarium, no. 562959, collected on Mineral Creek, Sierra County, at an altitude of 2,250 meters, September 26, 1904, by O. B. Metealfe (no. 1414). ADDITIONAL SPECIMENS EXAMINED: Tularosa Creek, August 18, 1899, Wooton; Ruidoso Creek, alt. 2,100 meters, June 30, 1895, Wooton. A very different plant from any of the other species of the genus, differing most noticeably in the width and shape of the leaf blades and the glabrous involucres. ‘ 146 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Zygophyllidium exstipulatum (Engelm.) Wooton & Standley. Euphorbia exstipulata Engelm. in Torr. U. 8. & Mex. Bound. Bot. 189. 1859. ANACARDIACEAE, Rhus choriophylla Wooton & Standley, sp. nov. Low shrub, 1 or even 2 meters high, with few branches; stems stout, the younger ones purplish, finely pubescent; leaflets 3 or 5, ovate, abruptly acumi- nate, 30 to GO mm. long, thick and coriaceous, glabrous, pale green, rather dull on the upper surface, mostly unequally rounded at the base; rachis finely puberulent; terminal leaflet larger and broader than the others; flowers in dense panicles much shorter than the leaves; bracts ovate, acuminate, densely pubescent; flowers almost sessile; fruit spherical, 5 mm. in diameter, orange, densely hirtellous, Type in the U. S. National Herbarium, no. 560279, collected in Guadalupe Canyon, on the Mexican boundary, near the southwest corner of New Mexico, August 16, 1892, by Dr. E. A. Mearns (no. 699). ADDITIONAL SPECIMENS EXAMINED: Guadalupe Canyon, 1893, Mearns 2524; Pena Blanca, November 30, 1901, Mae Gilmore; San Andreas Mountains, February 9, 1902, Wooton. In addition to the specimens cited, we have seen others from western Texas, southern Arizona, and northern Chihuahua. Evidently this is closely related to R. virens Lindh., but the vegetative characters are noticeably different, the leaflets being uniformly broader, rounded at the base, abruptly acuminate, glabrous, and of about the same shade of pale green on both surfaces. The leaflets, too, are fewer. ACERACEAE. Acer brachypterum Wooton & Standley, sp. nov. Tree of medium size, with spreading, smooth, brownish, slender branches; young twigs finely and densely pubescent; petioles usually shorter than the blades, mostly densely and finely velvety pubescent even in age; leaf blades averaging 5 cm. long and 6 to 8 em. broad, 5-lobed, all the lobes divergent, the two basal ones much smaller than the others or sometimes wanting; leaves cordate of truncate at the base, dull green and glabrous above, paler and pubescent beneath, thick, the lobes triangular-lanceolate or oblong-lanceolate, entire or with one or two low, rounded, lateral lobes, acutish; flowers not seen ; calyx persistent ; inflorescence short and few-fruited; pedicels slender. 20 to 35 mm. long, pubescent; body of the fruit turgid, glabrous, strongly reticulate, the wings small, 15 mm. long and 10 mm. wide or smaller, glabrous, strongly veined, reddish. Type in the U. S. National Herbarium, no. 560270, collected in the San Luis Mountains, July 19, 1892, by Dr. E. A. Mearns (no. 535). ADDITIONAL SPECIMENS EXAMINED: San Luis Mountains, 1892, Mearns 71, 569; San Luis Mountains, alt. 1.800 meters, October 1, 1893, Mearns 2481. A species near A. grandidentatum, but the leaves have very different lobes and are more densely pubescent, and the wings of the fruit are much shorter. _A note by Doctor Mearns on one of the sheets says: “This is the hard maple of this region. I also saw it in the Mogollon Mountains of Arizona, where the leaves turned red in October. Usually it is a small tree in ravines or canyons of the mountains. There is a photograph of the tree in the Report ' which grew *Report of the Boundary Commission upon the Survey and Re-marking of the Boundary between the United States and Mexico west of the Rio Grande, pl. facing p. 15. 1898, WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 147 in a moist canyon (Turkey Canyon) on the west side of the San Luis Moun- tains. It is the largest seen, measuring 225 cm. in circumference one meter above the ground. The trunk is of about the same size for 6 meters, when it divides into 3 branches of nearly equal size. The height was estimated at about 18 meters. There are several tyees in the neighborhood which approach this one in height.” The tree also occurs in southeastern Arizona. Possibly the Mogollon Moun- tain specimens mentioned above are true Acer grandidentatum. MALVACEAE. Malvastrum micranthum Wooton & Standley, sp. noy. Stems numerous, stout, erect, 20 em. high or less, much branched above, the branches ascending, densely and finely stellate-pubescent; leaves mostly about 12 mm. long, rarely as much as 18 mm., 3-cleft, the divisions all of about the same length, most of them deeply 3-lobed, the lobes oblong-oblanceolate, entire, rounded at the apex, densely stellate-pubescent beneath, sparingly so above; flowers few, only 3 to 5, approximate at the end of each branch; pedicels stout, 8 mm. long; calyx 3 to 4 mm. high, the lobes ovate-lanceolate, acute, longer than the tube; petals orange, 8 mm. long or shorter; fruit depressed ; ecarpels 9 or 10, finely stellate-pubescent, not cuspidate, semioblong, aintly reticulate on the inner surface: seeds solitary, filling the carpel. Type in the U. S. National Herbarium, no. 561154, collected by E. O. Wooton near Tiznitzin, August 4, 1904 (no. 2673). ADDITIONAL SPECIMENS EXAMINED: Mountains southeast of Patterson, August 16, 1900, Wooton. A very distinct species because of its small flowers and small, peculiarly divided leaves. Sphaeralcea arenaria Wooton & Standley, sp. nov. A low perennial, 80 em. high or less, from a thick, woody root; stems slender, erect or spreading, much branched in age, densely stellate-pubescent with a rather grayish, close pubescence ; petioles one-third as long as the blades or less; blades lanceolate, subhastate, rather obtuse, broadly cuneate or rounded at the base, about 25 mm. long, densely and finely grayish stellate pubescent on both surfaces; flowers axillary. solitary; leaves of the inflorescence but little re- duced; pedicels slender, 5 to 10 mm. long; calyx 7 or 8 mm. high, the lobes lanceolate, acute, densely stellate-pubescent ; petals obovate, emarginate, 12 mm. long, orange red; carpels numerous, 2-ovuled, 1-seeded, 5 or 6 mm. high, with a slender beak more than 0.5 mm. long, smooth above, reticulate below, densely stellate-pubescent on the back. Type in the U. 8. National Herbarium, no, 380390, collected by E. O. Wooton on the White Sands, Otero County, July 17, 1897 (no. 165). Altitude, 1,200 meters. ADDITIONAL SPECIMENS EXAMINED: Albuquerque, Herrick; Providencia Lake, July 3, 1900, Wooton; mesa west of Organ Mountains, August 26, 1899, Wooton; near Suwanee, August 1, 1906, Wooton; White Sands, August 51, 1904, Wooton 2662: between Tularosa and Mescalero Agency, June 22, 1895, Wooton. A common plant of the sandy mesas of southern New Mexico, coming into flower usually in late summer. It is low, with many spreading branches. It is similar to Sphaeralcea subhastata, but is finely instead of coarsely pubescent, and has thinner, mostly less lobed leaves, and the flowers are on long and slender pedicels rather than short, stout ones. 148 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Doctor .Coulter’s Sphaeralcea subhastata was a composite species, judging from material in the National Herbarium. One specimen of Wright’s collecting which he has named “8S. subhastata, n. sp.,” is evidentiy our Sphaeralcea arenaria, Another, however, marked in the same way is what we take to be S. subhastata. The original description points rather plainly to the second plant. Sphaeralcea tenuipes Wooton & Standley, sp. nov. Perennial from a thick, woody root; stems 30 em. high or less, very slender, much branched at the base, simple above, erect or ascending, sparingly stellate- pubescent with scattered, yellowish hairs; petioles as long as the blades or shorter; blades pedate, the lobes cuneate-oblanceolate, obtuse, 10 to 15 mm. long, entire or with 1 or 2 obtuse lateral lobes, rather bright yellowish green, sparingly stellate-pubescent on both surfaces; flowers in terminal racemes, soli- tary, rather distant, on slender pedicels 7 to 22 mm long; bracts linear-subu- late, reddish; calyx 8 mm. high, cleft half or two-thirds the way to the base, the lobes lanceolate, attenuate, densely stellate-pubescent; petals cuneate-ob- lanceolate to narrowly obovate, obtuse or retuse, 15 mm. long, 5 to 7 mm. wide, orange red. Type in the U. S. National Herbarium, no. 564301, collected on Tortugas Moun- tain southeast of Las Cruces, May 6, 1906, by Paul C. Standley. ADDITIONAL SPECIMENS EXAMINED: Tortugas Mountain, alt. 1,320 meters, Sep- tember 1, 1908, Wooton & Standley; Tortugas Mountain. April 22, 1894, March 2, 1902, August 27, 1894, August 29, 1902, Wooton; between El Paso and Monu- ment 53, September, 1892, Mearns 992. In general appearance this is similar to 8. pedata, but it is less pubescent and greener, the petals are narrower, and the flowers are solitary on long, slender pedicels instead of fascicled and on short, stout pedicels, This may be Sphaeralcea pedata angustiloba A. Gray.’ We have seen no material of that subspecies, but the description seems to define a different plant. Sphaeralcea tenwipes is rather common among the rough limestone rocks on Tortugas Mountain. Tt is a handsome plant, with its almost naked, slender racemes of bright colored flowers. Doubtless it occurs in similar situations about El Paso, Texas, and in northern Chihuahua. LOASACEAE. Mentzelia asperula Wooton & Standley, sp. nov. Annual with erect, branching stems 30 to 50 em. high, at first scabrous but becoming smooth below, the upper branches strongly ascending, the lower ones divergent then erect; petioles 1 em. long or less: leaf blades narrowly ovate to lanceolate, coarsely and irregularly serrate-dentate, sometimes la- ciniately 2 to 4-lobed near the base, hispid with barbed hairs; flowers solitary, appearing axillary, really terminal, the stem branching below after the flower is well grown, sessile: ‘alyx tube terete, short-clavate, elongating in fruit, densely hispid with barbed hairs, the lobes at first narrowly lanceolate, acuminate, becoming subulate, persisting on the fruit, 83 to 5 mm. long: petals 5, ovate to obovate, 6 to 8 mm. long, short-apiculate, orange, deciduous; fila- ments shorter than the petals, none of them dilated: fruit cylindric to long- clavate, 18 to 25 mm. long; seeds about 8, pyriform, obscurely and bluntiy angled, gray, with fine, parallel, curved strix. *Proc. Amer. Acad. 22: 292. 1887. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 149 Type in the U. S. National Herbarium, no, 498146, collected on Trujillo Creek, Sierra County, at an altitude of 2400 meters, September 14, 1904, by O. B. Metcalfe (no. 1864). ADDITIONAL SPECIMENS EXAMINED: New Mexico—Organ Mountains, Septem- ber 10, 1899, Wooton. Arizona—Sonoita Valley, 1874, Rothrock 642; near Fort Huachuca, 1894, Wileor 481; Bowie, 1884, Jones 4308; pass of the Chiricahua Mountains, 1851, Wright 1981. Trexas—Limpio Canyon, 1889, Nealley 659. Mexico—Near Durango, 1896, Palmer 484. The material from southern New Mexico and Arizona has mostly been re- ferred to Mentzelia aspera, a West Indian species with larger flowers, the outer row of filaments petaloid, the leaves acuminate and with more finely toothed margins and longer petioles. The Mexican representatives have been referred to Mentzelia hispida Willd., which has much larger flowers and leaves. Some of the Texas specimens have been called Mentzelia oligosperma, but that is a tuberous-rooted perennial with different leaves and habit. Mentzelia monosperma Wooton & Standley, sp. nov. Tuberous-rooted perennial with divaricately branched stems 30 to 40 em. high, forming a plant of as great or greater diameter; cortex on young stems yellowish green, hispid, becoming smooth, white, and papery; leaves broadly ovate in outline, obscurely 8-lobed, with a few coarse, sinuate teeth, broadly cuneate at the base, on petioles 2 or 8 mm. long, almost sessile above, acute or obtuse, bright green; flowers solitary, resembling those of M. oligosperma but smaller and with broader, shorter petals and calyx lobes; outer row of filaments about twice as wide as the inner ones; fruit clavate, woody when mature, about 1 cm. long, with a single large seed 38 mm. long, this elliptic- oblong more or less triangular in cross section, dull brownish, finely striate with undulating lines. Type in the U. S. National Herbarium, no. 690230, collected in the Organ Mountains, August 29, 1894, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Thirty-five miles west of Roswell, 1900, Earle 521. In root and seed characters this plant is very similar to Mentzelia oligo- sperma, to which it has been referred. The flowers and habit of the plant are different from those of that species. It has been collected only once in the type locality, therefore is probably very rare. Nuttallia gypsea Wooton & Standley, sp. nov. Tufted, herbaceous perennial, 20 to 40 em. high, branching freely above, leafy to the top; cortex white and smooth below, scabrous above; leaves green, oblong in outline, 2 to 4 em. long, about 1 cm. wide or less, pinnately divided into linear, mostly obtuse segments barely 1 mm. wide, short-petiolate, never elasping or auriculate at the base, rough with scattered, recurved, stout, white, barbed hairs swollen at the bases; flowers small, on short, terminal pedicels, subtended by 1 or 2 small, linear bracts; hypanthium campanulate, 3 to 4 - mm. long; sepals ovate-lanceolate, acuminate in bud, about 5 mm. long, becoming triangular-subulate; petals lanceolate, broader than the 5 inner staminodia, pale yellow, tapering to the base, acute, 12 to 15 mm. long, 2 to 38 mm. wide; stamens numerous, the outer filaments somewhat dilated, about as long as the petals, the inner ones shorter; capsules almost hemispheric, about 7 mm. long; seeds numerous, flat, winged, minutely tuberculate. Type in the U. S. National Herbarium, no. 564614, collected on pure gypsum near Lakewood, August 6, 1909, by E. O. Wooton. 150 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. ADDITIONAL SPECIMENS EXAMINED: On gypsum soil on plains 35 miles south of Torrance, August 10, 1909, Wooton. This species seems to be affiliated with Nuttallia laciniata, but is much smaller in every way, and the flowers are pale yellow. The time of opening of the flowers is not known. The specimens collected in the morning about 11 o’clock were as wide open as those taken about sundown. Nuttallia laciniata (Rydb.) Wooton & Standley. Touterea laciniata Rydb. Bull. Torrey Club 31: 565. 1904. Nuttallia procera Wooton & Standley, sp. nov. Perennial herb, 60 to 100 cm. high, slender, strict, sometimes branching at the base; cortex white and papery, smooth except on the young stems; leaves small, sessile, oblong, obtuse, 30 to 50 mm, long, about 5 mm. wide, with 5 to 10 coarse, rounded teeth on each side, very rough with short, stout, white barbed hairs; flowers rather small, on slender, terminal peduncles, or the stems somewhat corymbosely branched above; sepals narrowly lanceolate, abruptly acuminate; petals rather bright yellow, about 1 cm. long, oblanceolate, acute, the 5 staminodia narrower and sometimes short-acuminate; outer filaments somewhat dilated; capsules oblong-cylindric, 10 to 12 mm. long, 6 or 7 mm. in diameter; seeds numerous, disk-shaped, surrounded by a broad wing, white, tuberculate. Type in the U. S. National Herbarium, no. 516161, collected on the White Sands, August 18, 1907, by E. O. Wooton and Paul C. Standley. The plant has been collected in the vicinity of the White Sands several times. Wooton’s 571, distributed as Mentzelia pumila?, is the same species collected on the Sands in 1897. It has also been collected above Tularosa in the White Mountains (Wooton 567 in 1897). . The species is most nearly related to Nuttallia multiflora, but is of very dif- ferent habit, much more slender, and has smaller, less pinnatifid leaves and smaller flowers and fruit. Nuttallia strictissima Wooton & Standley, sp. nov. Fifty to 70 em, high or more; stems simple below, with a few erect branches near the top, whitish, scabrous; lower leaves linear or linear-elliptic, acutish, scabrous, very shallowly dentate, sessile; uppermost leaves linear or linear- lanceolate, reduced and bract-like, crowded, mostly entire, attenuate; calyx lobes about 1 em. long, very thick, scabrous with short, stiff, whitish hairs, narrowly triangular, attenuate; flowers few; petals 18 to 22 mm. long, linear- oblanceolate, very acute, with as many or twice as many petal-like staminodia ; stamens numerous, the filaments of the outer ones broad and flattened; cap- sules 2 cm. long, 8 mm. in diameter, cylindric, scabrous. Type in the U. 8. National Herbarium, no. 496766, collected on the Arroyo Ranch, near Roswell, in September, 1903, by David Griffiths (no. 5701). ADDITIONAL SPECIMENS EXAMINED: Twenty miles south of Roswell, alt. 1,080 meters, August, 1900, Farle 317. Related to V. multiflora, but distinguished by its strict habit, narrow petals, and the peculiar bract-like upper leaves. ONAGRACEAE. Anogra amplexicaulis Wooton & Standley, sp. nov. Stems stout, reddish, much branched, sparingly hirsute and with a sparse, fine, cinereous pubescence; lower leaves short-petioled, the upper clasping by an auriculate base, oblong-lanceolate, oblong, or triangular-lanceolate, acute, entire or runcinate-toothed near the base, green, puberulent and more or less villous ; ’ WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 151 calyx tips free; calyx tube slender, 30 to 40 mm. long, villous; sepals 15 to 20 mm. long, sparingly villous; petals white fading pink, about 13 mm. long; pistil slightly exserted; capsules ascending, cylindric, 40 mm. long and 2 or 3 mm. in diameter. villous. Type in the U. S. National Herbarium, no. 497937, collected by O. B. Metcalfe on a sandbar along the Mimbres River, July 1, 1904 (no. 1054). Very like Anogra neomexicana, but with much smaller flowers, clasping, more pubescent leaves, longer capsules, and more pubescent stems. Anogra ctenophylla Wooton & Standley, sp. nov. Biennial, 30 cm. high or less; stems stout, with divergent branches, densely and finely cinereous, often with a few long hairs; leaves 35 to 60 mm. long, short-petiolate or sessile, deeply pinnatifid almost to the midrib, the divisions elliptic-oblong, acute, densely puberulent and more or less hirsute; tips of the calyx lobes free in bud; calyx segments 2 cm. long, densely and minutely cinereous, sparingly hirsute; petals 35 to 40 mm. long, white; pistil shorter than the petals; capsules ascending, cylindric, 35 mm. long, minutely cinereous and hirsute. Type in the U. 8S. National Herbarium, no. 564751, collected near Zuni, in 1902, by Mrs. Matilda Coxe Stevenson (no. 99). ADDITIONAL SPECIMENS EXAMINED: Crawfords Ranch, June 21, 1906, Wooton; Reserve, July 9, 1906, Wooton; Defiance, June 22, 1883, Marsh 121; Burro Mountains, alt. 2,100 meters, August 4, 1906, Blumer 1827; Ruidoso Creek, August 20, 1897, Wooton. The last two specimens are somewhat doubtful, but probably belong here. The species is related to Anogra runcinata, but it has long hairs among the appressed pubescence and its leaves are deeply pinnatifid. Anogra engelmanni (Small) Wooton & Standley. Ocnothera albicaulis trichocalyx Engelm. Amer. Journ. Sci. II. 34: 335. 1862, not O. trichocalyr Nutt. Anogra pallida engelmanni Small, Bull. Torrey Club 23: 176. 1896. The type of this came from Las Vegas, New Mexico, collected by Wislizenus in 1846. We have seen only a single additional collection from the State, one gathered by Mr. Geo. L. Fisher near Nara Visa in 1910 (no, 54). Anogra leucotricha Wooton & Standley, sp. nov. Low and spreading or prostrate, with many divergent branches; stems stout, densely hirsute, 20 cm. long or less; leaves short-petiolate, 30 to 40 mm. long, deeply pinnatifid, the segments triangular-lanceolate and acute, densely strigil- lose and somewhat hirsute; calyx tube about 25 mm. long; sepals 15 mm. long, finely appressed-pubescent and hirsute; corolla 20 mm. long, white; capsules divergent, cylindric, 30 to 40 mm. long, hirsute, Type in the U. S. National Herbarium, no. 690235, collected on the San Augustine Plains, July 22, 1904, by E. O. Wooton (no. 2735). ADDITIONAL SPECIMENS EXAMINED: Willard, August 26, 1904, Wooton; plains 10 miles east of Horse Spring, June 20, 1892, Wooton. Related to Anogra engelmanni, but with deeply pinnatifid and short-petiolate leaves and different pubescence. Anogra runcinata (Engelm.) Wooton & Standley. Oenothera albicaulis runcinata Engelm. Amer. Journ. Sci. II, 34: 334. 1862. Anogra pallida runcinata Small, Bull. Torrey Club 23: 175. 1896. A common species found in almost all parts of New Mexico in the Lower and Upper Sonoran zones. 60541 °—13——4 152 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Galpinsia camporum Wooton & Standley, sp. nov. Low perennial, 15 em. high, from a thick, woody base; stems very numerous, simple or branched, slender, flexuous, finely and densely villous, glandular; leaves very numerous, green, elliptic-lanceolate, 15 to 20 mm. long, acute, entire or nearly so, glandular-pubescent or glandular-puberulent; calyx tube 4 cm. long, slender, sparingly glandular or villous; sepals 12 mm. long, slightly villous, the free tips 2 mm. long; petals 12 to 18 mm. long, rounded-obovate; capsules cylindric, erect, 13 mm. long, soft-pubescent and glandular, Type in the U. 8. National Herbarium, no. 564592, collected at Knowles, July 29, 1909, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Highest point of the Llano Estacado, June 17, 1903, Bailey 518; near Causey, August 17, 1909, Wooton; Buchanan, Au- gust 12, 1909, Wooton; Hondo Hill, July 28, 1905, Wooton; Nara Visa, October 2, 1907, W. Belknap. The type was collected in sandy soil on the broad plains near Knowles. It is similar to Galpinsia greggii and G. lampsana. From the former it is dis- tinguished by its larger flowers and the different form of the leaves, and from the latter by its much smaller flowers and shorter, less abundant pubescence. Galpinsia lampsana (Buckl.) Wooton & Standley. Oenothera lampsana Buckl. Proc. Acad. Phila. 1861: 454. 1862. A common species of western Texas and eastern New Mexico. Gaura brassicacea Wooton & Standley, sp. nov. Stems numerous, slender, ascending, 90 em. high or less, purplish, densely hispid; basal leaves oblanceolate, 5 to 9 ecm. long, sinuate-toothed; cauline leaves oblong to oblanceolate, sessile, acutish or obtuse, deeply sinuate-dentate, abundantly hirsute, especially along the veins; branches of the inflorescence glabrous, slender; bracts broadly obovate, acuminate, short-ciliate; calyx glabrous; fruit glabrous, narrowly ovoid, sharply angled, short-stipitate, 8 to 10 mm. long, acute: Type in the U. S. National Herbarium, no. 45764, collected at Socorro in May, 1881, by G. R. Vasey. Another specimen of the same collection is mounted on sheet 45763. Gaura cinerea Wooton & Standley, sp. nov. Probably tall and much branched; stems slender, ascending, covered. with a dense, loose, rather stiff pubescence; leaves elliptic or narrowly oblong, 20 to 25 mm, long, abruptly acuminate with subulate tips, sessile, sinuate-serrate below the middle, covered with a dense, coarse, curled pubescence: branches of the inflorescence densely cinereous; bracts ovate-lanceolate, with attenuate, subulate tips, pubescent; calyx and ovary densely cinereous; fruit on a slender cinereous stipe as long as the body. Type in the U. 8S. National Herbarium, no. 382592, collected 20 miles south of Roswell, in August, 1900, by F. S. and Esther 8. Earle (no. 538). Altitude 1,080 meters. We have also seen a specimen collected at Big Spring, Texas, in September, 1881, by Dr. V. Havard. Both sheets are in poor condition, showing neither open flowers nor mature fruit, but the plants are so distinct that one can not hesitate to describe them. The only species with which to compare this is Gaura villosa, but in that the branches of the inflorescence are glabrous and the pubescence on the flowers and stems is of an entirely different kind. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 153 Gaura induta Wooton & Standley, sp. nov. Gaura glabra Rydb. in part, not Lehm. Low perennial, 30 cm. high or less, with numerous branched, slender, glabrous stems: lower leaves lanceolate-oblong, sparingly toothed, acute, sessile, gla- brous, the upper ones linear; bracts linear, much exceeding the ovary; calyx tube 7 or 8 mm. long, it and the lobes densely strigose; petals 6 mm. long, rhombic, long-clawed; fruit 5 or 6 mm. long, with a thick stipe, densely and minutely cinereous. Type in the U. S. National Herbarium, no, 498956, collected on the dry, clay hills near Pecos, August 15, 1908, by Paul C. Standley (no. 4933). Altitude 2,010 meters. ADDITIONAL SPECIMENS EXAMINED: Santa Fe, alt. 2,160 meters, 1897, Heller 2659 ; 1847, Fendler 231b; Las Vegas, June 24, 1891, Dewey; Sandia Mountains, 1898, Herrick 276; near Tesuque, August 20, 1904, Wooton; Patterson, August 15, 1900, Wooton; mouth of Pino Canyon, 1898, Herrick 276; Farmington, alt. 1,600 meters, 1911, Standley 6919; Cedar Hill, alt. 1,900 meters, 1911, Standley 7911; Dulce, alt. 2,150 meters, 1911, Standley 8165; Nutritas Creek, alt. 2,250 meters, 1911, Eggleston 6638; Raton, alt. 2,100 meters, 1911, Standley 6297; north of Ramah, July 25, 1906, Wooton; Estancia, September 22, 1907, M. B. Atkinson; Hebron, September 21, 1907, C. de Foresta; Santa Fe, 1908, Standley 4482, Most, if not all of the material placed by Doctor Rydberg under Gaura glabra in the Flora of Colorado belongs here. That species was described as being glabrous, and such a plant is well represented in the National Herba- rium by specimens from Montana and adjoining States. Our plant has a.wide range outside of New Mexico, extending from Arizona and Utah to Wyoming and South Dakota. It occurs chiefly in the low foothills and on the dry plains of the Upper Sonoran Zone. Gaura glandulosa Wooton & Standley, sp. nov. Stems usually numerous, slender, much branched, erect, 90 cm. high or less, copiously hirsute; basal leaves oblanceolate, somewhat sinuate-dentate, 7 to 10 cm. long, slender-petioled; cauline leaves linear or linear-oblong, entire or re- motely repand-toothed, pubescent especially on the margins and veins, acute to obtuse, bright green, sessile; racemes slender, the branches glabrous except among the flowers and buds, there glandular; bracts ovate, acute, ciliolate, glandular; calyx tube 5 mm, long, glandular; petals oblanceolate, obtuse, 6 mm. long; fruit sessile, broadly ovoid, sharply angled, glabrous, not stipitate. Type in the U. S. National Herbarium, no. 561072, collected at Reserve, July 9, 1906, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Gila Hot Springs, August 20, 1900, Wooton ; eons Fork of the Gila, August 5, 1900, Wooton; Sapello Creek, August 22, 1900, Wooton; N Bar Ranch, August 2, 1900, Wooton. his has always passed as Gaura nealleyi Coulter, but that is a nearly glabrous plant with short-stipitate fruit. Gaura gracilis Wooton & Standley, sp. nov. Stems very slender, with numerous corymbose, ascending branches, 50 to 80 em. high, villous; cauline leaves linear, 25 to 35 mm. long, entire, bright green, glabrous or sparingly puberulent, acute, short-petiolate; branches of the inflo- rescence glabrous; bracts ovate, acuminate, ciliolafe, strigillose; calyx tube 4 mm. long, strigillose; petals 5 mm. long; fruit sessile, elliptic-ovoid, sharply angled, glabrous, 7 or 8 mm. long. not stipitate. Type in the U. S, National Herbarium, no. 499693, collected at the Forest Nursery at Fort Bayard, August 29, 1905, by J. C. Blumer (no. 44), 154 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. ADDITIONAL SPECIMENS EXAMINED: Filmore Canyon, September 4, 1897, Wooton; Filmore Canyon, alt. 1,800 meters, September 23, 1906, Wooton € Standley; Mimbres River, alt. 1,650 meters, 1904, Metcalfe 1088. Gaura linearis Wooton & Standley, sp. nov. Stems slender, erect or ascending, branched, glabrous; leaves linear, bright green, entire, acute, sessile, 10 to 15 mm. long, numerous; branches of the in- florescence cinereous-puberulent among the flowers; bracts lanceolate or ovate, acuminate, usually less than half as long as the ovary; calyx tube 3 mm. long, strigose like the lobes; petals G or 7 mm. long, oblanceolate, obtuse, long- clawed; mature fruit not seen, but the ovary densely whitish-strigose. Type in the U, 8. National Herbarium, no. 564598, collected on gypsum soil near ‘Lakewood, August 6, 1909, by E. O. Wooton. Related to Gaura induta, but readily distinguished by the short, broad, acumi- nate bracts and narrow leaves. Gaura podocarpa Wooton & Standley, sp. nov. Stems slender, branched from the base and again above, erect, hirsute, red- dish; leaves narrowly oblanceolate or oblong, 4 to 6 cm. long, with a few low, repand teeth, acute, the uppermost leaves entire, linear, hirsute along the veins, ciliate; branches of the inflorescence glabrous; bracts ovate, acute, ciliolate; calyx tube glabrous, 5 mm. long; petals oblanceolate, obtuse, 6 or 7 mm long; fruit ovoid, 8 mm. long, acutish, very sharply angled, contracted into a short stipe below. Type in the U. 8. National Herbarium, no. 495277, collected by O. B. Metcalfe on Bear Mountain near Silver City, Grant County, June 17, 1903 (no. 166). Altitude 1,500 meters. ADDITIONAL SPECIMENS EXAMINED: West Fork of the Gila, alt. 2,040 meters, 1903, Metcalfe 841; West Fork of the Gila, alt. 2,100 meters, August 6, 1900, Wooton; Van Pattens, August 29, 1894, Wooton; Filmore Canyon, October 29, 1904, Wooton. The plant also occurs in the Huachuca Mountains of Arizona. This, with Gaura strigillosa, G. gracilis, and G. glandulosa, described here, has passed as G. suffulta Engelm., a plant originally described from Lindheimer’s collections. All four of our plants have much narrower leaves, broader and much shorter bracts, smaller flowers, and larger fruit; while each one, in addi- tion, differs from that species in other particulars. Gaura strigillosa Wooton & Standley, sp. nov. Stems slender, ascending, much branched, 60 em, high or less, reddish, hir- sute; leaves oblong or linear-oblong, or the uppermost linear, the larger ones sinuate-dentate, sessile, acutish, glabrous except for the short-hirsute mid- vein and ciliolate margins; branches of the inflorescence glabrous; bracts ovate or ovate-lanceolate, acute, strigillose, ciliolate; calyx strigillose, the tube 6 mm. long; petals G mm. long; fruit glabrous, narrowly ovoid, sharply angled, short- stipitate, 8 mm. long. Type in the U. 8. National Herbarium, no. 5610738, collected by E. O. Wooton at Wingfields Ranch on Ruidoso Creek in the White Mountains, July 8, 1895. Lavauxia hamata Wooton & Standley, sp. nov. Cespitose perennial with a short, thick stem 5 to 6 em. long; leaves narrowly elliptic-lanceolate, deeply and irregularly pinnatifid, the segments acute, attenu- ate, long-petioled, bright green, glabrous except along the puberulent margins; calyx tube 5 to 7 cm. long, slender, glabrous or nearly so, the lobes 15 to 20 mm. long, nearly glabrous; petals 2 cm. long; capsules 20 to 25 mm. long, 10 mm. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 155 thick, sharply angled, the angles furnished below the top with stout, divaricate, hooked processes, the beak of the capsule stout, the whole finely puberulent. Type in the U. S. National Herbarium, no. 45766, collected at Socorro, in May, 1881, by G. R. Vasey. The stout, hooked beak on the wings distinguish this from all our other species; otherwise it is similar to Lavauria flava. Lavauxia taraxacoides Wooton & Standley, sp. nov. Caudex short and thick; leaves 20 to 30 cm. long, narrowly oblanceolate, deeply pinnatifid near the base into narrow, acute, distant lobes, the terminal portion merely slightly toothed, long-petioled, glabrous; calyx tube 18 to 20 cm. long, slender, glabrous; sepals 85 mm. long, glabrous; petals 35 to 40 mm. long; capsules oblong, 25 mm. high, acute at the apex, narrowly winged, glabrous or slightly puberulent. Type in the U. 8S. National Herbarium, no. 563856, collected in James Canyon of the Sacramento Mountains, July 6, 1899, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Tularosa Creek, August 18, 1899, Wooton; White Mountain Peak, August 1, 1901, Wooton; James Canyon, June 26, 1899, Wooton; White Mountains, alt. 3,000 meters, August 16, 1897, Wooton 664. A species readily distinguished by its large leaves and flowers and by its lack of pubescence. Oenothera irrigua Wooton & Standley, sp. nov. Probably a biennial, 2 meters high or less, with very numerous stout, spread- ing branches; stems stout, terete, densely and finely canescent, also with numerous short, spreading hairs; basal leaves not seen, the cauline ones nar- rowly elliptic-lanceolate, 14 en’. long and 2 cm. wide or less, acute, narrowed to the base, sessile or on short, winged petioles, obscurely repand-denticulate, rather densely appressed-pubescent on both surfaces, grayish green; inflores- cence a short, dense raceme; ovary densely strigose; calyx tube rather stout, 35 to 45 mm. long, strigose; sepals about 40 mm. long, separate when reflexed, the tips connivent in bud, 5 to 8 mm. long; petals 385 mm. long, yellow, drying purplish red; pistil slightly exserted; mature capsules not seen, but the im- mature ones columnar, densely silky-strigose, much exceeded by the floral leaves. Type in the U. S. National Herbarium, no, 561366, collected in the Mesilla Valley, Dona Ana County, in June, 1906, by E. O. Wooton and Paul C. Stand- ley. Altitude about 1,150 meters. The plant is very abundant along the banks of irrigating ditches and in moist cultivated fields. ADDITIONAL SPECIMENS EXAMINED: Mesilla Valley, July 25, 1907, Wooton & Standley; Farmington, August 8, 1904, Wooton 2732; Albuquerque, October 18, 1894, Herrick; Aztec, July 1, 1895, Grigin; Mesilla Valley, [vah Afead,. . Aplant similar to O. hookeri, but much larger and more abundantly branched, and with very different pubescence. Oenothera macrosiphon Wooton & Standley, sp. nov. Perennial, 50 cm. high or less, with slender, weak, decumbent branches; stems several from each root, branched, hirsute, the hairs rising from papille, also minutely cinereous, but sparingly so; cauline leaves lanceolate or elliptic-lanceolate, 7 to 9 cm. long, 25 mm. wide or less, undulate, acute, cuneate at the base or rounded to a short, winged petiole, sparingly repand- denticulate, finely appressed-pubescent, hirsute along the veins, the leaves of the inflorescence slightly reduced; flowers few; calyx tube 15 to 19 cm. long, 2 mm. thick, sparingly pilose; sepals 50 to 60 mm. long, the subulate tips 5 mm. long or more; pistil and stamens included; petals 50 to 55 mm, long, deep 156 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. yellow turning purplish; capsules 30 to 40 mm. long, columnar, obtusely angled, sparingly hirsute. Type in the U. 8. National Herbarium, no. 241248, collected in the Organ Mountains, August 29, 1894, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Organ Mountains, alt. 1,860 meters, July 8, 1897, Wooton 114; Organ Mountains, 1881, Vasey; Van Pattens, June 11, 1906, Standley; Van Pattens, September 10, 1899, Wooton; Dripping Springs, August, 1898, Cockereil. A beautiful plant with larger flowers than any other species of the genus. It occurs in the Organs in deep, rocky canyons, principally about the edges of pools. It has been called O. jamesii, but that species has much smaller flowers and abundant, appressed pubescence. Oenothera procera Wooton & Standley, sp. nov. A slender biennial or perennial, 40 to 100 cm. high; stems simple, sparingly and loosely hirsute, also with a few inconspicuous, appressed, curled hairs; basal leaves not seen, the cauline ones elliptic-lanceolate or mostly oblanceolate, 8 to 10 ecm. long, 15 mm. wide or less, acute, narrowed at the base to a slender petiole, bright green, thin, entire or faintly repand-denticulate, sparingly appressed-pubescent on both surfaces; leaves of the inflorescence considerably reduced; racemes short, few-flowered; calyx tube slender, about 25 mm. long, loosely pubescent or nearly glabrous; sepals distinct in anthesis, 15 mm. long; petals 12 to 14 mm. long, golden yellow fading purplish; pistil not exserted; capsule 20 to 25 mm, long, obtusely angled, 3 to 4 mm. thick, sparingly hirsute. Type in the U. S. National Herbarium, no. 498579, collected along Winsor Creek in the Pecos River National Forest, July 5, 1908, by Paul C. Standley (no. 4212). Altitude 2,550 meters. ADDITIONAL SPECIMENS EXAMINED: Mouth of Mora River, alt. 2,460 meters, July 7, 1908, Standley 4246; Fendler 218; West Fork of the Gila, alt. 2,250 meters, August 4, 1908, Metcalfe 379; Ruidoso Creek, August 20, 1897, Wooton; James Canyon, August 11, 1899, Wooton; Beulah, August, 1899, Cockerell; tilmores Ranch, alt. 2,220 meters, August 25, 1907, Wooton & Standley; Pajarito Park, August, 1908, Bartlett; Upper Pecos River, 1898, Maltby & Coghill 75; Gilmores Ranch, July 14, 1895, Wooton; White Mountain Peak, July 6, 1895, Wooton; Harveys Upper Ranch, alt. 2,880 meters, 1908, Standley 4672; Santa Fe, 1908, Standley 4523. A common plant in the mountains in the Transition Zone. It grows usually on moist open slopes, but sometimes along streams. Seldom or never does it exceed a meter in height, and the stems ..are invariably. simple. The type collection was distributed as O. strigosa Rydb., but that is a plant with much larger flowers and different pubescence. Part of the specimens referred to Onagra strigosa by Doctor Rydberg in the Flora of Colorado belong here. Pachylophus australis Wooton & Standley, sp. nov. Acaulescent, cespitose; leaves on short, rather slender petioles, the blades 10 to 14 em. long, narrowly oblanceolate, abruptly acute, remotely denticulate near the apex, toward the base pinnatifid into distant, triangular segments, finely cinereous on both surfaces, green; hypanthium tube 14 cm. long, about 12 mm. wide in the throat, minutely cinereous; sepals nearly linear, cinereous; petals white, obovate, 5 ecm. long; mature capsules not seen, the ovaries densely appressed-pubescent. Type in the U. S. National Herbarium, no. 690245, collected on the South Fork of Tularosa Creek, July 31, 1897, by E. O. Wooton. Related to P. montanus (Nutt.) A. Nels., but with much larger flowers and a long hypanthium tube. £ WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 157 Pachylophus eximius (A. Gray) Wooton & Standley. Oenothera eximia A. Gray, Mem, Amer. Acad. II. 4: 45. 1849. Pachylophus exiguus Rydb. Colo. Agr. Exp. Sta. Bull. 100: 246. 1906. CORNACEAE. Garrya goldmanii Wooton & Standley, sp. nov. A low. shrub, 1 meter high or less; young branches densely covered with fine curled hair, the pubescence persisting often for several years; leaves small, 40 mm. long and 20 mm. wide or smaller, usually about 25 mm. long and 12 mm. wide, elliptic, narrowed toward the base and the mucronate apex, very thick, yellowish green, somewhat crispate, not at all conspicuously veined, pubescent on both surfaces, densely so beneath, the hairs long and only slightly curled or crinkled; margins of the leaves thickened, yellowish, muriculate; the stout petioles 5 to 7 mm. long; fruit in racemes 2 em. long or less, sessile, subtended by lanceolate, abruptly acuminate bracts about 7 mm. long; fruit glabrous, ovoid to spherical, 6 mm, in diameter or less. Type in the U. S. National Herbarium, no. 562808, collected on limestone ledges near Queen, New Mexico, July 31, 1909, by E. O. Wooton, Altitude about 1,770 meters. ADDITIONAL SPECIMENS EXAMINED: New Mexico—Big Hatchet Mountains, alt. 2,010 meters, 1908, Goldman 1319, 1318; Sheep Mountain, San Andreas Range, 1902, Gaut 86. Trxas—Guadalupe Mountains, 1901, Bailey 452; Chisos Moun- tains, 1901, Bailey, 371. ‘In Coulter’s Botany of Western Texas this is referred to as a narrow-leaved variety of G. ovata Benth. It is undoubtedly closely related to that species of central Mexico, but it differs in its lower growth, and small, narrow, more pubescent, crispate leaves. The leaves are much less conspicuously veined than in G. ovata and the fruit is much smaller. ARALIACEAE. Aralia bicrenata Wooton & Standley, sp. nov. Somewhat branched herb about 1 meter high; stems stout, sparingly pubescent on the older parts, abundantly pubescent on the younger branches; petioles long and slender; leaflets ovate, 6 to 8 cm. long, the lower ones 3-parted, abruptly acuminate, oblique to cordate at the base, very thin, bright green, doubly crenate almost to the base, nearly glabrous above, puberulent beneath, especially along the veins; inflorescence much branched; peduncles short, 10 to 25 mm. long, puberulent; pedicels 7 to 8 mm. long, numerous; bracts linear, 1 to 2 mm. long; petals ovate, obtuse; ovary glabrous; mature fruit not seen. Type in the U. S. National Herbarium, no. 563963, collected near Holts Ranch in the Mogollon Mountains, July 20, 1900, by BE. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Mogollon Creek, alt. 2,250 meters, July 23, 1908, Metcalfe 303; Las Vegas Hot Springs, August, 1901, H. S. Barber 151; South Bonito Creek, 2 miles above the forks, 1899, Turner 216; Gallinas Planting Station, 1908, Bartlett 301; Sierra Grande, alt. 2,600 meters, 1911, Standley 6136. The plant is related to A. pubescens DC. and A. hwmilis Cav. (if they are separable species) of Mexico, but it has much larger and thinner leaves, doubly instead of simply crenate, the pubescence is much more sparse, and the stems are not at all woody. Our plant is also an ally of A. racemosa, but the form of the inflorescence is different, the leaves are not deeply cordate at the base, and they are not so conspicuously acuminate. 158 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. APIACEAE. Phellopterus utahensis (Jones) Wooton & Standley. Cymopterus montanus purpurascens A, Gray in Ives, Rep. Colo. Riv. 15. 1860. Cymopterus utahensis Jones, Proce. Calif. Acad. II. 5: 684, 1895. Cymopterus utahensis monocephalus Jones, op. cit. 685. Phellopterus purpurascens Coult. & Rose, Contr. U. 8. Nat. Herb. 7: 168. 1900. Pseudocymopterus filicinus Wooton & Standley, sp. nov. Roots elongated, 15 mm. thick or more; stems very densely clustered, low, 20 to 25 cm. high, sparingly branched or nearly simple, slender, glabrous; basal leaves very numerous, 20 to 25 em. long; petioles slender, 6 to 8 cm. long; blades broadly triangular or rhombie in outline, 8 to 14 em. long and usually almost as broad, the length of the lower divisions causing the blades to appear ternate; most of the blades thrice parted; ultimate segments linear or linear-elliptic, bright green, thin, glabrous, very numerous, crowded, short, 15 mm. long or less; principal divisions of the leaves appearing sessile because of the presence of lobes at their bases; peduncles scarcely exceeding the leaves; umbels 15 mm. wide or less, dense, the short branches often puberulent; involu- cels linear; flowers bright yellow. Type in the U. S. National Herbarium, no. 564352, collected on Bear Moun- tain near Silver City, Grant County, June 17, 1908, by O. B. Metcalfe (no. 165). Another specimen of the same collection is mounted on sheet 560402, ADDITIONAL SPECIMENS EXAMINED: Mangas Springs, September 1, 1897, Met- calfe; Holts Ranch, July 20, 1900, Wooton; Pinos Altos, 1891, Nealley 46, A very handsome plant for the family, its leaves strongly suggesting some of the ferns. It is distinguished from our other species by the very numerous leaves of peculiar form, and by the small umbels usually but slightly exceeding the leaves. PRIMULACEAE. Steironema validulum Greene, sp. nov. Stem 30 to 60 cm. high, robust, whitish and somewhat polished, densely leafy and floriferous from below the middle; leaves lanceolate or lance-oblong, acute, entire, glabrous, only the short, broad petiole fringed, and that loosely and coarsely; flowers copious, rather crowded at the ends of the branches; sepals ovate-lanceolate, not indistinctly feather-veined above the middle; seg- ments of the corolla nearly orbicular, quite as broad as long, shortly cuspidate- acute; capsule globose, much shorter than the calyx. Type in the U. 8S. National Herbarium, no. 45865, collected along Oak Creek, near Mlagstaff, Arizona, in July, 1884, by J. G. Lemmon and wife. The best specimens are from northern Arizona collected by MacDougal, Lemmon, and others. One sheet of not very good specimens from McKinneys Park in the Mogollon Mountains, collected by O. B. Metcalfe, seems to represents the species in New Mexico. OLEACEAE. Menodora laevis Wooton & Standley, sp. nov. A low perennial about 25 em. high, from a thick, woody root; stems slender, woody below, very numerous, simple below, corymbosely branched above, bright green and shining, glabrous, angled; leaves obovate to lanceolate, 15 mm. long or less, obtuse or acute, contracted at the base into a short petiole, glabrous; pedicels about 5 mm. long, glabrous; tube of the calyx 1.5 mm. high, the 7 to WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 159 10 lobes linear, acute, glabrous, 5 to 6 mm. long; tube of the corolla very short, the lobes obovate, 8 or 9 mm. long, acute; capsules glabrous, 7 or 8 mm. high. Type in the U. 8S. National Herbarium, no. 45767, collected in the Organ Mountains in August, 1881, by G. R. Vasey. ADDITIONAL SPECIMENS EXAMINED: La Luz Canyon, August 27, 1901, Wooton; Duck Creek Flats, 1903, Metcalfe 770. The type was distributed as VM. scoparia Engelm., but that has fewer, shorter calyx lobes. The species is more closely related to J/. scabra, but differs in being glabrous throughout instead of scabrous. GENTIANACEAE, Dasystephana rusbyi (Greene) Wooton & Stamdley. Gentiana rusbyi Greene; A. Gray, Syn. FI. 2': 406. 1878. APOCYNACEAE. Apocynum angustifolium Wooton, sp. noy. Stems slender, tall, much branched, the branches strongly ascending, glabrous ; leaves narrowly oblong-lanceolate, 7 to 10 cm. long, 1 to 2 cm, wide, glabrous, bright green on both surfaces, acute or rather abruptly acute, mucronate, acute at the base or the lowest rounded, all at least short-petioled, the upper with petioles 5 mm. long; cymes few, densely flowered, on slender peduncles 8 to 5 em. long; bracts small, linear-lanceolate, attenuate, somewhat scarious; calyx 1.5 mm. long or less, glabrous, the lobes lanceolate, acute; corolla greenish white, 3 to 3.5 mm. long, narrow, with short, erect lobes; follicles slender, 9 to 12 cm. long, 3 to 4 mm. in diameter, glabrous. Type in the U. 8. National Herbarium, no. 564822, collected in the Gila River bottom near Cliff, Grant County, June 13, 1903, by O. B. Metcalfe (no. 182). Altitude 1,350 meters. ADDITIONAL SPECIMENS EXAMINED: Mimbres, alt. 1,650 meters, July 1, 1904, Metcalfe 1070; Lower Plaza, July 25, 1900, Wooton; Eagle Creek, 1899, Turner 129. . A very distinct species of the cannabinum group, distinguished by its narrow, bright green leaves, nearly all of them acute at the base. Apocynum viride Wooton & Standley, sp. nov. Plant 1 meter high or more, with numerous erect, slender, glabrous branches ; leaves narrowly elliptic-oblong, bright green, glabrous, 6 to 10 cm, long, 15 to 30 mm. wide, acute or abruptly short-acuminate, rounded or acutish at the base, on slender petioles 3 mm. long or less; cymes few, compact, 80 to 385 mm. wide, many-flowered, on slender peduncles 15 to 30 mm. long; bracts linear, thin; calyx 1 to 1.5 mm. wide, glabrous, with lanceolate, acute lobes; corolla 2 to 3 mm. long, pinkish, glabrous, with short, erect lobes, Type in the U. S. National Herbarium, no. 499829, collected at Gilmores Ranch on Eagle Creek in the White Mountains, Lincoln County, August 25, 1907, by E. O. Wooton and Paul C. Standley (no. 3451). Altitude 2,220 meters. ADDITIONAL SPECIMENS EXAMINED: Farmington, alt. 1,550 meters, 1911, Stfand- ley 6970; Cedar Hill, alt. 1,900 meters, 1911, Standley 8003; Las Vegas, July, 1881, Vasey; near Pecos, alt. 2,010 meters, 1908, Standley 5044; Reserve, July 9, 1906, Wooton; Fresnal, July 21, 1899, Wooton; Sapello Creek, August 22, 1900, Wooton; Gilmores Ranch, July 27, 1901, Wooton. Similar to A. cannabinum L., but differing in its narrow, bright green, glabrous leaves, aS well as in other minor characters. 160 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. DICHONDRACEAE. Dichondra brachypoda Wooton & Standley, sp. nov. Perennial from a slender root; stems slender, creeping, 50 cm. long or less, seldom rooting at the nodes, villous; petioles slender, erect, 12 to 80 mm. long, villous; blades reniform, with broadly rounded lobes and a narrow sinus, emarginate, 2 to 3 cm. wide, bright green, pubescent on both surfaces, more densely so beneath; pedicels stout, 4 or 5 mm. long; sepals oblong-obovate, rounded at the apex, 3 to 4 mm. long, villous; capsules pubescent, 4.5 mm. high, much exceeding the calyx at maturity. Type in the U. S. National Herbarium, no. 564085, collected in Filmore Canyon in the Organ Mountains, high up in deep ravines, September 23, 1906, by E, O. Wooton and Paul C. Standley. Altitude about 1,800 meters, ADDITIONAL SPECIMENS EXAMINED: New MeExtco—Kingston, alt. 2.010 meters, 1904, Metcalfe 1377; Organ Mountains, 1890, Wooton; Queen, alt. 1,770 meters, July 31, 1909, Wooton; Mexican Boundary Survey 1005; 1851-2, Wright 1620. TEexas—1849, Wright 515, 516. A most distinct species, evidently related to D. caroliniana Michx. That, how- ever, has smaller leaves, much longer petioles and pedicels, and its capsules are shorter than the calyx. POLEMONIACEAE. Eriastrum Wooton & Standley, nom. noy. Hugelia Benth. Edwards's Bot. Reg. 19: pl. 1622. 1833, not DC. 1830. Gilia section Hugelia A. Gray, Proc. Amer. Acad. 8: 271. 1870. Eriastrum filifolium (Nutt.) Wooton & Standley. Gilia filifolia Nutt. Journ. Acad. Phila. n. ser. 1: 156. 1848. Navarretia filifolia Brand in Engl. Pflanzenreich 27: 167. 1907. Gilia brachysiphon Wooton & Standley, sp. nov. Perennial from a usually slender root; stems stout, 15 to 50 em. high, cinereous-tomentulose, simple or branched; leaves petiolate, pinnately parted into linear, spinulose-tipped segments, cinereous to glabrate; inflorescence thyrsiform, often short and somewhat congested, in age elongated, the flowers collected in small, pedunculate, close clusters; calyx 8 to 4 mm. long, villous, sparingly viscid, with lanceolate-subulate, spinescent lobes nearly as long as the tube; corolla bluish, 8 to 10 mm, long, the tube usually not at all exserted, about equaled by the oblong, apiculate lobes; stamens exserted ; capsules obtuse, 4 mm. long. Type in the U. S. National Herbarium, no. 561092, collected at Van Pattens Camp in the Organ Mountains, August 29, 1894, by FE. O. Wooton. The plant is not uncommon in this range, growing on slopes near the summit, in the thin shade of yellow pines. ADDITIONAL SPECIMENS EXAMINED: Kingston, alt. 1,980 meters, 1904, Metcalfe 1269; near Carlisle, August 13, 1902, Wooton; mountains southeast of Patter- son, August 16, 1900, Wooton; Filmore Canyon, September 20, 1908, Wooton & Standley ; Organ Mountains, September 4, 1898, Cockerell, September 23, 1906, Wooton & Standley, in 1897, Wooton 440. This is related to G. multiflora, but the corolla is very different, the tube in that species equaling the lobes, while in G. brachysiphon it is often shorter. Gilia campylantha Wooton & Standley, sp. nov. Perennial with numerous clustered stems from running rootstocks; stems tall, 30 em. high, very slender, densely tomentulose; leaves numerous, pecti- WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXiIco. 161 nately parted into linear, abruptly acuminate, aristate-tipped lobes, tomentose or tomentulose, petiolate; flowers numerous, in few-flowered pedunculate or sessile clusters arranged in a secund, raceme-like thyrse; pedicels short; calyx 83 mm. long, green, only slightly scarious, puberulent, cleft to the middle or lower into triangular-subulate, pungently pointed lobes; corolla white, the tube about 8 mm. long, thick, sharply bent downward just above the calyx, the lobes elliptic-oblong, obtuse, narrowed at the base into a short claw, about 4 mm. long. Type in the U. 8S. National Herbarium, no. 233429, collected in the San Luis Mountains, September 11, 1893, by Dr. E, A. Mearns (no. 2242). A remarkable species, distinguished from all the related ones by its small, white flowers and the peculiarly formed corolla tube. Otherwise it suggests G. glomeriflora Benth., but that has a very different calyx besides. Gilia formosissima (Greene) Wooton & Standley. Callisteris formosissima Greene, Leaflets 1: 160. 1905. Batanthes formosissima Greene, op. cit. 224. 1906. Gilia greeneana Wooton & Standley, nom. nov. Callisteris collina Greene, Leaflets 1: 159. 1905, not Gilia collina Eastwood. Batanthes collina Greene, op. cit. 224. 1906. Gilia attenuata collina Cockerell, Univ. Mo, Stud. Sci. 27: 197, 1911. Gilia texana (Greene) Wooton & Standley. Callisteris texana Greene, Leaflets 1: 160. 1905. Batanthes texana Greene, op. cit. 224. 1906. Gilia viscida Wooton & Standley, nom. nov. Gilia pinnatifida Nutt.; A. Gray, Proc. Amer. Acad. 8: 276. 1870, not Moc. & Sessé, 1837. Phlox grayi Wooton & Standley. Phlox longifolia stansburyi forma brevifolia A. Gray, Proc. Amer. Acad. 8: 255. 1870. Phlox longifolia brevifolia A. Gray, Syn. Fl. 2': 183. 1878, not P. brevifolia Baum. This plant is distinct enough from both P. longifolia and P, stansburyi to receive specific rank; in fact, it is much more distinct from those species than are most of the members of the genus from each other. It is marked chiefly by its lower habit, more branched stems, and especially by the short, broad, rather crowded leaves. The corollas of our New Mexican specimens have a very small limb, only about 13 mm. wide and each lobe of the limb is retuse rather than rounded, as in most of our species. Phlox tenuis Wooton & Standley, sp. nov. Slender perennial 15 to 20 cm. high; stems numerous, erect, tortuous, finely villous, slightly glandular; internodes long; leaves linear, 35 to 65 mm. long, divaricate, acute, villous; pedicels slender, 12 to 30 mm. long, glandular-villous ; calyx 12 to 14 mm. high, densely glandular-villous, the linear-subulate lobes about as long as the tube; corolla tube only slightly exceeding the calyx; corolla lobes cuneate-oblanceolate, 13 to 14 mm. long, 5 mm. wide, broadly rounded at the apex; fruit not seen. Type in the U. 8. National Herbarium, no. 306405, collected at Barranca, Taos County, May 26, 1897, by A. A. and E. Gertrude Heller (no. 3589). Alti- tude 2,070 meters. ; On the same sheet with the type, and distributed under the same number, are plants of what appear to be P. stansburyi. The collection was distributed 162 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, under that name. Our plant, however, is very unlike P. stansburyi, having the corolla tube only slightly longer than the calyx instead of twice as long, while the leaves are larger and more slender, and the corolla lobes longer and narrower. Baker’s no. 60 from Cerro Summit, Colorado, is apparently the same species, HYDROPHYLLACEAE. Marilaunidium foliosum Wooton & Standley, sp. nov. Annual, 12 to 30 em. high; stems with very numerous, dense, spreading branches, these rather stout, hirsute and puberulent; leaves obovate to oblanceo- late or broadly oblong, very numerous especially about the inflorescence, obtuse, acute at the base, sessile, flat, glandular throughout, hispid on the upper sur- face; flowers very numerous, in terminal or axillary clusters, sessile or nearly so; calyx 5 or 6 mm. long, the linear lobes coarsely hirsute; corolla about 7 mm. long, purple, the very short tube much exceeded by the calyx, the lobes broad and rounded; seeds very numerous, minute, ovoid, with a thin, rugulose coat. Type in the U. S. National Herbarium, no. 480933, collected by F. S. Earle on saltgrass flats near Roswell, August 30, 1900 (no, 531). ADDITIONAL SPECIMENS EXAMINED: Roswell, alt. 1,140 meters, 1900, Harle 558; near Lake Arthur, August 1, 1905, Wooton; Fort Stanton, July 26, 1905, Wooton. This was distributed as Nama stenocarpum A. G ‘ay, but that is a prostrate plant with decurrent leaves and different pubescence. It is more closely re- lated, perhaps, to Marilaunidium hispidum, but that, too, has more abundant pubescence, narrower leaves, and different corollas, Marilaunidium tenue Wooton & Standley, sp. nov. Annual; stems slender, with few erect branches, 9 cm. high or less, puberulent and hirsutulous; leaves few, linear-spatulate, obtuse, 16 mm. long or less, nar- rowed at the base into a short petiole, hirtellous, glindular, the margins revo- lute; flowers few, axillary or cymulose, on slender pedicels 1 to 4 mm. long; calyx 3 mm. long in anthesis, increasing in size in age, the lobes linear, green, hirsute; corolla 4 to 5 mm, long, light blue, with a thick tube and narrow limb ; capsules glabrous, two-thirds as long as the calyx; seeds few, brown, pitted. Type in the U. 8. National Herbarium, no. 498090, collected on limestone hills 3 miles south of Hillsboro, Sierra County, by O. B. Metcalfe, September 6, 1904 (no, 1291). Altitude 1,500 meters, This was originally determined as Conanthus demissus (A. Gray) Heller, with which it has nothing to do. It is more closely related to Marilaunidium angustifolium, but is a lower, more slender plant, scarcely at all viscid, with longer corolla and much shorter calyx, and shorter floral leaves. Marilaunidium xylopodum Wooton & Standley, sp. nov. Perennial from a very thick, woody root; stems numerous, slender, sparingly branched, ascending, 8 em. long or less, hispid; leaves numerous, oblanceolate or oblong, acute, mostly flat, attenuate to the base, glandular, hispid; flowers few, cymulose, short-pedicellate; calyx 2.5 to 4 mm. long, with linear-subulate, hispid lobes; corolla purple, 6 or 7 mm, long, with a very thick tube not ex- ceeding the calyx, the lobes somewhat spreading, rounded; capsule about half as long as the calyx. Type in the U. 8. National Herbarium, no. 564567, collected in crevices of limestone rocks near Queen, July 31, 1909, by E. O. Wooton. Altitude about 1,770 meters. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 163 We have another specimen of the species collected in the same range of mountains (the Guadalupes) but across the line in Texas, in October, 1881, by Dr. V. Havard. This differs from all our other New Mexican species in being perennial. It does not seem to be closely related to any of the perennial species found else- where. Phacelia bombycina Wooton & Standley, sp. nov. Annual, 12 to 20 cm. high; stems few, stout, nearly scapose, hirsute and glandular-puberulent; basal leaves on stout petioles 4 to 5 cm. long, obtuse, pinnatifid into oblong-ovate, rounded, crenate lobes, sericeous and glandular on both surfaces; cauline leaves few or absent. petiolate, lobed; inflorescence narrow, With few branches; flowers numerous, crowded, on very short, stout pedicels; sepals oblong, obtuse, 2.5 mm. long or less, glandular-hirsute; corolla 5 to 6 mm, long, the lobes rounded, entire; stamens much exserted; styles hairy below; capsules subglobose, 2 to 2.5 mm. in diameter, hirtellous and glandular; seeds 1.5 mm. long, dark brown, finely pitted on the back. Type in the U. S. National Herbarium, no. 45771, collected on “ gravelly banks” at Mangas Springs in March or April, 1880, by H. H. Rusby (no. 276). ADDITIONAL SPECIMENS EXAMINED: Bear Mountains, alt. 1500 meters, 1903, Metcalfe 75. » Similar in general appearance to P, intermedia, but with hirsute pubescence, silky leaves, and narrower calyx lobes, Phacelia depauperata Wooton & Standley, sp. nov. Annual, 6 cm. high or less; stems erect, slender or stout, simple or branched at the base, densely glandular-puberulent and hirsute; leaves linear-oblong in outline, 35 mm. long or less, once pinnatifid, the oblong-oval segments crenate or lobed, obtuse, scaberulous and glandular on both surfaces; inflorescence short and dense; pedicels stout, less than 1 mm. long; calyx 2 mm. long, the segments rounded-obovate, hirsute and glandular-puberulent; corolla 5 mm. long, the lobes broadly rounded, undulate-margined ; stamens much exserted. Type in the J. S. National Herbarium, no. 496292, collected on the Arroyo Ranch, near Roswell, in 1903 by David Griffiths (no. 4249). This was determined as P. arizonica, but it differs from that species in its much longer flowers, the form of the calyx lobes, and the character of the pubescence. Phacelia tenuipes Wooton & Standley, sp. nov. Annual, 15 to 20 em. high; stems slender, with a few ascending branches above the base, hirsute and sparingly glandular; lower leaves broadly oblong, with a few small lobes near the base, the terminal part lobed and crenate, sparingly sericeous or with spreading pubescence on both surfaces and glandu- jar, all the leaves petioled; upper leaves rather numerous, ovate, obtuse, crenate or lobed; inflorescence open, slender, few-flowered, the flowers not crowded, at least in age; pedicels slender, 2 or even 3 mm. long; calyx lobes oblong, obtuse, 2.5 mm. long, hirsute, glandular; corolla about 4 mm. long, the lobes rounded, entire; stamens well exserted; style hirsute near the base; capsules globose, shorter than the sepals, hirtellous, glandular. Type in the U. S. National Herbarium, no. 45770, collected at Carrizalillo Spring, April 17, 1892, by Dr. E. A. Mearns (no. 91). Related to P. bombycina, but a more slender planf. with much longer pedicels, fewer flowers and very different leaves. 164 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. EHRETIACEAE. Eddya gossypina Wooton & Standley, sp. nov. Perennial from a somewhat woody base; stems stout, prostrate, 20 cm. long, densely canescent; leaves spatulate, thick and fleshy, 8 mm. long or less, on slender petioles, obtuse, the blades hispid and short-pubescent, the petioles densely canescent; floral leaves crowded, densely white-villous; flowers few, axillary, sessile or nearly so; calyx lobes linear-lanceolate, white-villous, 3 mm. long; corolla purplish, 10 mim. long. with a spreading limb, the rounded lobes with nearly entire margins; stamens unequally inserted, unequal, the bases of the filaments much enlarged and filling the tube of the corolla; style 2-parted at the top, glabrous; mature fruit not seen. Type in the U. 8S. National Herbarium, no. 690234, collected on Tortugas Mountain, September 2, 1894, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Tortugas Mountain, 1911, Standley 6439. In general this resembles HF. hispidissima, but has a white, cottony appear- ance very different from that species, while the corolla is much larger and the stamens are somewhat different. BORAGINACEAE. Lappula grisea Wooton & Standley, sp. nov. Biennial from a short, thick taproot; stems stout, 80 em. high or less, simple below, canescent; basal leaves numerous, narrowly oblanceolate, obtuse, 3 to 5 cm. long; cauline leaves narrowly oblanceolate to linear-oblong, obtuse or acute, 25 to 40 mm. long, the upper sessile, the lower attenuate to a broad, winged petiole; all densely canescent on both surfaces, the hairs with large, white, bulbous bases; inflorescence sparse, with slender, loosely few-flowered, erect ‘branches; bracts very small, lanceolate; pedicels 3 to 4 mm. long; sepals ovate, obtuse; corolla blue, small, 2.5 mm. long; nutlets small, 2.5 mm. long, the prickles as long as the body, free to the base, the back of the nutlets papillose. Type in the U. S. National Herbarium, no. 562188. collected in James Canyon of the Sacramento Mountains, August 6, 1905, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Tularosa Creek, alt. 2040 meters, July 30, 1897, Wooton 252. Apparently a recognizable species, distinguished from the related L. flori- bunda by its grayish appearance, caused chiefly by the enlarged white bases of the hairs. The pubescence, too, is more abundant, longer, stiffer, and harsher than in any of the related species. Lappula hirsuta Wooton & Standley, sp. nov. Biennial from a short tap-root; stems stout. erect, a meter high or less, coarsely appressed-pubescent, more or less hirsute above, simple up to the in- florescence; basal leaves narrowly spatulate, obtuse, long-petiolate, 10 em. long or less; cauline leaves narrowly oblanceolate to oblong, 5 to 8 cm. long, acute, sessile by a rounded base, or the lower ones attenuate to a winged petiole, canescent or appressed-pubescent on both surfaces, glandular, long-ciliate; in- florescence with numerous slender, ascending branches, the racemes long and distantly flowered; bracts elliptic or lanceolate, small; pedicels 4 to 8 mm. long; sepals narrowly oblong. obtuse; corolla deep blue, 8 mm. broad, the lobes orbicular, the appendages finely papillose; marginal bristles of the nutlets sepa- rate to the base. about equaling the body, this nearly smooth on the back but minutely hirtellous. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 165 Type in the U. S. National Herbarium, no. 306419, collected by A. A. and E. Gertrude Heller in Santa Fe Canyon, 9 miles east of Santa Fe, July 2, 1897 (no. 3793). Altitude about 2,400 meters; Transition Zone. ADDITIONAL SPECIMENS EXAMINED: Gallinas Planting Station, 1908, Bartlett 302; mountains near Las Vegas, July, 1881, Vasey; Winsors Ranch, alt. 2.520 meters, 1908, Standley 4106; 1847, Fendler 6383; Beulah, 1899, Porter; Water Canyon, 1897, Herrick 750; Upper Pecos River, September, 1904, Bartlett, in 1898, Maltby & Coghill 66; Beulah, 1899, Helen Blake; Mogollon Creek, alt. 2,400 meters, 1908, Metcalfe 249. Related to ZL. floribunda, but differing in its hirsute pubescence and larger flowers. Mr. C. V. Piper was the first to suggest’ that this might be a distinct species. Fendler’s specimens were referred by Doctor Gray’ to LZ. ursina, but this has nothing to do with that species. Mertensia amplifolia Wooton & Standley, sp. nov. Perennial from a thick rootstock; stems several, erect, stout, 25 to 35 em. high, retrorsely appressed-pubescent, simple, or branched above the base, with loosely ascending branches above; basal leaves lance-oblong, 11 em. long and 5 cm. wide or smaller, acutish, unequal at the base and somewhat decurrent upon the slender petiole, 9 cm. long or less; cauline leaves very numerous, the upper ones little reduced, lanceolate, sessile, or the lowest petiolate, 5 to 9 cm. long, acute, rounded at the base, all the leaves finely appressed-pubescent on both surfaces; inflorescence ample, very leafy, the branches slender, few- flowered, strigose; pedicels slender, about 10 mm. long; calyx in anthesis 5 mm. high, in age about twice as long, cleft to about the middle, the lobes triangular- lanceolate, acute, strigose; corolla 10 mm. long, the limb about 4 mm. and the tube 1.5 mm. wide, the limb and tube about equal in length, the throat with a ring of short hairs; nutlets 2 to 3 mm. long, dark brown, minutely papillose. Type in the U. S. National Herbarium, no. 45768, collected at Glorieta, June, 1881, by G. R. Vasey. Additional material of the same collection is mounted on sheet 45769. Our plant is related to M. bakeri Greene, but is larger, has larger leaves of different shape, is less pubescent, and has much smaller flowers. Mertensia grandis Wooton & Standley, sp. nov. Perennial from a slender rootstock; stem stout, fleshy, erect or ascending, glabrous, 40 to 100 cm. high; leaves ovate to lanceolate, rather fleshy, acute, rounded at the base, 8 to 18 cm. long, the basal long-petioled, the cauline sessile, the uppermost reduced, glabrous beneath and glaucescent, scaberulous on the upper surface; inflorescence rather ample, the slender branches white-strigose ; pedicels stout, 8 to 12 mm. long, strigose; calyx 4 to 5 mm. high, cleft to near the base, the lobes oblong, obtuse, glabrous or strigose, ciliate, the tube strongly strigose, the whole enlarging in age; corolla bright blue, 12 to 17 mm. long, the limb 5 to 9 mm. wide, much shorter than the tube, the latter 2.5 to 5 mm. thick, the throat with a ring of short hairs. Type in the U. S. National Herbarium, no. 498109, collected on a shaded slope of Hillsboro Peak, in the Black Range, September 11, 1904, by O. B. Metcalfe (no. 1819). Altitude 2,550 meters. ADDITIONAL SPECIMENS EXAMINED: Holts Ranch, July 20, 1900, Wooton ; Mogollon Mountains, August, 1881, Rusby 291; Hagle Peak, August 2, 1900, Wooton; Mogollon Creek, alt. 2,250 meters, July 28, 1903, Metcalfe 302. *Bull. Torrey Club 29:545. 1902. * Syn. FI. 2':422. 1878. 166 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Related to M. ciliata (Torr.) Don, but with a larger, differently formed and abundantly pubescent calyx. It is a larger plant, too, than most of our western Mertensias. It is related also to J/. franciscana Heller, but the corolla and calyx are very different. Oreocarya urticacea Wooton & Standley, sp. nov. Perennial from a woody caudex covered by the bases of the dead leaves; stems very stout, about 30 cm. high, hispid and densely canescent, simple up to the inflorescence; basal leaves spatulate, rounded at the apex, 7 to 9 cm. long, petiolate, green but hirsute and finely canescent; cauline leaves narrowly oblan- ceolate, obtuse, attenuate to a winged petiole, very numerous; inflorescence a dense, narrow thyrse with numerous linear-lanceolate or linear-oblong, strongly hispid leaves, its branches densely setose-hispid with yellowish hairs; lateral glomerules of the inflorescence on peduncles 10 to 35 mm. long, the flowers in very dense clusters; calyx lobes nearly linear, 6 mm. long, densely hispid; corolla white, 5 or 6 mm. long, the tube shorter than the calyx; mature fruit not seen, Type in the U. 8. National Herbarium, no. 306425, collected at Canyoncito, Sante Fe County, June 18, 1897, by A. A. and E. Gertrude Heller (no. 3781). Altitude 2,160 meters. ADDITIONAL SPECIMENS EXAMINED: Glorieta, June, 1881, Vasey; Sierra Grande, alt. 2,200 meters, 1911, Standley GO56. The type collection was distributed as O. glomerata (Pursh) Greene, but that is a lower, rather less hispid plant with a dense, spike-like inflorescence and broader leaves. Our plant is also related to O. perennis (A, Nels.) Rydb., but differs from that species in about the same manner as from O. glomerata. VERBENACEAE. Verbena imbricata Wooton & Standley, sp. nov. Erect plant, 35 em. high, with numerous strict, herbaceous stems from the crown of a stout, woody root, sparingly hispid throughout; leaves cuneate- obovate in outline, 3 em. long or less, pinnately 3-lobed, the middle lobe largest and pinnately toothed or lobed, decurrent below into a short petiole; flowers fn crowded, bracted spikes resembling those of V. bracteosa but more crowded and the bracts shorter; bracts lanceolate, 5 to 10 mm. long; fruiting calyx barely 3 mm. long; corolla deep blue, about 4 mm. long, shorter than the bracts; nutlets as in V. bracteosa. Type in the U. 8S. National Herbarium, no. 562249, collected at’ Farmington, August 8, 1904, by E, O. Wooton (no. 2831). This may possibly be a hybrid between V. bracteosa and some erect species, but this seems improbable, since no species is found in that general region that would be likely to hybridize with V. bracteosa to produce such a plant as this. MENTHACEAE. Agastache cana (Hook.) Wooton & Standley. Cedronella cana Hook. Curtis’s Bot. Mag. pl. 4678, 1851. Cedronelia cana lanceolata A. Gray, Syn. Fl 21: 462. 1878, in part. Brittonastrum lanceolatum Heller, Mublenbergia 1: 4. 1900. SPECIMENS EXAMINED: Headwaters of the Pecos, August, 1905, Bartlett; hills near Santa Rita, October 9, 1904, Metcalfe; Hillsboro, alt. 1,650 meters, 1904, Jfetcaife 1015; Organ Mountains, August, 1881, Vasey; Dona Ana Moun- tains, October 28, 1906, Wooton & Standley; Organ Mountains, alt. 1,660 meters, WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 167 1897, Wooton 437; San Augustine Ranch, alt. 1,740 meters, August 16, 1895 Wooton; Soledad Canyon, September 24, 1905, Wooton. This is without doubt the plant which Hooker described and figured as Cedronella cana. The locality cited in Wright’s field notes indicates that the type was collected in the low mountains, probably the Huecos, about 30 miles east of El Paso, possibly near the Hueco Tanks, which were then a watering place on the route leading across the plains from El Paso, the one probably fol- lowed by Wright’s party. The Organ Mountains, where we first found this plant, are the next chain to the northwest, and only a little farther away than El Paso. The plant collected by G. R. Vasey and referred to by Doctor Gray in his description of Cedronella cana lanceolata is C. cana from the Organ Moun- tains, in all probability. This is the only Agastache in the National Herbarium collected by Vasey, and this herbarium should contain specimens of all his collections. We have not seen the Wright specimen referred to by Gray in his description of lanceolata, which was probably his no. 1533, collected at Santa Rita. The Rusby plant from Mangas Springs that deserves the name of lanceolata is Cedronella rupestris, an excellent species which has been collected several times since in the immediate region. The Burro Mountain plant col- lected by Bigelow is also that species. Agastache greenei (Brig.) Wooton & Standley. Brittonastrum greenei Briq. Ann. Cons. Jard. Genéve 6: 157. 1902. We do not agree with Doctor Rydberg in considering this a synonym of A. pallidifora (Heller) Rydb. A rather extended series of the Arizona plant from various localities and also from western New Mexico in the Mogollon Mountains, shows that it has a green calyx and pale whitish flowers, while A. greenei has the calyx teeth and upper part of the tube decidedly pink or pur- plish tinged and the corolla also brighter colored. Besides these more con- spicuous calyx differences, the corollas in A. greenei are noticeably longer, more arched, and wider at the throat. These differences in color and size of corolla and calyx and noticeable ones in the calyx teeth seem to be the most important diagnostic characters in a group of closely related but distinct species which have until recently been considered as belonging to two or three very variable ones. Agastache mearnsii Wooton & Standley, sp. nov. Herbaceous, branched perennial, about 70 cm. high, with several erect stems from the base, finely puberulent throughout; stems distinctly quadrangular, suleate above, below almost terete; petioles mostly 10 to 12 mm. long; blades 25 to 50 mm. long, about two-thirds as broad, triangular to hastate, sometimes ovate, mostly truncate at the base, acute to abruptly short-acuminate, coarsely crenate-dentate, finely and closely puberulent beneath and pale; flowers numer- ous, in crowded, terminal, spike-like clusters 10 cm. long and 3 to 4 cm. wide when in full flower; peduncles and pedicels 2 or 3 mm. long, the linear-subulate bracts about 5 mm. long; calyx tubular, reddish purple, 10 to 12 mm. long, the triangular-subulate, almost equal teeth fully 2 mm. long, erect; corolla about 20 mm. long, reddish purple, somewhat arched, puberulent outside, the upper lip retuse, the lower 3-lobed; stamens 4, the longer pair exceeding the corolla; nutlets brown, scabrous at the apex. Type in the U. S. National Herbarium, no. 233421, collected in a canyon on the east side of the San Luis Mountains in extreme southwestern New Mexico, nearly on the Mexican boundary line, September 11, 18938, by Dr. HE, A. Mearns (no. 2251). 60541°—13——5 168 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. ADDITIONAL SPECIMENS EXAMINED: Animas Valley, 1893, Mearns 2502; Burro Mountains, alt. 2190 meters, 1908, Goldman 1517; Pinos Altos Mountains, Sep- tember 10, 1880, Greene. This is one of the species of southern New Mexico and northern Mexico which has been passing as Cedronella mexicana, a much larger plant with larger flowers from some hundreds of miles farther south. It is nearest to Brittonastrum barberi and B. ionocalyr Robinson, both of which have larger corollas, the former a longer calyx and the latter a shorter one, and both much broader and relatively shorter calyx teeth. The species is named for Dr. E. A. Mearns, whose extensive collections along the Mexican boundary have added records of many little known species to the flora of New Mexico. Agastache micrantha (A. Gray) Wooton & Standley. , Cedronella micrantha A. Gray, Proc. Amer. Acad. 8: 369, 1872. Agastache verticillata Wooton & Standley, sp. nov. Perennial herb, 40 to 80 cm. high, branched, puberulent throughout, green, the younger stems sulcate-quadrangular, slightly scabrous; leaves ovate-lanceolate, the blades 2 to 4 cm, long, on petioles 10 to 15 mm. long, rather thin, slightly paler beneath, the margins distantly and coarsely crenate-dentate, broadly cuneate or truncate at the base, the apex acute and somewhat attenuate in the upper leaves; flowers small for the genus, in an interrupted, verticillate spike 10 to 12 cm. Jong, the verticels crowded, the internodes slightly longer than the fruiting calyces; peduncles and pedicels very short, 1 to 2 mm., the linear bracts but little longer; calyx tubular. 5 to 6 mm. long, the subulate lobes about one-fourth as long, unequal, two of them shorter than the others, the tube green below, the lobes purple; corolla slender, 10 to 12 mm. long, pale, slightly, if at all, curved; nutlets smooth. Type in the U. S. National Herbarium. no. 561455, collected in the Organ Mountains, altitude about 2,250 meters, September 23, 1906, by E. O. Wooton and Paul C. Standley. ADDITIONAL MATERIAL EXAMINED: Organ Mountains, August, 1881, Vasey; Organ Mountains, August 16, 1895, Wooton; Filmore Canyon, October 23, 1904, Wooton; West Fork of the Gila, altitude 2.250 meters, 1908, Metcalfe 348; West Fork of the Gila, August 6, 1900, Wooton; Old Tiptop. Organ Mountains, Octo- ber 18, 1903, Metcalfe. This species is nearest A. wrightii, but is easily Separated. The flowers are about twice as large, the calyx lobes are purple instead of white, and the leaves are larger and have fewer teeth. The dried calyces of the different species of the genus have different odors when pulverized. This species gives a pro- .houned odor of camphor. Agastache wrightii (Greenman) Wooton & Standley. ; Cedronella wrightii Greenman, Proc. Amer. Acad. 41: 244. 1905. Brittonastrum wrightii Robinson, Proc. Amer. Acad. 48: 26. 1907. Hedeoma pulcherrima. Wooton & Standley, sp. nov. Perennial herb, much branched, from a woody root, the stems slender, dif- fusely ascending. finely and sparingly puberulent throughout, about 30 em. high; leaves opposite, elliptic-oblong, 2 cm. long or less, the largest 7 mm. wide, taper- ing below into a short petiole, rather thin, green, obtuse, entire; flowers large for the genus, in 1 to 5-flowered cymose clusters in the axils of the leaves, the upper internodes and leaves somewhat reduced; floral bracts linear-lanceolate, hardly longer than the short (1 to 2 mm. long) pedicels; calyx 6 to 7 mm. long, 15-ribbed, slightly constricted above the middle, sparingly hispidulous, the three upper lobes triangular-lanceolate, about 1 mm. long, the two lower ones subu- WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXxiIco. 169 late, about twice as long, slightly curved upward, all of them sparingly bearded in the throat; corolla bluish purple, 12 to 14 mm. long, narrowly funnelform, gaping, finely pubescent outside, the upper lip erect, retuse, the lower lip 3-lobed, the central lobe largest; stamens 2, surpassing the upper corolla lobe; nutlets oval, dark brown, smooth, acute. Type in the U. 8. National Herbarium, no. 330465, collected in the White Mountains, Lincoln County, altitude about 1,950 meters, July 30, 1897, by E. O. Wooton (no. 241). ADDITIONAL SPECIMENS EXAMINED: Cloudcroft, June 30, 1899, Wooton; Tula- rosa Creek, August 19, 1899, August 6, 1901, Wooton; Toboggan, July 31, 1899, Wooton; Dark Canyon, 1907, Wooton & Standley 3480; Mescalero Reservation, July 21, 1905, Wooton. The plant is most nearly related to H. jucunda Greene, from Durango, Mexico, and was distributed under that name, having been determined from the descrip- tion alone. Ours, however, is a stouter plant, with flowers about half again as large, and very different calyx. The species seems to be on the dividing line between the two principal groups of our species: its upper calyx lobes are not foliaceous-expanded as in one group, nor narrowly subulate as in the other, but are triangular-lanceolate, and they are hispid-ciliate like those of the second group. In habit and general appearance our plant resembles H. jucunda, but it is a little more leafy and has larger and fewer flowers. Salvia earlei Wooton & Standley, sp. nov. Tall, slender, herbaceous perennial, 30 to 100 em. high, branched above, the stems erect, sometimes 3 or 4 from a single root, finely puberulent throughout; leaves linear to narrowly oblong-lanceolate, 4 to 8 cm. long (mostly about 5 em.) including the petiole, acute, cuneate and decurrent at the base into a petiole about 1 cm. long, entire or undulate with occasionally a few obscure teeth, glabrate but finely glandular above; flowers in terminal, interrupted, glomerate clusters, about 12 flowers in each verticel; calyx 6 to 8 mm. long, campanulate-tubular, obscurely 8 to 10-ribbed, becoming somewhat urceolate in fruit, densely white or blue tomentose with very fine short hairs, the limb short-trunecate, obscurely 2-lipped, the upper lip entire, the lower with 2 minute teeth; corolla bright blue to almost violet, about twice as long as the calyx, the upper lip notched, erect, tomentose on the back, the lower lip 3-lobed, the middle lobe much the largest; style densely bearded with bluish hairs; nutlets smooth, brown. ; Type in the U. S. National Herbarium, no. 382519, collected 35 miles west of Roswell, in August, 1900, by F. S. and Esther S. Earle (no. 375). ADDITIONAL SPECIMENS EXAMINED: NEw MExico—Twenty miles west of Ros- well, July 28, 1905, Wooton; South Berendo River near Roswell, June, 1899, Tinsley; Sixteen Spring Canyon, August 23, 1901, Wooton. Trexas—Toyah Creek, 1902, Tracy & Earle 1388; southwestern Texas, 1880, Palmer 1066; Frio Water Hole, Edwards County, 1895, R. T. Hill 36; Tom Green County, 1880, Tweedy 255; Baird, 1882, Letterman 72; Ballinger, 1889, Nealley 391a. The plant has long been confused with S. farinosa Benth., which it resembles very closely as to its flowers, although these are smaller, but the leaves are of an entirely different character, much more like those of S. pitcheri Torr. The distribution area is farther west than that of the one and farther south than that of the other, in a climate much more arid than either of the others endures. Salvia pinguifolia (Fernald) Wooton & Standley. Salvia ballotaeflora pinguifolia Fernald, Proc. Amer. Acad. 35: 523. 1900. The type was collected by Wright in 1851 or 1852 (no. 1524), probably in New Mexico somewhere near the Copper mines of Santa Rita. 170 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. This is distinguished from 8. ballotaeflora by the very large, scarcely rugose, broad leaves, pale beneath because of the presence of abundant whitish pubes- cence, and by the denser and more ample inflorescence. The species has been collected at various times in southern New Mexico and southeastern Arizona. Salvia vinacea Wooton & Standley, sp. nov. A low, suffrutescent plant, about 30 em. high; stems slender, sparingly tomen- tulose above; leaves broadly ovate, about 30 mm. long and 25 mm. wide, rather obtuse, coarsely crenate, mostly truncate at the base or slightly decurrent, glabrous above and almost so beneath but slightly puberulent, of about the same color on both surfaces; petioles slender, half as long as the blades or more; flowers in short, congested spikes, numerous; calyx ampliate in age, its lobes very broad and obtuse, the whole calyx 12 or 13 mm. long, of a deep wine color, greenish at the base, on a short, deflexed pedicel; corolla dark blue, 20 to 22 mm. long, its tube much exceeding the calyx, the upper lip oblanceolate. Type in the U. 8. National Herbarium, no. 499430, collected in the Florida Mountains at an altitude of about 1,950 meters, September 8, 1908, by E. A. Goldman (no, 1501). ADDITIONAL SPECIMENS EXAMINED: Florida Mountains, September 7, 1903, M. E, Jones; Martins Spring, Florida Mountains, 1895, Mulford 1067. Nearest S. pinguifolia, but differing in its much larger corolla with a nar- rower upper lip, wine colored calyx, and more congested inflorescence, and in not having its leaves conspicuously whitened beneath. Tetraclea angustifolia Wooton & Standley, sp. nov. Herbaceous perennial with several branching, obscurely 4-angled, ascending or spreading stems 40 to 50 cm. high from a woody root, the whole plant scabrous with short, stout, recurved, whitish hairs, these most abundant on the stems; leaves narrowly oblong, tapering into a short petiole, serrate-dentate with a few coarse teeth on each side, acute, the hairs mostly on the petioles, veins, and margins of the leaves; flowers in few-flowered axillary clusters with narrowly linear bracts; calyx campanulate, the lobes narrowly lanceolate, acuminate in flower, accrescent and persistent in fruit; corolla and stamens as in 7. coulteri but smaller and the tube narrower; nutlets slightly more reticulate and of the same size. Type in the U. 8S. National Herbarium, no. 330627, collected on the plains south of the White Sands, August 238, 1897, by E. O. Wooton (no. 403). We have one other specimen collected from the same locality, August 26, 1899, by EE. O. Wooton. This plant is more slender, taller, less pubescent and with shorter hairs, and has narrower calyx lobes, smaller corolla, and more reticulate nutlets than the only other species of the genus, 7. cowlteri. The leaves, too, are narrower and toothed. SOLANACEAE. Androcera novomexicana (Bartlett) Wooton & Standley. Solanum heterodowum novomexricanum Bartlett, Proc. Amer. Acad. 44: 628. £909. Type collected by Fendler in New Mexico, doubtless near Santa Fe, in 1847 (no. 673). Although Solanum heterodorum has been reported from New Mexico at various times, it is to this species that all such specimens belong. The plant, while seldom abundant in any one locality, has a rather wide range in New Mexico, having been collected in the region about Santa Fe and as far south as Santa Rita. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 171 ° SCROPHULARIACEAE. Castilleja eremophila Wooton & Standley, sp. nov. Perennial from a slender root, the stout, solitary stems 10 to 15 cm. high, cinereous-puberulent; a few of the lowest leaves linear-oblanceolate, the others pinnatifid, each with 1 or 2 pairs of oblong-linear, divergent or ascending, obtuse lobes, all dull green, thin, cinereous-puberulent, sessile, not conspicuously veined ; flowers rather few, crowded; bracts narrow, acute, usually with several linear lateral lobes, glandular-puberulent or slightly villous; calyx 16 to 20 mm. long, about equally cleft before and behind, the lateral divisions each cleft at the summit into 2 oblong-lanceolate, acute segments, glandular and villous, scarlet at the tips; corolla about 25 mm. long, the short tube about 10 mm. and the galea 15 mm. long, the tube glabrous, greenish, the galea puberulent, yellowish green faced with scarlet, the lower lip of 2 glabrous, green, oblong- lanceolate, acute teeth about 1 mm. long. Type in the U. S. National Herbarium, no. 687232, collected on arid, sandy mesas about the north end of the Carrizo Mountains, July 31, 1911, by Paul C. Standley (no. 7464). Additional material is mounted on sheet 686431. The plant is somewhat like C. chromosa A. Nels., but is much lower and has always solitary stems; the pubescence is very different besides. It grows in the most arid situations in the Upper Sonoran Zone. No other Castilleja was found growing at so low an altitude, C. integra having a habitat most nearly approaching that of this species. Dasystoma wrightii (A. Gray) Wooton & Standley. Gerardia wrightit A. Gray, U. S. & Mex. Bound. Bot. 118. 1859. Mimulus parvulus Wooton & Standley, sp. nov. Slender annual with almost filiform, prostrate or ascending branches not more than 4 or 5 cm. long, often rooting at the nodes; stems mostly glabrous; leaves thin, 4 to 8 mm. long and fully as wide or wider, broadly rounded-ovate, obtuse, subcordate at the base, short-villous with white hairs, entire or ob- scurely denticulate, on slender petioles half as long as the blades or longer; pedicels axillary, very slender, exceeding the leaves, glabrous or sparingly villous; calyx turbinate or narrowly campanulate, sharply angled, 5 mm. long in fruit, in flower not much shorter, purplish, oblique at the mouth, sparingly villous with coarse, white hairs; corolla bright yellow, 8 or 9 mm. long, the slender tube more than twice as long as the calyx; capsules 2 mm. long, nar- rowly oblong, abruptly acute. Type in the U. 8. National Herbarium, no. 660448, collected in Rocky Canyon, Grant County, August 9, 1911, by J. M. Holzinger. A most distinct species, very unlike any of our other southwestern ones and apparently very different from any of those from farther west or south. Pentstemon cardinalis Wooton & Standley, sp. nov. Perennial, forming dense clumps 40 cm. in diameter or more; stems rather slender, erect, simple, glabrous, green; basal leaves elliptic-spatulate, obtuse, long-petiolate ; cauline leaves various, the lower oblong-lanceolate, obtuse, 13 cm. long or less, narrowed at the base, the upper ones oblong to triangular-ovate, acutish or acuminate, sessile by a truncate or clasping base, thin, glabrous; bracts triangular-subulate, very small; inflorescence thyrsiform, secund, loosely many-flowered, glabrous; pedicels slender, erect, conspicuous; sepals 4 mm. long, lance-ovate, acute, scarious-margined; corolla 22 to 28 mm. long, eardinal red, considerably dilated in the throat, contracted at the mouth, the upper 172 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, lobes erect, the lower deflexed, nearly orbicular, 2 or 2.5 mm. long, strongly bearded in the throat with yellow hairs; anthers echinate along the commissure. Type in the U. 8. National Herbarium, no. 563916, collected on White Moun- tain Peak just above the forks of Ruidoso Creek, J uly 6, 1895, by E. O. Wooton. Altitude about 2,400 meters. This is somewhat, but not very closely, related to the group of P. puniceus and P. wrightii. The plants are not at all glaucous, however, and the form of the corolla is altogether different, especially in the contracted mouth. Pentstemon crassulus Wooton & Standley, sp. nov. Perennial from slender, fleshy roots and a stout caudex; stems stout, 45 em. high or less, simple, erect, glabrous, purplish above; basal leaves narrowly oblanceolate, short-petiolate, acute or obtuse, 10 em. long or less; cauline leaves all sessile, narrowly oblong to triangular-lanceolate, acute or abruptly acumi- nate, glabrous, thick and rather fleshy; inflorescence secund, loosely few- flowered, glabrous; pedicels slender, erect; sepals 6 mm. long, broadly ovate, obtuse or abruptly short-acuminate, scarious-margined; corolla about 25 mm. long, red, much dilated in the throat and sparingly bearded, contracted at the mouth, the upper lobes erect, the lower deflexed, nearly orbicular, short, about 2 mm. long. Type in the U. 8. National Herbarium, no. 563082, collected in the Lincoln National Forest in 1903 by Fred G. Plummer. Similar in general appearance to P, cardinalis, but the calyx lobes are shorter and broader and obtuse, and the leaves thick and fleshy and of different outline. Pentstemon neomexicanus Wooton & Standley, sp. nov. Perennial from a slender, creeping rootstock; stems stout, erect, simple, 50 to 70 em. high, glabrous; basal leaves linear-oblanceolate, acute, petiolate, 8 cm. long or less; cauline leaves oblong to linear, acutish or obtuse, 6 to 10 em. long, thick and fleshy, glabrous, rather numerous, scarcely reduced above: leaves of the inflorescence reduced, the lowest sometimes longer than the flowers ;-inflorescence much elongated, thyrsiform, secund, many-flowered, gla- brous; pedicels stout, short; calyx lobes obovate to oblong, 4 to 5 mm. long, truncate or obtuse, abruptly short-mucronate, erose-denticulate at the apex, Scarious-margined ; corolla 22 to 25 mm. long, deep bright blue, with a dilated throat and spreading limb, the lobes rounded, strongly hairy in the throat; capsules ovoid-conic, 5 or 6 mm. high. Type in the U. S. National Herbarium, no. 561371, collected in pine woods near Gilmores Ranch on Eagle Creek in the White Mountains, altitude 2,220 meters, August 15, 1907, by E. O. Wooton and Paul C. Standley (no, 3507). ADDITIONAL SPECIMENS EXAMINED: Capitan Mountains, 1900, Farle 200; James Canyon, August 5, 1899, Wooton; Cloudcroft, alt. 2,550 meters, 1909, Fisher 23; Capitan Mountains, 1903, Plummer; White Mountains, alt. 1,890 meters, 1897, Wooton 238; White Mountain Peak, August 1, 1901, Wooton; Gilmores Ranch, July 27, 1901, Wooton; Wingfields Ranch, July, 1895, Wooton, Mescalero Reservation, July 21, 1905, Wooton; Ruidoso Creek, alt. 1,980 meters, July 3, 1895, Wooton; Cloudcroft, June 30, 1899, Wooton. A very common and handsome plant in the higher parts of the Sacramento, White, and Capitan Mountains. It is most nearly related to P. unilateralis Rydb., but has very different calyx lobes and a hairy instead of glabrous throat. That species is found in New Mexico only in the higher mountains near the Colorado border. Pentstemon oliganthus Wooton & Standley, sp. nov. Stems slender, erect, simple, 20 to 30 cm. high, glabrous below, glandular xbove; basal leaves oblong or oval, petiolate, obtuse, 2 cm. long or less; cauline WOOTON AND STANDLEY—NEW PLANTS FROM NEW MExICcO. 173 leaves few and remote, lanceolate to narrowly oblong, acute, erect, thick, glabrous beneath, minutely puberulent above; inflorescence loosely few-flowered, its branches glandular and slightly villous; pedicels mostly slender, sometimes 1 em. long; calyx 4 mm. high, the lobes elliptic-oblong, acute, glandular-villous ; corolla 2U to 25 mm. long, the tube slightly widened upward, the spreading lobes oblong, obtuse, bearded in the throat; sterile stamen strongly bearded with yellow hairs; capsules conic-ovoid, 6 or 7 mm. high, acute. Type in the U. 8. National Herbarium, no. 259061, collected in the moun- tains west of Grants Station, August 1, 1892, by E. O. Wooton. This belongs to the group of P. confertus and P. procerus, but differs decidedly from those species in its larger flowers and loose, few-flowered inflorescence. Pentstemon spinulosus Wooton & Standley, sp. nov. Stems slender, ascending, 20 to 35 cm. high, purplish, minutely puberulent; leaves linear-oblanceolate to linear-lanceolate, numerous, obtuse or acufe, slightly reduced upward, glabrous, narrowed at the base or sessile, 5 cm. long or less; bracts linear-lanceolate, 1 to 2 cm, long; inflorescence few-flowered ; pedicels short, stout; sepals 7 mm, high, the lobes lanceolate, rather abruptly acuminate, not scarious, glabrous, the tips spreading; corolla 8 cm. long, di- lated in the throat, not bearded, the spreading limb 2 cm. wide; stamens included; anthers sagittate, dehiscent for half their length, finely spinulose along the sutures. Type in the U. S. National Herbarium, no. 156865, collected in the Magdalena Mountains in June, 1881, by G. R. Vasey. This is more closely related to P. bridgesii than to any of the southwestern species, but may be separated by the glabrous instead of glandular inflo- rescence and the much dilated corolla tube. Whether the -corollas are red 48 in that species can not be told from the faded dried specimens. Scrophularia laevis Wooton & Standley, sp. nov. Tall perennial, 1 meter high or more; stems slender, bright green, glabrous, erect, simple or with a few weak, spreading branches; petioles long, slender, usually half as long as the blades, these ovate or broadly lanceolate, 4 to 7 cm. long, acute, neither attenuate nor acuminate, bright green, thin, glabrous, scarcely if at all paler beneath, few, truncate or rounded and usually somewhat decurrent at the base, coarsely laciniate-dentate, the teeth triangular, acute or attenuate; inflorescence rather sparse and short, consisting of 5 or fewer pairs of few-flowered corymbs on spreading penduncles; pedicels stout, ascending, 1 to 2 em. long, glabrous or nearly so; flowers not seen; calyx lobes triangular- lanceolate, very acute, 3 to 4 mm. high; capsules narrowly ovoid, attenuate, about 8 mm. high, terminated by the persistent filiform style 4 mm. long. Type in the U. 8. National Herbarium, no. 561409, collected on a moist, shaded slope high up on Organ Peak above Filmore Canyon, altitude about 2,400 meters, September 23, 1906, by E. O. Wooton and Paul C. Standley. ADDITIONAL SPECIMENS EXAMINED: Old Tiptop, Organ Mountains, October 18, 1903, Metcalfe. On account of its slender habit, green stems, and pale leaves, and of its long petioles, this plant appears very different from the other western species. It is also distinguished from our others by its broad, short leaves and very acute calyx lobes, as well as by its almost complete lack of indument. Scrophularia parviflora Wooton & Standley, sp. nov. Perennial, about a meter high; stems rather slender, dull green or purplish, finely and densely puberulent throughout, simple or sparingly branched ; petioles short, less than one-third as long as the blades, these mostly triangular-lance 174 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. olate, 5 to 10 cm. long, truncate to cuneate at the base and unequal, often de- current, attenuate, coarsely laciniate-dentate, the teeth mostly triangular and not very acute, dull green, conspicuously veined, puberulent on both surfaces; in- florescence short, of about 6 pairs of corymbs or fewer, these on stout, spreading peduncles, finely glandular-puberulent; pedicels slender, 15 mm. long or less; calyx lobes short, triangular-ovate, acute or acutish; corolla 6 mm. long, dull purplish; mature capsules not seen. Type in the U. 8. National Herbarium, no. 495413, collected in the Mogollon Mountains on the West Fork of the Rio Gila, Socorro County, altitude about 2,250 meters, August 2, 1903, by O. B. Metcalfe (no. 345). ADDITIONAL SPECIMENS EXAMINED: Graham, July 21, 1900, Wooton. Related to 8. californica Cham. & Schlecht., but with sparser inflorescence, smaller flowers, thicker and more strongly veined leaves not cordate at the base, and different pubescence. We have seen two specimens of S. parviflora from Arizona, the one collected by G. C. Nealley in 1891 (no. 90), no locality given, and the other from the canyon of the Blue River near Coopers Ranch, Graham County, collected by Walter Hough, July 5, 1905. Veronica micromera Wooton & Standley, sp. nov. Slender, stoloniferous perennial; stems slender, 10 to 20 em. long, ascending, freely rooting at the lower nodes, succulent, glabrous; leaves small, 1 to 2 em. long, oval or obovate, the upper scarcely reduced, mostly shorter than the inter- nodes, obtuse, thin, entire or obscurely and remotely seriulate, all the upper Sessile, some of the lower contracted into petioles 1 mm. long, or all sessile; racemes axillary, slender, 3 to 7 cm. long, glabrous; pedicels very slender, ascending, or divergent and curved upward at the tip, subtended by very small, linear bracts; sepals 3 mm. long, narrowly lanceolate or elliptic-lanceolate, glabrous, very acute, in fruit evidently exceeding the capsule: corolla nearly white, bluish, scarcely exceeding the sepals; capsules small, 8 mm. long, glab- rous, broadly oval, nearly orbicular, scarcely as broad as long, rounded at the apex but not broadly so, Type in the U. 8S. National Herbarium, no. 686250, collected along ditches about Shiprock, on the Navajo Reservation, July 25, 1911, by Paul C. Standley (no. 7283). Altitude 1,425 meters. The plant is similar to V. americana, but is much smaller and more slender in all its parts. The leaves are almost all sessile instead of petiolate, and the Sepals are longer and narrower, . BIGNONIACEAE. Stenolobium incisum Rose & Standley, sp. nov. A low shrub 1 meter high or less, the stems simple or very sparingly branched ; leaves 17 cm. long or less, with 5 to 11 leaflets, usually with 9; leaflets linear- lanceolate, mostly about 6 cm. Jong and a little less than 1 em. wide, acuminate, _ attenuate to the base, the uppermost sessile, the lower conspicuously petiolulate, all sharply incised-serrate with deep, salient teeth, glabrous, or sometimes sparingly puberulent beneath; flowers about 4 em. long, on peduncles about 5 mm. long, in simple racemes, each peduncle subtended by a linear bract; calyx with 5 acute, cuspidate teeth; fruit 12 to 15 cm. long, smooth, or with numerous light colored lenticels. Type in the U. 8. National Herbarium, no 46776, collected on hills near Chihuahua, Mexico, November 15, 1886, by C. G. Pringle (no. 960). Also col- lected in the same locality by the same collector, October 2, 1885 (no. 360). ADDITIONAL SPECIMENS EXAMINED: Santa Eulalia Mountains, Chihuahua, 1885, Wilkinson; near Concepcion del Oro, Zacatecas, 1902, Palmer 389; Durango, WOOTON AND STANDLEY—NEW PLANTS FROM NEW mexico. 175 1896, Palmer 131, 507; Saltillo, 1898, Palmer 193; near San Juan Capistrano, Zacatecas, 1897, Rose 2495. The plant occurs as far north as western Texas and southern Arizona. In New Mexico it is known only from the Dona Ana Mountains, where it was col- lected October 28, 1896, by Wooton & Standley. This has always passed as S. stans (Tecoma stans L.), but is distinguished by its low stature, narrow leaflets with sharp and salient teeth, and the narrower and longer bractlets. The leaflets, too, are usually more numerous than in g. stans, The latter is a shrub often 3 meters high or more, or even a low tree with well-defined trunk. The proposed species is hever more than a very small shrub, often not more than 60 cm. high. It grows in the driest places in the southwestern mountains, on exposed slopes among rocks. RUBIACEAE. Houstonia rigidiuscula (A. Gray) Wooton & Standley. Houstonia angustifolia rigidiuscula A. Gray, Syn. Fl. 17: 27. 1884. Readily. distinguished from H. angustifolia by the lower, stouter, less erect stems, the thick, rigid leaves, and the few, more closely glomerate flowers. A common plant of western Texas and eastern New Mexico, growing on the plains and low hills of the Upper Sonoran Zone. CAPRIFOLIACEAE. Sambucus vestita Wooton & Standley, sp. nov. Shrub 3 meters high or less, with numerous stout stems from a single root; young branches minutely and densely velvety-pubescent ; leaflets lanceolate or narrowly so, 8 to 15 cm. long, long-attenuate, very unequal at the base and usu- ally rounded, puberulent beneath, puberulent above along the veins, thin, rather pale green, coarsely serrate, the teeth not incurved, acute, or acutish; petioles and petiolules densely and finely pubescent; cyme broad (10 to 20 cm.), flat- topped, with numerous open, slender, pubescent branches; flowers small, 3 to 4 wm. in diameter; fruit abundant, 5 mm, in diameter, black, glaucous, Type in the U. 8. National Herbarium, no. 560944, collected by Paul Cc. Stand- ley in Ice Canyon above Van Pattens Camp in the Organ Mountains, June 11, 1906., ADDITIONAL SPECIMENS EXAMINED: West Fork of the Gila, alt. 2,250 meters, 1903, Metcalfe 344; 4 miles west of Kingston, 1909, Goldman 1822; Eagle Peak, August 2, 1900, Wooton; San Mateo Peak, alt. 3,000 meters, 1909, Goldman 1740; Black Range, alt. 2,550 meters, 1904, Metcalfe 1184: Organ Mountains, 1908, Bailey 1469, May 15, 1892, Wooton, The plant is common in the canyons of the southwestern mountains. It is related to S. ncomexicana, but has smaller flowers and pubescent instead of glabrous branches. In habit the two are dissimilar, for S. neomezicana has usually a well-developed trunk with branches, while S. vestita consists of a clump of mostly simple shoots. CICHORIACEAE. Crepis chamaephylla Wooton & Standley, sp. nov. Perennial from a thick, fleshy root; stems glabrous, glaucous, 80 em. high, erect or ascending, very slender, the branches ascending, nearly naked, bearing only a few small, linear, bract-like leaves; basal leaves oblanceolate, entire, acute, 9 cm. long or less, glabrous, glaucous especially beneath, thick and succu- 176 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. lent, all lying flat upon the ground, sessile, or on short, broadly winged petioles; heads few; involucre 8 mm. high, glabrous, the bracts linear-lanceolate, pale yellowish green, the few outermost much reduced ; corollas deep yellow; achenes dark brown, glabrous, scarcely discoid at the summit, with copious pappus of bright white bristles 4 mm. long. Type in the U. S. National Herbarium, no. 686386, collected at the north end of the Carrizo Mountains, July 30, 1911, by Paul C. Standley (no. 7419). The plants were common in the wet ground about a small hillside spring flowing out from among juniper and pinyon trees. Upper Sonoran Zone. The proposed species is a very distinct one by its entire, glaucous leaves which are reflexed upon the ground instead of erect as in most of our species. The heads are comparatively small and the bracts are a clear, pale yellowish green rather than black or brownish as we find them in most species. The aspect of the plant is strikingly like that of two of the species of Cynthia. Crepis mogollonica Greene, sp. nov. in herb. Basal leaves numerous, some of them linear and entire, others narrowly linear-oblong or linear-oblanceolate and runcinate-pinnatifid into linear lobes, glabrous, bright green, thin, 18 to 24 em. long; stems slender, 40 to 60 cm. high, glabrous, with few, ascending branches, bearing linear, bract-like leaves at the nodes; heads few, long-pedunculate; involucre 10 to 12 mm. high, of linear- oblong, glabrous bracts; flowers bright yellow; achenes slender, brown, 5 mm. long, shorter than the slightly sordid pappus. Type in the U. S. National Herbarium, no. 495570, collected in the Mogollon Mountains, Socorro County, on the West Fork of the Gila, altitude 2,400 meters, August 23, 1903, by O. B. Metcalfe (no. 576). This is related to ©. glauca, or is of that group at least. It is distinguished from the related species by its large heads and the very long, extremely narrow leaves, Crepis neomexicana Wooton & Standley, sp. nov. Basal leaves in a dense cluster, about 9 cm. long, nearly sessile by winged bases, oblanceolate or oblong, deeply runcinate-pinnatifid with broadly trian- gular lobes, rather obtuse at the apex, thick and firm, bright green, glabrous; stems about 30 cm. high, stout, glabrous, sparingly branched, with reduced, linear, bract-like leaves at the nodes; peduncles sparingly tomentose; invol- ucres 8 mm. high, the bracts linear, slightly attenuate upwards, glandular; flowers bright yellow; mature achenes not seen, the young ones very short, brown. Type in the U. 8. National Herbarium, no. 563748, collected on Tularosa Creek, Socorro County, July 14, 1906, by E. O. Wooton. Similar to C. perplerans Rydb., but with smaller heads and very thick, broad, deeply lobed leaves. Ptiloria bigelovii (A. Gray) Wooton & Standley. Hemiptilium bigelovii A. Gray, U. S. & Mex. Bound. Bot. 105. 1859, CARDUACEAE. Coelestina sclerophylla Wooton & Standley, sp. nov. Slender, rigid perennial, about 35 cm. high, with numerous puberulent, nearly simple stems from a suffrutescent base; leaves opposite, numerous, lanceolate to ovate or somewhat deltoid, 25 to 40 mm. long, thick and rigid, acute or attenuate, attenuate or acute at the base, coarsely crenate, puberulent on both WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXIco. 177 surfaces, glandular-dotted beneath, all on slender petioles 7 mm. long or less; inflorescence of few, clustered heads terminating long, naked branches; heads campanulate, 7 mm. high, on short, glandular or puberulent peduncles; bracts linear, appressed, attenuate, puberulent, striate: flowers but little exceeding the bracts; achenes 3 mm. Jong, 5-angled, glabrous; pappus a short, obtusely toothed, glabrous crown. Type in the U. S. National Herbarium, no, 232780, collected in Guadalupe Canyon, Sonora, August 27, 1893, by E. C. Merton (no. 2031). Also on Cajon Creek in Chihuahua along the New Mexico line, August 16, 1892, Mearns 700. Guadalupe Canyon extends into New Mexico, and doubtless the plant occurs on the north as well as on the south side of the boundary. Doctor Gray reported this plant from this same region as Ageratum corym- bosum Zuce., but the northern plant is very unlike true corymbosa, which occurs much farther south. It differs especially in the form of the leaves and in the pubescence and inflorescence. Kuhnia chlorolepis Wooton & Standley, sp. nov. Perennial with numerous stems from a thick, woody root; stems slender, erect, 30 to 50 cm. high, simple up to the inflorescence, there abundantly branched, densely puberulent; upper leaves linear, the lower lanceolate or Jance-linear, all obtuse, densely puberulent, sessile, 3-nerved, glandular-dotted on the lower surface; heads numerous, on long, slender peduncles, large, 15 mm. high; bracts firm, green tinged with purple, linear-oblong, broad, the outer short and acute, the inner obtuse or acuminate, finely pubescent, conspicu- ously striate, in several series, the outer ones gradually and successively shorter; corolla lobes oblong-lanceolate, rather long; achenes pubescent, 6 mm. long, finely striate; pappus 6 mm. long, whitish, copper colored at the base. Type in the U.S. National Herbarium, no. 560399, collected at Mangas Springs, June 2, 1903, by O. B. Metcalfe (no. 104). ADDITIONAL SPECIMENS EXAMINED: Gila River bottom near Cliff, alt. 1,350 meters, 1903, Metcalfe 152; 1851-2, Wright 1394 and 1182 (in part); Alamo Viejo, 1892, Mearns 188; Mexican Boundary Survey 458 (in part). This is most nearly related, perhaps, to A. gooddingii A. Nels., but that has° smaller hends and densely glandular bracts. In that species, too, the bracts are much thinner and not green. Coleosanthus chenopodinus Greene, sp. nov. in herb. Small, much-branched shrub; stems stout, with exfoliating bark, puberulent below, glandular above, densely branched ; leaves small, ovate or lanceolate, 35 mm. long or less, rather thick and succulent, glabrous or nearly so, acute, rounded or cuneate at the base, somewhat serrate; heads paniculate, large, about 12 mm. long, on slender, leafy, densely viscid peduncles 2 to 4 cm. long; outer bracts often foliaceous, lanceolate, the others lanceolate to oblong-linear, conspicuously nerved, glandular-viscid, acute, or the inner obtuse; achenes faintly strigose. Type in the U. S. National Herbarium, no. 495728, collected in the Gila River bottom near Cliff, Grant County, September 22, 1903, by O. B. Metealfe (no. 776). Altitude 1,850 meters. A peculiar species, somewhat related to C. floribundus, but with very long peduncles, larger heads, and peculiarly succulent leaves. Coleosanthus venosus Wooton & Standley, sp. nov. Low perennial, 50 cm. high or less, with sev eral stems from a woody root; stems slender, simple below, sparingly branched above, cinereous-puberulent ; leaves narrowly oblong to linear, obtuse, sessile, entire or obscurely serrate, 178 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. thick, conspicuously veined, canescent, 55 mm. long or less; heads few, racemose or narrowly paniculate, 10 to 12-flowered, 10 to 12 mm. high; peduncles slender, bearing 1 to 3 heads, 10 to 35 mm. long; bracts much im- bricated in several series, the outer orbicular to broadly oblong or ovate, obtuse or emarginate and mucronate, cinereous, the inner lanceolate, acute, ciliolate, all prominently striate; achenes equally striate, strigose, brown, with firm, white, scaberulous pappus. Type in the U. S. National Herbarium, no. 495629, collected at Mangas Springs, September 5, 1903, by O. B. Metcalfe (no, 653). Altitude 1450 meters, ADDITIONAL SPECIMENS EXAMINED: 1851, Wright 1185; Burro Mountains, Sep- tember, 1880, Rusby; San Luis Mountains, 1893, Mearns 2211 and 2284. The plant also occurs in southern Arizona and northern Chihuahua. In the Synoptical Flora it was listed as Brickellia oliganthes (Less.) A. Gray, but that name was applied originally to a plant from much farther south in Mexico. C. venosus differs in having much narrower, sessile, more pubescent leaves, as well as different inflorescence, bracts, and pubescence, Doctor Gray at various times remarked that the plant from New Mexico and Arizona was different from that of Mexico, but probably he had not sufficient Mexican material to warrant separation of the two. Grindelia neomexicana Wooton & Standley, sp. nov. Erect biennial or perennial, 50 cm. high or less; stems slender, sparingly branched from the base but abundantly branched above, the branches erect, leafy throughout, glabrous; cauline leaves narrowly oblong, or oblong-lanceolate, 45 mm. long and 10 mm. wide or less, sessile, obtuse or acute, glabrous, sharply and evenly serrate, rarely entire; heads few, large, 15 mm. wide and 12 mm. high or smaller; rays numerous, showy, stiff, 15 mm. long, narrowly spatulate, obtuse; bracts many, elongated-linear, the outermost thick and green through- out, with lax, slightly spreading tips, the innermost wide, scarcely if at all viscid; heads subtended by 1 or 2 linear or Nnear-lanceolate, bract-like leaves; achenes light brown, faintly striate, the pappus smooth or sparingly and very faintly barbellate. Type in the U. 8. National Herbarium, no. 561099, collected in the mountains horth of Santa Rita, August 23, 1900, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Mountains southeast of Patterson, August 16, 1900, Wooton; G O S Ranch, 1911, Holzinger. We have seen no other New Mexican plant with involucral bracts like those of this species. They are unusually numerous, narrow, green, and only slightly spreading, none of them being recurved. Grindelia pinnatifida Wooton & Standley, sp. nov. Perennial or biennial, 30 to 45 em. high, with numerous stout stems from each root, these simple below but with numerous erect, slender, corymbose branches above; stems glabrous, striate, reddish above; lower leaves long-petio- late, laciniate-pinnatifid, the segments irregularly dentate, oblong, acutish ; upper leaves linear-oblong, laciniate-serrate, sessile by clasping bases, all glabrous or nearly so, conspicuously glandular-punctate; heads solitary at the ends of the branches, depressed-hemispheric, 15 to 17 mm. in diameter; bracts linear, with flat, green, slightly reflexed tips, densely viscid; immature achenes glabrous; pappus bristles smooth. Type in the U. S. National Herbarium, no. 685628, collected on open slopes about Chama, altitude about 2,400 meters, J uly 9, 1911, by Paul C. Standley (no. 6606). The plant was very abundant about Chama, in the Transition Zone. It differs from all our other species in its evidently pinnatifid lower leaves. The WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 179 truly basal ones were not secured, but doubtless they are even more deeply divided than the lower cauline ones. Grindelia setulifera Wooton & Standley, sp. nov. Stems slender, with numerous erect branches, leafy throughout, glabrous, reddish; cauline leaves oblanceolate, 25 mm. long or less, obtuse, sessile, with numerous setose teeth, the teeth toward the apex usually broader; heads numerous, mostly 10 to 12 mm. broad, subtended by 1 or 2 oblanceolate, bract- like leaves; outer bracts lanceolate, acute, thick and green at the tips, the inner mostly oblong, acute or abruptly acuminate, thin, straw colored, scarcely at all viscid; all the bracts erect and appressed; rays numerous, narrowly spatulate, obtuse, about 10 mm. long; achenes brown, glabrous, about 3 mm. long. Type in the U. S. National Herbarium, no. 45772, collected on high summits of the Mogollon Mountains, September, 1881, by Dr. H. H. Rusby (no. 206). This was originally determined as G. arizonica, but it differs from that species in its much smaller heads and very different leaves. The bracts, too, are not alike in the two species. Chrysopsis cryptocephala Wooton & Standley, sp. nov. Perennial with several stems from a slender, woody root; stems very slender, 30 to 40 cm. high, erect, glandular-puberulent, sometimes sparingly hirtellous above, simple, with rather few leaves; leaves sessile, oblong to lanceolate or ovate, obtuse, often abruptly acuminate and mucronate, 25 mm. long or less, green, thin, minutely glandular, scaberulous, especially on the upper surface; heads mostly solitary at the ends of the simple stems, sometimes corymbose, about 1 cm. broad, almost hidden by the numerous, thin, broad, often ciliate, ovate, acute, bract-like leaves; rays numerous, pale yellow, 8 or 9 mm. long. Type in the U. S. National Herbarium, no. 563739, collected by E. O. Wooton in section 23 of the V Pasture in the White Mountains, July 23, 1905. ADDITIONAL SPECIMENS EXAMINED: Gavilan Creek, August 19, 1897, Wooton 512. This formed a part of Doctor Greene’s C. fulcrata, as originally described, but it is very different from the type of that species. It is most like C. rest- nolens A. Nels., but has broader, fewer heads, different leaves, and very different bracts. Chrysopsis nitidula Wooton & Standley, sp. nov. Perennial with several stems from a slender root; stems slender, erect, 20 to 85 cm. high, finely and rather sparingly sericeous, leafy; leaves oblanceolate, or the uppermost lanceolate, obtuse or acutish, sessile, the lower tapering to the base, finely sericeous, the leaf as a whole appearing green and remarkably soft and smooth; heads 1 to several, on slender, erect peduncles, subtended by a few elliptic to oblanceolate, thin, bract-like leaves; disk about 1 cm. broad, the bracts linear-lanceolate, acute, sericeous; rays numerous, bright yellow, 10 to 12 mm. long; achenes compressed, sericeous. Type in the U. S. National Herbarium, no. 495550, collected by O. B. Met- calfe in the Mogollon Mountains on the West Fork of the Gila, at an altitude of about 2,250 meters, August 20, 1903 (no. 552). ADDITIONAL SPECIMENS EXAMINED: North of Ramah, July 25, 1906, Wooton; Middle Fork of the Gila, August 5, 1900, Wooton. This is very unlike any of our other species, being strongly marked by its peculiar pubescence and long rays. Chrysopsis senilis Wooton & Standley, sp. nov. Stout perennial with several stems from each root; stems simple below, corym- bosely branched above, the branches ascending, densely villous throughout, very leafy, the leaves mostly longer than the internodes; leaves sessile, oblong or 180 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. oblong-lanceolate, obtuse, often abruptly short-acuiminate, thick, canescent; heads crowded, 3 to 8 at the end of each branch, about 1 cm. broad, subtended by numerous ovate to elliptic, thin, bract-like leaves, these long-ciliate; in- volucral bracts thin and membranous, linear or linear-lanceolate, in several series, acute, canescent; the very short peduncles densely white-villous; rays numerous, dull yellow, about 9 mm. long; achenes compressed, densely villous. Type in the U. 8S. National Herbarium, no. 330718, collected in the Organ Mountains at an altitude of 1,440 meters, September 1, 1897, by E. O. Wooton (no. 509). ADDITIONAL SPECIMENS EXAMINED: Pena Blanca, October 21, 1906, Wooton & Standley; San Augustine Ranch, September 1, 1897, Wooton; 1851, Wright; Organ Mountains, September 4, 1898, Cockerell; Van Pattens, September 10, 1899, Wooton. This is related to C. fulerata Greene, which grows in the same range of mountains, but it has abundant, long, white pubescence, while in that species the pubescence is short and not conspicuously white. C, senilis, too, has mostly grayish, narrower leaves and there are several heads clustered at the ends of each branch, while in C. fulcrata the heads are usually solitary. Sideranthus laevis Wooton & Standley, sp. nov. Perennial from a thick, woody root; stems slender, much branched, erect, the branches ascending, bright green, glabrous; leaves linear or linear-oblong, bright green, glabrous, entire or sparingly serrate, the teeth low and incon- spicuous; heads few, solitary, 8 or 9 mm. broad, on slender peduncles; bracts linear or linear-lanceolate, acute, glabrous or sparingly puberulent ; rays pale yellow, 4 mm. long; achenes small, densely sericeous, the pappus evidently barbellate, Type in the U. 8. National Herbarium, no. 564582, collected on gypsum hills near Lakewood, August 6, 1909, by E. O. Wooton. From its lack of pubescence this might be confused with S. glaberrimus Rydb., but in appearance the two are really very unlike. Our plant is much more slender, is much branched, and has fewer heads and narrow, shallowly toothed leaves, and the whole plant is of a bright green, while the stems and foliage of S. glaberrimus are dull and somewhat glaucous. Sideranthus viscidus Wooton & Standley, sp. nov. Stems stout, branched, ascending or spreading, densely glandular-puberulent, 50 cm. high or less; cauline leaves thick, humerous, oblong, obtuse, sessile, coarsely serrate, the teeth not spinulose, densely viscid; heads few, on slender peduncles, cumpanulate, about 12 mm. broad; bracts linear, with evident green tips, glandular-puberulent; rays few, short, not more than 5 mm. long; achenes nearly 3 mm. long, densely sericeous, with tawny pappus 5 or 6 mm. long. Type in the U. S. National Herbarium, no. 690240, collected near Hope, August 3, 1905, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Dayton, October 3, 1907, Z. BS. Wigsdale. Very different from our other species in the form of its leaves and in its densely viscid pubescence. Isocoma oxylepis Wooton & Standley, sp. nov. Perennial, about 30 em. high, with numerous stems from a thick, woody root; stems slender, glabrous, densely leafy, erect, ending in a corymblose inflorescence; leaves linear, acute, glabrous or scaberulous, 4 cm. long or less, weak and spreading or reflexed; heads very numerous, all on slender peduncles 1 em. long lor shorter; involucre narrowly campanulate, about 4 mm. high; bracts glabrous or nearly so, lanceolate or oblong-lanceolate, acute; throat of WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 181 the corolla but slightly inflated, the lobes lanceolate, acute; achenes densely sericeous. Type in the U. S. National Herbarium, no. 233968, collected near White Water, Chihuahua, September 11, 1898, by Dr. E. A. Mearns (no. 2288). ADDITIONAL SPECIMENS EXAMINED: Dog Spring, New Mexico, September 22, 18938, Mearns 2407. This is a very striking species, distinguished by its numerous, very narrow, long, and weak leaves, the many pedunculate heads, and the acute bracts. Isocoma wrightii (A. Gray) Wooton & Standley. Linosyris wrightit A. Gray, Pl. Wright. 1: 95. 1852. Bigelovia wrightit A. Gray, Proc. Amer. Acad. 8: 639. 1873. This has been confused with J. heterophylla, but is amply distinct in its narrow leaves and slenderly pedunculate heads. Both species occur in New Mexico. Chrysothamnus baileyi Wooton & Standley, sp. nov. Low, densely branched shrub; older branches covered with rough, dark gray bark, the younger ones straw colored, slender, puberulent, angled, densely leafy; leaves erect or appressed, linear or linear-oblong, 15 mm. long or less, abruptly acuminate, sharp-pointed, thick, dull pale green, minutely ciliolate with short, stout hairs, otherwise glabrous, sessile; heads numerous, clustered at the ends of the branches, turbinate to cylindric, not strongly angled; in- volucres 10 mm. high or less, the numerous bracts ovate to oblong or lanceolate, thin, glabrous, abruptly acuminate, mostly bristle-pointed; achenes glabrous, striate; pappus yellowish, 8 to 10 mm. long. Type in the U. 8S. National Herbarium, no, 443565, collected at the north end of the Guadalupe Mountains, September 4, 1902, by Vernon Bailey (no. 498). ADDITIONAL SPECIMENS EXAMINED: White Mountains, alt. 1,620 meters, 1897, Wooton 508; Buchanan, August 12, 1909, Wooton. This is similar in general appearance to C. pulchellus, That species does not have ciliolate leaves and has much larger heads and longer pappus, Solidago arizonica (A. Gray) Wooton & Standley. Solidago canadensis arizonica A. Gray, Proc. Amer. Acad. 17: 197. 1882. Stems stout, usually simple, 1 meter high or often more, striate, finely and densely cinereous-puberulent; leaves very numerous, often crowded, elliptic to elliptic-lanceolate or oblanceolate, 7 to 18 cm. long, acute, attenuate at the base to a broad, margined petiole or sessile, sharply serrate or often only obscurely and remotely serrate, bright green, evidently triple-veined, copiously scaberulous on both surfaces; inflorescence a broad, pyramidal panicle, 25 cm. long and as broad or smaller, all the branches recurved-ascending, slender, densely puberu- lent, sparsely viscid; pedicels slender, erect or ascending, 5 to 7 mm. long; heads campanulate, about 5 mm. high, with very numerous flowers and short, narrow, inconspicuous rays; achenes short, loosely pubescent, less than half as long as the white, scabrous pappus. Although first described from Arizona, this plant is a common and showy species of southern New Mexico. It has always been associated with 8S. cana- densis, but seems remarkably distinct from that species, especially by its very large heads. Solidago howellii Wooton & Standley, sp. nov. Stems low and stout, 40 cm. high or less, purplish below, densely canescent ; basal leaves not seen; cauline leaves narrowly elliptic to oblanceolate or linear- oblanceolate, 25 to 75 mm. long, acute, or the lower obtuse, very thick and stiff, ‘tapering at the base, mostly entire, prominently 3-nerved, densely scabrous- 182 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. canescent on both surfaces; inflorescence pyramidal or narrow, 10 to 15 cm. long, all the branches recurved, densely canescent; leaves of the inflorescence elliptic to ovate-lanceolate or linear-lanceolate, acute; pedicels 3 to 5 mm. long; heads 5 to 7 mm. high; involucral bracts oblong, obtuse, straw colored, glabrous or nearly so; rays 2 to 2.5 mm. long, pale yellow; achenes conspicuously strigose. Type in the U. S. National Herbarium, no. 495104, collected on the Sierra Grande, August 15, 1908, by A. H. Howell (no, 219). ADDITIONAL SPECIMENS EXAMINED: Clayton, 1891, Carleton 393; Sierra Grande, 1903, Howell 286, 237; Trinchera Pass, September 7, 1903, Howell 190; Folsom, 1908, Howell 162; Capitan Mountains, 1903, Gaut 91; Sierra Grande, altitude 2,300 meters, 1911, Standley 6097; Nara Visa, 1911, Fisher 204. This is related to S. radula, but has larger heads and 3-nerved cauline leaves. The plant, too, is not bright green, but dull yellowish or grayish, because of the more abundant pubescence of the various parts. It is also related to S. mollis Bartl., but has very different leaves. The plant is abundant upon the lower slopes of the Sierra Grande, growing among the volcanic rocks. Solidago neomexicana (A. Gray) Wooton & Standley. Solidago multiradiata neomexicana A. Gray, Proc. Amer. Acad. 17: 191. 1882. Stems stout, erect, 50 to 60 cm. high, purplish, striate, nearly glabrous below, above viscid-puberulent ; basal leaves not seen, the cauline ones oblanceolate, 5 to 10 cm. long, acute, pale green, narrowed at the base into broadly winged petioles, glabrous, finely and inconspicuously reticulate-veined, never triple- veined; inflorescence rather loosely thyrsoid, often with loose axillary clusters, viscid-puberulent; peduncles short, stout, 2 to 4 mm. long; heads large, 7 mm. high, campanulate, with numerous disk and ray flowers; bracts unequally im- bricated, narrowly oblong, obscurely puberulent, acute; achenes about 2 mm. long, terete or nearly so, stout, densely appressed-pubescent, about half as long as the copious pappus. The type was collected by Rusby, in September, 1881, on high, rocky summits of the Mogollon Mountains of New Mexico. The plant seems not to have been found by any other collector. Solidago tenuissima Wooton & Standley, sp. nov. Stems slender, glabrous, pale, about 1 meter high or a little less, simple; basal leaves linear-oblanceolate, acute, entire or with a few remote, low teeth, 10 to 20 cm. long; cauline leaves linear-oblanceolate, the uppermost nearly linear, glabrous, bright green, conspicuously veined but never triple-nerved, acute, long-attenuate to the base, 5 to 15 cm. long, usually entire or the lower remotely serrate, the leaves rather numerous on the stems, spreading or in- clined to be ascending; inflorescence a narrow panicle with few ascending branches, often a few pedunculate racemes in the axils of the leaves, the whole very narrow and slender, 10 cm. long or less and less than 5 em. wide; branches of the inflorescence and pedicels puberulent, the latter 3 to 5 mm. long; bracts linear, 5 mm. long or less; heads rather numerous, 3.5 or 4 mm. high; involucral bracts oblong-lanceolate, obtuse, ciliolate at the apex, glabrous or sparingly puberulent; rays small, 1.5 mm. long. Type in the U. 8. National Herbarium, no. 591665, collected in Guadalupe Canyon near Cloverdale, July 15, 1892, by Dr. E. A. Mearns (no. 466). ADDITIONAL SPECIMENS EXAMINED: Head of Guadalupe Canyon near Clover- dale, July 15, 1892, Mearns 478; along streams, Mogollon Mountains, July, 1881, Rusby 227. Doctor Rusby’s specimens were distributed as S. missouriensis, and our species belongs to the group of which that species is a member. It is related -'WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 183 to 8. glaberrima, but is a taller, much more slender plant, with very narrow leaves and a different, narrow inflorescence. Petradoria graminea Wooton & Standley, sp. nov. Tufted perennial from a thick, woody caudex, low, 10 to 15 em. high; old leaves persistent at the base in a dense mass, grayish; leaves linear, rigid, sharp-pointed, 1-nerved, 3 to 5 cm. long, glabrous, dilated at the base, very lumerous; corymbs with few cylindric heads 6 to 7 mm. high; bracts acute to abruptly acuminate or obtuse, smooth, stramineous; flowers usually 5; achenes brown, glabrous. Type in the U. 8S. National Herbarium, no. 45778, collected in northwestern New Mexico, July 14, 1883, by C. C. Marsh (no. 209). ADDITIONAL SPECIMENS EXAMINED: South of Gallup, 1904, Wooton 2560; Gallup, 1897, Herrick 816; Tunitcha Mountains, 1911, Standley 7777. A species of different aspect from P. pumila because of the very numerous, short, grass-like leaves and smaller stature. In general appearance the plants suggest some of the narrow-leaved species of Arenaria. Leptilon integrifolium Wooton & Standley, sp. nov. Annual or possibly biennial, 20 to 40 em. high; stems stout, much branched at the base or above, glandular-villous and arachnoid; leaves numerous, oblong, obtuse, entire or rarely with a few teeth, sessile by a clasping base, villous; heads rather few, racemose, subtended by small lanceolate leaves, on slender peduncles 1 to 3 cm. long, 6 to 8 mm. broad; bracts green, lanceolate, acute, villous and arachnoid; rays very short, nearly obsolete, white; achenes small, glabrous, the fine, nearly white pappus 3 mm. long. Type in the U. S. National Herbarium, no. 495594, collected on the West Fork of the Gila in the Mogollon Mountains, altitude about 2,100 meters, August 28, 1903, by O. B. Metcalfe (no. 610). ADDITIONAL SPECIMENS EXAMINED: Mineral Creek, alt. 2,250 meters, 1904, Metcalfe 1419; White Mountains, alt. 2,100 meters, 1897, Wooton 356; East Las Vegas, September 24, 1907, S. Y. Parnay; Gilmores Ranch, alt. 2,220 meters, 1907, Wooton & Standley 3698. Similar to LZ. subdecurrens, but that has all its leaves prominently toothed, its heads short-pedunculate, its stem less branched, and its pubescence rather different. * Erigeron deminutus Wooton & Standley, sp. nov. Perennial with creeping rootstocks; stems stout, 30 cm. high or less, simple, canescent; basal leaves spatulate, oblanceolate, obtuse or acutish, 6 cm. long or less; cauline leaves oblanceolate to oblong-lanceolate, acutish, sessile, thick, canescent, the upper much reduced; peduncles 1 to 5, long, stout, canescent and sparingly glandular; disk 10 to 12 mm. wide; bracts in about 2 series, linear, acute, appressed, densely canescent; rays pale purplish, narrow, about 100, 7 to 8 mm. long; achenes strigose, brown, the pappus in 2 series. Type in the U. 8. National Herbarium, no. 690244, collected north of Ramah, July 25, 1906, by E. O. Wooton, ADDITIONAL SPECIMENS EXAMINED: Mountains west of Grants Station, alt. 1.800 meters, August 2, 1892, Wooton. A species of the glabellus group, readily distinguished by its thick, canescent leaves. Erigeron eastwoodiae Wooton & Standley, sp. nov. Low annual, 10 to 14 ecm. high, with very numerous spreading, slender branches arising from or near the base; stems much branched, smooth, sparingly cinereous, the tips of the hairs all directed upward; leaves linear or 60541°—13——-6 184 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. linear-oblanceolate, obtuse, 20 cm. long or less, the uppermost shorter and narrower, cinereous but not densely so; heads small, about 7 mm. wide; peduncles very slender, terminating the branches, mostly naked; involucral bracts in about 2 series, green, thin, nearly linear, slightly glandular, sparingly hirsute, abruptly acute; rays 25 to 30, pale purple, 3 mm. long; achenes slender, conspicuously compressed, nearly glabrous; pappus deciduous, uniseriate, of few weak, white bristles. Type in the U. 8. National Herbarium, no. 687233, collected on dry hills at the north end of the Carrizo Mountains, July 30, 1911, by Paul C. Standley (no. 7433). ADDITIONAL SPECIMENS EXAMINED: Shiprock, alt. 1,425 meters, 1911, Standley 7275. The plant was first collected by Miss Alice Eastwood along the San Juan River in southeastern Utah, July 12, 1895. Probably it is what is reported as E. bellidiastrum in Rydberg’s Flora of Colorado, said to come from McElmo Canyon. It is nearly related to that species, but is a very low, diffusely branched plant, very unlike H. bellidiastrum in habit. The rays are brighter colored and the leaves shorter and narrower. Erigeron gilensis Wooton & Standley, sp. nov. Biennial or possibly only annual, 35 cm. high, branched from the base; stems several, slender, erect, with numerous leaves, copiously strigose, the pubescence near the base mostly spreading; leaves oblanceolate to narrowly oblong or linear-lanceolate, the lower obtuse, the upper acute, sessile or attenuate to the base, sparingly strigose on both surfaces; heads few, on slender, erect peduncles, the disk S or 9 mm. wide; bracts linear, acute or attenuate, villous, slightly viscid; rays numerous, about 100, lavender, linear, 5 mm. long. Type in the U. S. National Herbarium, no. 690243, collected on the North Fork of the Rio Gila, August 4, 1900, by E. O. Wooton. This is related to Erigeron divergens, but the pubescence is mostly appressed instead of spreading, and the plant is greener, more slender, lower, and less branched. Erigeron rudis Wooton & Standley, sp. nov. Perennial from a stout, creeping, branched rootstock, 4 or 5 stems or more in each clump; stems erect, a meter high or fess, stout, striate, hirsute throughout, usually simple below; radical leaves oblanceolate to spatulate, acute or obtuse, entire, 12 cm. long and 14 mm. wide or smaller, 3-nerved, bright green, scabrate or glabrate, long-petiolate; cauline leaves lanceolate or oblong-lanceolate, 6 cm. long and 2 cm. wide or smaller, very numerous, the plants appearing densely leafy, the leaves little reduced above, acute, sessile by broad, somewhat clasping bases, scabrous on both surfaces, ciliate; peduncles stout, short, villous, glan- dular, ascending; heads 8 to 8, the disk about 15 mm. wide and 7 mm. high; rays numerous, very narrow, light purple, about 12 mm. long; bracts linear, appressed, glandular-puberulent, in about 2 series; achenes strigose, the pappus in 2 series. Type in the U. 8. National Herbarium, no. 330495, collected in the White Mountains, Lincoln County, August 1, 1897, by E. O. Wooton (no. 270). Altitude 2,100 meters. ADDITIONAL SPECIMENS EXAMINED: Burro Mountains, alt. 2,100 meters, 1906, Blumer 1881; Luna, July 28, 1900, Wooton; West Fork of the Gila, August 6, 1900, Wooton; Middle Fork of the Gila, August 5, 1900, Wooton; 1851, Wright 1165: Wheelers Ranch, July 11, 1906, Wooton; Graham, July 21, 1900, Wooton; Gilmores Ranch, July 14, 1895, Wooton; Capitan Mountains, 1900, Harle 397; Capitan Mountains, alt. 2,100 meters, 1900, Harle 196. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 185 One of the most abundant species in the mountains of southern New Mexico; a tall, coarse plant, usually growing along streams. It has passed as Hrigeron macranthus, but has different pubescence and is a stouter, more leafy plant. Erigeron semirasus Wooton & Standley, sp. nov. Perennial from a slender rootstock; stems usually slender, erect, densely leafy, 60 cm. high or less, mostly glabrous below, above puberulent and some- what glandular, reddish; basal leaves oblanceolate-spatulate; lower cauline leaves oblanceolate, the upper lanceolate or lance-ovate and but little reduced, acute, the uppermost sessile by clasping bases, scabrous, not ciliate, bright green, thick, entire; peduncles few, short, stout, ascending, nearly concealed by the leaves; disk about 12 mm. broad; bracts in 2 series, linear, acute, glandular- puberulent; rays purple, narrow, numerous, 10 mm. long; achenes strigose, the pappus in 2 series. Type in the U. 8. National Herbarium, no. 495395, collected in the Mogollon Mountains on Mogollon Creek, July 23, 1908, by O. B. Metcalfe (no. 320). Altitude 2,400 meters. ADDITIONAL SPECIMENS EXAMINED: Mountains near Las Vegas, July, 1881, Vasey; 1847, Fendler 376; Harveys Upper Ranch, alt. 2,880 meters, 1908, Standley 4606; Beulah, August, 1899, Cockerell; Beulah, 1899, Porter; Upper Pecos, 1904, Bartlett 1238. Erigeron senilis Wooton & Standley, sp. nov. Biennial or a short lived perennial from a slender root; stems branched from near the base, usually simple for about 2 cm. then branched, the branches very slender, 30 cm. long or less, prostrate or ascending, densely hirsute at the base, above strigose; basal leaves spatulate-obovate, obtuse, long-petiolate; cauline leaves rather distant, spatulate-obovate to oblanceolate, thin, bright green, considerably reduced toward the extremities of the branches, abruptly acuminate or acutish, strigose; heads 10 to 12 mm. in diameter, on long, slender, sparingly strigose peduncles; bracts nearly equal, linear, acute, hirsute; rays numerous, pale purplish, very narrow, 4 to 5 mm. long; achenes glabrous, stramineous, the pappus in a single series. Type in the U. 8. National Herbarium, no. 560777, collected in a canyon above Van Pattens Camp in the Organ Mountains, June 9, 1906, by Paul C. Standley. ADDITIONAL SPECIMENS EXAMINED: Organ Mountains, alt. 1,650 meters, August 29, 1894, Wooton. In habit this is much like HZ. flagellaris, but the stems have different pubes- cence and the leaves are much broader, less reduced toward the extremities, and more obtuse. The method of branching, too, is different from that of JZ. flagellaris. Erigeron bakeri Wooton & Standley, sp. nov. Perennial from a slender rootstock; stems stout, erect, 45 cm. nigh or less, simple up to the inflorescence, glabrous near the base, elsewhere finely and densely soft-pubescent; basal leaves oblanceolate, acute, 20 cm. long or less, long-petiolate; cauline leaves oblanceolate below to lanceolate or elliptic-oblong above, acute or acuminate, numerous, thin, sparingly appressed-pubescent, bright green; peduncles stout, 10 cm. long or less, soft-pubescent with fine, dark hairs, bearing a few small leaves; disk 12 to 14 mm. wide; bracts in about 2 series, linear, acute, appressed, canescent; rays purplish, narrow, numerous, 7 or 8 mm. long; achenes strigose, the pappus in 2 series. Type in the U. S. National Herbarium, no. 369203, collected near Chama, September 9, 1899, by C. F. Baker (no. 678). 186 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. ADDITIONAL SPECIMENS EXAMINED: Chama, 1899, Baker; Chama, alt. 2,400 meters, 1911, Standley 6744; Dulce, alt. 2,400 meters, 1911, Standley 8088 and 8179. This is another species of the macranthus group, apparently, with pubescence very different from that of the related species. The plant is common about Chama and Dulce, growing in the moist meadows along the small streams. Erigeron tonsus Wooton & Standley, sp. nov. Biennial or a short-lived perennial, from a slender, short root; stems very slender, flagelliform, spreading and forming dense mats, 20 cm. long or less, glabrous or with a few scattered, appressed hairs, bright green; basal leaves narrowly oblanceolate-spatulate, 6 cm. long or less, abruptly acuminate or acutish, attenuate to a slender petiole, entire; cauline leaves linear-oblanceolate to nearly linear, much reduced upward, numerous, acutish or acuminate, sessile, bright green, glabrous or thinly strigose; heads small, about 5 mm. broad, on long, naked, nearly glabrous, slender peduncles; bracts nearly equal, linear, acute, purplish, strigose; rays very narrow, about 50, 2.5 mm. long. Type in the U. S. National Herbarium, no. 690241, collected near the N Bar Ranch, August 2, 1900, by E. O. Wooton, ADDITIONAL SPECIMENS EXAMINED: Luna Valley, July 27, 1900, Wooton; north of Ramah, July 25, 1906, Wooton. This is clearly related to Frigeron fiagellaris, resembling that species in habit, but it is nearly glabrous and has smaller heads and more brightly colored rays. Eschenbachia tenuisecta (A. Gray) Wooton & Standley. Conyza coulteri tenuisecta A. Gray, Syn. Fl]. 17: 221. 1884. Very distinct from #. coulteri in its smaller and more numerous heads, and in its bipinnately parted leaves with very narrow segments. In EF. coulteri the leaves are mostly entire, but sometimes toothed or shallowly pinnatifid. Herrickia Wooton & Standley, gen. nov. Perennial herb with alternate, thick, rigid, toothed, sessile leaves; stems branched; heads solitary at the ends of the slender, leafy branches; rays purple; disk flowers perfect, tubular, yellow drying purplish, with short, nar- rowly ovate lobes; ray flowers pistillate; bracts of the involucre in several series, about equal, conspicuously keeled, with green, foliaceous tips and spinescent points, the outer bracts foliaceous and changing gradually into the proper leaves; receptacle convex, naked; style appendages lanceolate, acute ; achenes compressed, striate, glabrous; pappus simple, of numerous stout, simple, nearly equal, strongly barbellate bristles. The genus is named for the late Dr. C. L. Herrick, an enthusiastic and dis- tinguished scientist, for several years president of the University of New Mexico. Although chiefly interested in geology, he made large collections of plants in many parts of the State. Our plant seems not to agree with any of the published asteraceous genera. Doctor Gray, doubtless, would have placed it in the genus Aster, but to-day none would hold it congeneric with the typical representatives of that group. It is related to Xylorrhiza. The plants composing that genus, however, are mostly scapiform and have large heads with very different involucres and pubescent achenes. The rigid, spinulose teeth of the leaves of Herrickia are very characteristic. They give the plant a strange, xerophytic aspect strikingly like that of Perezia nana. Herrickia horrida Wooton & Standley, sp. nov. PLATE 50. Perennial with numerous slender, branched stems from a somewhat woody root; stems 30 cm, high or less, wiry, hirsute, glandular; leaves thick and Contr. Nat. Herb., Vol, 16, Plate 50. HERRICKIA HORRIDA WOOTON & STANDLEY. WOOTON AND STANDLEY—-NEW PLANTS FROM NEW MEXICO, 187 rigid, sessile, oblong or oblong-ovate, 20 to 45 mm. long, obtuse or acutish, sub- cordate or clasping, conspicuously veined, coarsely spinulose-serrate, puberulent ; involucres about 1 cm. high, the bracts linear-lanceolate to oblanceolate, glan- dular, spinulose-tipped; rays rather broad, 8 mm. long; achenes 3 to 4 mm. long; papus tawny, 6 mm. long. Type in the U. S. National Herbarium, no. 690258, collected at Baldy, August 14, 1910, by E. O. Wooton. The plant was first collected at Raton, August 22, 1897, by Dr. C. L. Herrick (no. 762). Specimens not yet in flower were gath- ered on the sides of Goat Mountain at Raton, June 22, 1911, by Paul C. Standley (no. 6330). The plants are very abundant about Raton, growing in the can- yons on the sides of the mountains in oak chaparral. EXPLANATION OF PLATE 50.—Part of type specimen. Slightly less than natural size. Aster hydrophilus Greene, sp. nov. in herb. Perennial or biennial; stems simple, bright green, 80 to 40 cm. high, stout, glabrous up to the inflorescence; cauline leaves linear or very narrowly linear- lanceolate, 6 cm. long or less, 3 to 4 mm. wide, attenuate, acute, sessile by a broad base, bright green, rather succulent, glabrous, with only the single mid- vein; heads few, 4 to 8, large, about 12 mm. wide; peduncles stout, ascending, 4 to 6 mm. long, densely glandular-puberulent, each bearing several small, re- duced, bract-like, lanceolate, acute, glandular leaves about 3 mm. long; involucral bracts irregularly imbricated, the outer successively shorter, green, oblong-linear, acute, densely glandular-puberulent; rays numerous, 10 to 12 mm. long, nearly 2 mm. wide, spreading; achenes strigose, the pappus of numerous slender, tawny, obscurely scaberulous bristles. Type in the U. S, National Herbarium, no. 498171, collected along the edge of Berendo Creek at the south end of the Black Range, Sierra County, Septem- ber 22, 1904, by O. B. Metcalfe (no. 1393). Altitude 1,500 meters, Similar to A. pauciflorus, but with much larger heads and greener herbage. Aster neomexicanus Wooton & Standley, sp. nov. Stems tall, up to 1 meter high, glabrous, smooth, green, sparingly branched below or simple; leaves thick and rather fleshy, green, glabrous, few and incon- spicuous, linear or linear-lanceolate, 8 cm. long or less, acute; inflorescence corymbosely much branched, the slender branches ascending, glabrous; heads solitary at the ends of the branches, 6 to 8 mm. high; peduncles with numerous linear-lanceolate, bract-like leaves; involucral bracts glabrous, in several series, green but with scarious margins; rays purple, narrow, 4 mm. long or less; achenes olive gray, striate, nearly glabrous. Type in the U. S. National Herbarium, no. 382479, collected at Roswell in August, 1900, by F. S. and Esther S. Earle (no. 327). Altitude about 1,140 meters. ADDITIONAL SPECIMENS EXAMINED: Near Lake Arthur, August 1, 1905, Wooton; Roswell, August 28, 1902, Cockerell. In general appearance the plants resemble A. paucifiorus, but that is usually a smaller plant and has strongly glandular peduncles and involucres. Machaeranthera amplifolia Wooton & Standley, sp. nov. Biennial or perhaps perennial, 70 em. high or lower; stems slender, sparingly paniculate-branched above, the branches ascending, almost glabrous below, puberulent above, the hairs very fine and short, occasionally one of them gland- tipped, but the stems scarcely viscid; leaves bright green, the basal ones broadly spatulate, 40 mm. long and 25 mm. wide or less, broadly obtuse, narrowed at the base to a long, slender, winged petiole, the petiole as well as the base of . 188 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. the leaf ciliate; leaves with coarse, shallow, rather obtuse teeth; upper leaves oblong-obovate or narrowly oval, sessile, often by clasping bases, serrate, obtuse, obscurely puberulent or glabrous; heads solitary at the ends of the branches, 15 mm. broad and 12 mm. high, with numerous narrow, violet rays 2 em. long; bracts linear, acute or acuminate, some of them with subulate tips, green for two-thirds their length, puberulent, only slightly glandular, the tips spreading, the lowest bracts reflexed; achenes straw-colored, glabrous. Type in the U. 8S. National Herbarium, no. 562446, collected high up in Fil- more Canyon of the Organ Mountains, September 28, 1906, by E. O, Wooton and Paul GC. Standley. ADDITIONAL SPECIMENS EXAMINED: Filmore Canyon, October 23, 1904, October 29, 1904, Wooton; Old Tiptop, Organ Mountains, October 18, 1908, Metcalfe. A very handsome plant with large heads and showy, bright-colored rays. It is related to M. asteroides, but the leaves are very different, likewise the pubescence. Machaeranthera angustifolia Wooton & Standley, sp. nov. Stems reddish, slender, erect, 80 cm. high or less, simple below, corymbosely branched above, cinereous; leaves linear-oblong to linear-oblanceolate, entire, densely scabrous, especially beneath, 1-nerved, acute, sessile by slightly nar- rowed bases, the uppermost leaves reduced: beads few, hemispheric, 15 mm. wide and 10 mm. high or smaller, with numerous bluish-purple rays; bracts unequal, thick, green above for half their length or more, linear-subulate, their tips squarrose, rather densely glandular near the tips; achenes almost or quite glabrous. Type in the U. S. National Herbarium, no. 45774, collected in New Mexico, probably in the Sandia Mountains, in 1853, by J. M. Bigelow. Another sheet of the same collection, which we have examined, is in the Gray Herbarium. ADDITIONAL SPECIMENS EXAMINED: Bear Canyon, Sandia Mountains, Septem- ber 24, 1897, Herrick; Dulce, alt. 2,200 meters, 1911, Standley 8124; Pecos, 1904, Bartlett 105; north of Glorie‘a, alt. 2.220 meters, 1908, Standley 5217, 5265; Tunitcha Mountains, 1911, Standley 7765, 7620. Machaeranthera aquifolia Greene, sp. nov. in herb. Stems very slender, erect, 80 cm. high or less, sparingly branched near the base, the branches simple up to the loosely paniculate inflorescence, sparingly scabrous, abundantly glandular; leaves linear-oblong, acute or nearly obtuse, thin, sparingly spinulose-dentate, sessile by scarcely narrowed bases, somewhat scabrous, bright green, the uppermost reduced; heads few, hemispheric, 14 mm. broad and 10 mm. high or smaller; bracts very unequal, green for half their length, with spreading tips, linear-subulate, the outermost reflexed, finely glandular-puberulent; achenes scaberulous. Type in the U. 8. National Herbarium, no. 495763, collected at the Gila Hot Springs in the Mogollon Mountains, Socorro County, August 26, 1908, by O. B. Metcalfe (no. 856). Altitude 1,950 meters. ADDITIONAL SPECIMENS EXAMINED: Hop Canyon, 1895, Herrick 628; Magda- lena Mountains, 1895, Zerrick 616; Grand Canyon of the Gila, August 19, 1900, Wooton; Mangas Springs, alt. 1,430 meters, 1903, Metcalfe 715; East Canyon, 1911, Holzinger; Gilmores Ranch, July 25, 1901, Wooton. The plant occurs in Arizona as well. The specimens from the Magdalena Mountains are stouter and have more heads than the typical form. Machaeranthera centaureoides Greene, sp. nov. in herb. Stems 70 cm. high or less, stout, green, simple below, corymbosely branched above, the branches ascending, glabrous below, rather densely scabrous and WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 189 glandular above, the pubescence especially abundant on the branches of the inflorescence; lower leaves firm and thick, oblong-linear, 11 cm. long or less, 15 mm. wide or narrower, acute, bright green, mostly glabrous, closely and sharply spinulose-dentate, sessile by narrow bases, the uppermost leaves sessile by broad, clasping bases, some of the leaves almost entire, obtuse; heads few, 14 mm. wide and 12 mm. high or smaller, hemispheric, with numerous narrow, violet rays; bracts linear-subulate, green for two-thirds their length, the tips all spreading and some of the lowest bracts reflexed, unequal, densely and coarsely viscid-pubescent ; achenes almost or quite glabrous, Type in the U. S. National Herbarium, no, 495484, collected in the Mogollon Mountains on the Middle Fork of the Rio Gila, Socorro County, August 9, 1903, by O. B. Metcalfe (no, 440). Altitude about 2,250 meters. ADDITIONAL SPECIMENS EXAMINED: West Fork of the Gila, August 25, 1903, Metcalfe; Luna, July 28, 1900, Wooton. Similar to M. asteroides, but distinguished by its stouter stems and narrow, thick leaves, and especially by the coarse, glandular pubescence Machaeranthera pygmaea (A. Gray) Wooton & Standley. Macheeranthera tanacetifolia pygmaea A. Gray, Pl. Wright. 2: 74. 1853. Aster tanacetifolius pygmaeus A. Gray, Syn. Fl. 17: 206. 1884. Machaeranthera simplex Wooton & Standley, sp. nov. Perennial or biennial from a thick, somewhat woody root; stems several, clustered, very slender and wiry, erect, 40 cm. high or less, simple or bearing 2 or 3 heads on short peduncles, reddish, glabrous below, sparingly cinereous above; leaves thin, green, oblanceolate to oblong-lanceolate, acute, tapering to the base, coarsely salient-serrate, nearly glabrous except on the ciliate margins and villous veins; heads about 12 mm. broad; bracts linear, with long, linear, spreading, green tips, cinereous, nowhere glandular; rays narrow, 8 to 10 mm. long. Type in the U. 8S. National Herbarium, no. 382533, collected in the Capitan Mountains at an altitude of 2,100 to 2,250 meters, August 31, 1900, by F. S. and Esther S. Earle (no. 390). This is near M. asteroides, but differs in having thin, oblanceolate, nearly glabrous leaves, few heads, and very slender, simple stems. Aphanostephus perennis Wooton & Standley, sp. nov. Slender perennial; stems wiry, erect, 830 cm. high or less, much branched, the branches ascending, rough-puberulent or hispidulous; leaves all linear, entire, bright green, 15 mm. long or less, cinereo-scaberulous; heads few, 8 or 9 mm. broad, on slender peduncles; bracts numerous, much imbricated, elliptic- lanceolate, acute, puberulent; rays rather few, pinkish; achenes dark brown, nearly terete, almost glabrous. Type in the U. S. National Herbarium, no. 564549, collected at Knowles, July 29, 1909, by E. O. Wooton. This may be readily distinguished from our other species by its perennial root, wiry stems, narrow leaves, and harsh pubescence. Dicranocarpus dicranocarpus (A. Gray) Wooton & Standley. Heterospermum dicranocarpum A. Gray, Pl. Wright. 1: 109. 1852. Dicranocarpus parviflorus A. Gray, Mem. Amer. Acad. n. ser. 5: 322. 1854. Wootonia parvifiora Greene, Bull. Torrey Club 25: 122. 1898. ° There is no doubt that the genera Dicranocarpus and Wootonia are the same, as suggested long since by Mrs. Brandegee and Mr. M. E. Jones. The types of the two genera came from the same general region, from localities probably not more than 100 miles apart. It is interesting to find that both are 190 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. included in Dalla Torre and Harms’s Genera Siphonogamarum, but with an interval of just one hundred genera between the two! Gymnolomia brevifolia Greene, sp. nov. in herb. Perennial; stems 50 to 60 cm. high, slender, brownish, with sparse, short, appressed, grayish pubescence; leaves ovate, rhombic-ovate, or elliptic, or the lowest oblanceolate, 20 to 40 mm. long, 10 to 20 mm. wide, acute, entire or obscurely serrate, the upper on short, winged petioles, the lowest on more slender petioles 10 mm. long, appressed-pubescent, scaberulous above, bright yellowish green; heads 10 mm. in diameter; bracts lanceolate, acuminate, canescent; rays showy, bright yellow; achenes obovate, glabrous, dark brown, without pappus. Type in the U. S. National Herbarium, no. 495518, collected in the Mogollon Mountains on the West Fork of the Rio Gila, altitude 3,300 meters, August 15, 1903, by O. B. Metcalfe (no. 511). Ividently this is closely related to G. multiflora, but not more so than is G. longifolia. It is distinguished especially by its broad and remarkably short leaves. It seems to grow at a far higher altitude than most of our species of the genus. Helianthus canus (Britton) Wooton & Standley. Helianthus petiolaris canescens A. Gray, Pl. Wright. 1: 108. 1852, not H. canescens Michx. Helianthus petiolaris canus Britton, Mem. Torrey Club 5: 334. 1894. This seems to be worthy of specific rank. It is nearest H. petiolaris, but in general appearance is very different, chiefly because of the abundant white pubescence on leaves and stems. The pubescence of the peduncles is of long, spreading hairs, while in H. petiolaris it consists of short, appressed ones. Helianthus neomexicanus Wooton & Standley, sp. nov. Tall, branched perennial; stems comparatively slender, reddish, hispid or hispidulous up to the inflorescence; leaves all except the uppermost opposite, on slender petioles one-fourth to one-third as long as the blades, these ovate- lanceolate, thick, 8 to 15 em. long, 6 em. wide or less, rounded or narrowed and acute at the base, attenuate or long-acuminate at the apex, sparingly ser- rate with low teeth, scabrous on the upper surface, beneath soft-villous; heads rather few, the disk 15 mm. broad, on long, slender, densely canescent pedun- cles; bracts lanceolate, with long, abruptly acuminate, spreading tips, ciliate below the middle, scaberulous on the back, 13 mm. long or less; achenes not seen. Type in the U. 8. National Herbarium, no. 563826, collected at Mangus Springs, August 19, 1902, by E. O. Wooton. It is not certain that this is a perennial plant, but in its various character- istics it agrees better with the perennial than the annual species. We have only the upper branches. The pubescence of the lower surface of the leaves is very different from that found in any of our other New Mexican species, exactly matching that of the southeastern Helianthus tomentosus Michx. Verbesina oreophila Wooton & Standley, sp. nov. Tall plant, probably perennial, 1 meter high or more; stems erect, stout, not winged, minutely puberulent; leaf blades triangular-lanceolate to deltoid-ovate, 8 to 11 cm. long, 7 cm. broad or less, acute, narrowed at the base to a short, slender petiole, coarsely serrate, scaberulous on the upper surface, beneath soft-pubescent; heads numerous, about 12 mm. in diameter, loosely cymose, on slender, naked peduncles; bracts oblanceolate, linear-oblong or lanceolate, WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 191 acute, 5 mm. long or less, appressed-pubescent; rays bright yellow, conspicuous, obovate-spatulate; achenes 5 mm. long, cuneate-obovate, brown, glabrous, closely invested by the pales. Type in the U. 8S. National Herbarium, no. 563951, collected at Cloudcroft, in the Sacramento Mountains, August 24, 1899, by E. O. Wooton. A species of the section Verbesinaria, but very unlike anything listed for the section by Robinson and Greenman in their revision of the genus.’ It has larger heads than most members of the section. Thymophylla hartwegi (A. Gray) Wooton & Standley. Hymenatherum berlandiert Benth. Pl. Hartw. 18. 1839, not DC. Hymenatherum hartwegi A. Gray, Pl. Wright. 1: 117. 1852. Thymophylla neomexicana (A. Gray) Wooton & Standley. Adenophyllum wrightit A. Gray, Pl. Wright. 2: 92. 1858, not Hymenatherum wrightit A. Gray, 1849. Hymenatherum neomesicanum A. Gray, Proc. Amer. Acad. 19: 40. 1883. Thymophylla thurberi (A. Gray) Wooton & Standley. Hymenatherum thurberi A. Gray, Proc. Amer. Acad, 19: 41. 1883. Hymenopappus fisheri Wooton & Standley, sp. nov. Probably biennial, from a long, slender root; stems several, erect, stout, 30 to 40 cm. high or more, leafy, the leaves gradually reduced toward the top; basal leaves not seen; cauline ones large, once pinnatifid into narrowly oblong to broadly linear, obtuse segments, densely arachnoid beneath, above sparingly woolly when young, but soon glabrate and bright green; heads numerous, corymbose, about 8 mm. in diameter, on slender peduncles; bracts sparingly tomentose, obovate, 5 mm. long, with thin, yellow, scarious tips; corolla bright yellow, glandular, the lanceolate lobes about equaling the campanulate throat; achenes dark brown, nearly naked on the faces, densely villous-ciliate on the angles; pappus conspicuous, about equaling the diameter of the achene at the summit. Type in the U. S. National Herbarium, no. 564948, collected near Nara Visa, in clay and sandy soil, September 8, 1910, by Mr. Geo. L. Fisher (no. 16). Also collected at Nara Visa, June 12, 1911, Fisher 177. The plant is suggestive of H. artemisiaefolius DC., but the bracts lack the very broad, white margins of that species, the leaf segments are narrower, and the corollas are bright yellow instead of nearly white. Hymenopappus nudatus Wooton & Standley, sp. nov. Perennial, 380 to 50 cm. high, with several stems clustered at the top of a thick, black root; leaves practically all basal, 10 cm. long, twice pinnate into linear divisions, densely arachnoid, the petioles very densely so, their bases covered with long, white wool; stems scapose, densely arachnoid, bearing but 1 or 2 much reduced leaves; heads few, 1 to 5, large, 12 to 15 mm. broad, on long, stout peduncles; bracts 8 mm. long or less, unequal, obovate or oblong, with scarious stramineous tips, densely arachnoid; corolla pale yellow, the oblong- ovate, obtuse lobes not more than one-third as long as the tube; achenes densely silky-villous with tawny hairs; pappus of prominent scales nearly 2 mm. long. Type in the U. S. National Herbarium, no. 495226, collected in the Burro Mountains, Grant County, altitude 1,650 meters, June 6, 1903, by O. B. Metcalfe (no. 107). ADDITIONAL SPECIMENS EXAMINED: West of Patterson, June 21, 1892, Wooton; Cactus Flat, July 5, 1906, Wooton; near Santa Rita, 1877, Greene; Silver City, June 2, 1880, Greene. Proc. Amer. Acad. 34: 534-566. 1899. 192 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. The type collection was originally determined as H. luteus Nutt., but that is a smaller plant with smaller heads and more leafy stems. Our plant sug- gests H. arenosus, but that, too, has smaller heads and lower, more leafy stems. Picradeniopsis dealbata (A. Gray) Wooton & Standley. Bahia dealbata A. Gray, Mem. Amer. Acad. n. ser. 4: 99, 1849. Bahia absinthifolia deatbata A. Gray, Pl. Wright. 1: 121. 1852. Hymenoxys brachyactis Wooton & Standley, sp. nov. Perennial or possibly biennial, from a thick tap-root; stems solitary, very stout, simple at the base, with a few erect, corymbose branches above; basal leaves long-petiolute, once or twice pinnatifid into numerous narrowly linear segments; cauline leaves very numerous, all once or twice pinnatifid into linear or filiform divisions, the lower leaves with very numerous divisions, the upper with but few; heads numerous, 6 mm. in diameter, on short, slender, glabrous peduncles; involucres glabrous, 5 mm. high, the outer bracts lanceolate, acute, strongly keeled, united for nearly half their length, the inner ones very obtuse, short-acuminate, tomentose on the margins; rays pale yellow, cuneate-obovate, much shorter than the involucral bracts, usually not more than half as long; disk flowers bright yellow, sparingly glandular; achenes villous with pale tawny hairs; palee of the pappus long-attenuate, two-thirds as long as the disk corollas. Type in the U. S. National Herbarium, no. 690242, collected near East View, August 4, 1906, by E. O. Wooton. The plant has the size and habit of H. rusbyi, but it has very different leaves and heads. Its most striking characteristics are its tall, stout, simple stems, its finely divided Artemisia-like foliage, and the numerous small, angled heads. Hymenoxys cockerellii Wooton & Standley, nom. nov. Hymenoxys chrysanthemoides jugta Cockerell, Bull. Torrey Club 31: 508. 1904. This seems to be a distinct species, differing conspicuously enough from H. chrysanthemoides, as orginally pointed out by Professor Cockerell. Ordi- narily, in raising a subspecies to specific rank we should not think of changing the subspecific name unless it were preoccupied. There is, however, so far as we know, no precedent for the use of a preposition as a specific name, nor do we think it desirable or even permissible that one be established. Hymenoxys mearnsii (Cockerell) Wooton & Standley. Hymenoxrys chrysanthemoides mearnsii Cockerell, Bull. Torrey Club 31: 506. 1904. A common species of southern New Mexico, ranging from the southwest _ eorner as far east as the Pecos Valley. It is the smallest of our annual forms. Tetraneuris formosa Greene, sp. nov. in herb. Perennial from a thick root; caudices cespitose, thickened; basal leaves oblanceolate-spatulate, 60 cm. long and 9 mm. broad or smaller, obtuse, green, sparingly silky-pubescent on both surfaces, glandular-punctate; peduncles about 20 cm. long, slender, bearing 2 or 38 small leaves near the base, rarely branch- ing, pubescent with few loosely appressed hairs; heads large, the involucres 12 mm. broad and 7 or 8 mm. high; bracts oblong, obtuse, densely silky-villous; rays large, extending 15 mm. beyond the involucre, 7 or 8 mm. broad, with 3 rounded teeth at the apex, conspicuously veined; achenes villous; palee of the pappus ovate, with long, abruptly acuminate tips. Type in the U. S. National Herbarium, no. 498042, collected on dry hills near Kingston, Sierra County, at an altitude of 1,980 meters, August 22, 1904, by O. B. Metcalfe (no, 1285). : WOOTON AND STANDLEY—NEW PLANTS FROM NEW MEXICO. 1938 A specimen collected in the Magdalena Mountains in June, 1881, by G. R. Vasey appears to be the same, although it has more densely pubescent leaves and stems. The species is related to 7. leptoclada, but has very much longer rays and larger heads and is a taller plant. The pappus scales are broader and not so acute. Tetraneuris pygmaea (Torr. & Gray) Wooton & Standley. Actinella depressa pygmaea Torr. & Gray, Mem. Amer. Acad. n. ser. 4: 100. 1849. The type came from the Raton Mountains of New Mexico. We have speci- mens from the Sandia Mountains that we take to be the same plant. It is similar to 7’. depressa, but the leaves are densely sericeous instead of villous, the heads are not nearly so large as in that species, and the rays are shorter than the involucre instead of equaling it. Artemisia albula Wooton, nom. nov. Artemisia microcephala Wooton, Bull. Torrey Club 25: 455. 1898, not Hillebr. 1888. Artemisia petrophila Wooton & Standley, sp. nov. Low shrub, 30 to 40 cm. high, the lower branches spreading, bearing numerous very slender, mostly simple, erect ones; stems densely tomentose; the lower leaves, i. e. those of the older thicker branches, cuneate, 2 to 3 cm. long, densely white-sericeous, tridentate at the apex, the teeth oblong or oblong- lanceolate, obtuse; leaves of the upper slender branches oblanceolate to linear- oblong, entire, obtuse or acutish, white-sericeous, rather distant, never crowded ; inflorescence of very narrow panicles, the lateral branches not more than 2 cm. long, slender; heads homogamous, few-flowered, solitary or in small clusters, sessile or pedunculate, the peduncles soon recurved; involucre campanulate, 2.5 mm. broad, the bracts few, in about two series, oblong, obtuse, densely tomentose. Type in the U. S. National Herbarium, no. 686323, collected on a dry sand- stone mesa at the north end of the Carrizo Mountains, July 28, 1911, by Paul C. Standley (no. 7355). ADDITIONAL SPECIMENS EXAMINED: NEW Mexico—Dry hills near Farmington, alt. 1,650 meters, July 17, 1911, Standley 7084. Arizona—Moki Reservation, 1896, Hough 58. ‘ . This low shrub is similar to Artemisia tridentata, especially in herbarium specimens. Even here, however, it may be distinguished by its broader inflores- cence, recurved peduncles, campanulate heads, and rather distant, mostly entire leaves. In the field the two are very unlike. A. tridentata isa densely branched shrub usually a meter high, growing on the plains, while this is a low plant with slender, open, never dense branches, growing on the hills, usually among rocks. Senecio metcalfei Greene, sp. nov. in herb. Perennial, multicipitous, from a rather stout, ascending, somewhat woody rootstock; stems scapiform, 20 cm. high or mostly less, stout, glabrous or with very scanty tomentum; leaves numerous, erect or somewhat spreading, less than half as long as the stem, rather thick and firm, dull green, oblanceolate or narrowly spatulate, obtuse, mostly entire or some with 2 or 3 low, incon- spicuous teeth, glabrous except the very youngest, these with a scanty tomen- tum, narrowed at the base into a petiole as long as the blade or shorter; leaf blades 45 mm. long and 15 mm. wide or smaller; stems almost naked but with a few much reduced leaves; heads few, on short peduncles, campanulate, mostly about 9 mm. long or smaller; peduncles mostly naked; bracts linear or linear- 194 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. lanceolate, acute, 6 mm. long or less, light green; rays oblong, rather bright yellow, 6 mm. long and 3 mm. wide or smaller; achenes striate, glabrous. Type in the U. S. National Herbarium, no. 497842, collected on open slopes on Hillsboro Peak, at the south end of the Black Range, May 27, 1904, by O. B. Metcalfe (no. 938). Altitude 3,060 meters. Most closely related, perhaps, to S. pentodontus, but with very different leaves and pubescence. Senecio remifolius Wooton & Standley, sp. nov. Perennial, multicipitous, from a rather stout, creeping or ascending rootstock ; stems scapiform, 25 cm. high or lower, glabrous, slender; basal leaves linear- oblanceolate, glabrous, or with an obscure and very sparse tomentum visible only under a lens, about 60 mm. long and 8 mm. wide, obtuse, entire or with 3 or sometimes a few more shallow teeth, thick and fleshy, deep green, gradually tapering at the base into a slender petiole as long as the blade or longer; lower cauline leaves like the basal ones but smaller; upper ones linear, sessile by a somewhat widened base; stems sometimes with an obscure tomentum in places ; branches of the inflorescence several, ascending, each bearing 1 to 3 rather long- pedunculate heads, these campanulate, about 12 mm. high; peduncles naked or with a few bractlets; involucral bracts 10 to 12, linear-lanceolate, acute, with membranous margins, about 7 mm. long; rays oblong, pale yellow, 10 mm. long and 2.5 mm. wide or smaller; achenes striate, ciliolate along the angles. Type in the U. 8. National Herbarium, no. 690231, collected along Willow Creek, August 8, 1900, by E. O. Wooton. Similar to the preceding but with different leaves, heads, and achenes, Senecio sacramentanus Wooton & Standley, sp. nov. Erect from a cluster of rather fleshy roots, 70 cm. high or less; stems mostly simple below but paniculately branched above, sparingly tomentose below, densely so above, the pubescence densest about the nodes; lee ves lanceolate to triangular-lanceolate, 14 em. long or less and 5 em. wide or narrower, abruptly acuminate, coarsely salient-dentate, tapering, truncate, or cordate at the base, sparingly puberulent beneath, glabrous above, bright green, thin; petioles of the lowest leaves 7 em. long, slender, dilated and clasping at the base, those of the upper leaves shorter and winged, the uppermost leaves sessile and often clasping by a broad base; stems leafy throughout, the upper leaves considerably smaller than the others ; inflorescence much branched, of paniculate racemes; heads very numerous, small, 10 mm. long and 9 mm. wide or less, rather nar- - rowly campanulate, nodding; bracts about 8, rather broadly oblanceolate, acute, with membranous, light-colored margins; heads subtended by 2 or 3 short, fili- form bracts; rays none; achenes light chestnut colored, striate, with abundant soft, barbellate pappus. Type in the U. S. National Herbarium, no. 690287, collected in the vicinity of Cloudcroft, near the summit of the Sacramento Mountains, altitude about 2620 meters, August 15, 1901, by E. O. Wooton. ADDITIONAL SPECIMENS EXAMINED: Cloudcroft, August 24, 1901, Wooton; James Canyon, August 11, 1899, Wooton. This is as nearly related to S. pudicus as to any species, but its foliage is very different. In that species the leaves are linear to oblanceolate, tapering to the base, and nearly entire. In S. sacramentanus the leaves vary from cordate-ovate to oblong-lanceolate, are coarsely serrate, and are mostly truncate to abruptly contracted at the base. The general appearance of the two is so unlike that at first glance one would not suspect the relationship between them. WOOTON AND STANDLEY—NEW PLANTS FROM NEW MExIco. 195 Carduus gilensis Wooton & Standley, sp. nov. Tall, sparingly branched biennial with slender, striate, villous stems; basal leaves oblanceolate, about 40 cm. long, acute, conspicuously lobed, the lobes dentate, the few teeth tipped with slender spines, glabrous on the lower surfaces, puberulent above; upper cauline leaves oblong or triangular-lanceolate, acute, clasping at the base, shallowly lobed, the lobes and their principal teeth with numerous long, slender, salient spines; heads usually solitary at the ends of the branches, campanulate, about 3 cm. broad or more and of the same height, subtended by many narrowly linear-lanceolate, spiny, bract-like leaves; outer bracts foliaceous, linear-lanceolate, with elongated tips, often arachnoid on the margins, pectinate with very numerous spiny teeth; inner bracts broader, scaberulous, little or not at all dilated at the tips and laciniate; corollas green- ish yellow. Type in the U. 8. National Herbarium, no. 495440, collected in the Mogollon Mountains on the West Fork of the Rio Gila, Socorro County, altitude 2250 meters, August 4, 1903, by O. B. Metcalfe (no. 377). This, like O. inornatus, was determined as CO. parryi. It is of that group, but is distinguished by its large, mostly solitary heads, as well as by its unusually large and thin leaves, and the very numerous foliaceous bracts. Carduus inornatus Wooton & Standley, sp. nov. Tall biennial about 1 meter high with a stout stem, this simple below, above with a few ascending branches; stems striate, sparingly arachnoid, densely so on the younger parts, nearly glabrous in age; basal leaves not seen; lower cauline leaves linear-lanceolate, 10 to 18 cm. long, 17 mm. wide or less, with few remote, triangular, spine-tipped teeth, the margins beset with fine spines, glabrous beneath except on the midvein, sparingly villous there as well as on the upper surface with long, weak, white hairs; upper cauline leaves lanceolate or oblong, acute or attenuate, sessile and clasping at the base, the auricles rounded, the margins irregular and bearing numerous slender, yellow spines; heads few, occasionally solitary at the ends of the branches but usually in clusters of about 3, pedunculate, campanulate, 25 mm. long and 20 mm. broad or smaller, subtended by numerous spiny, reduced, bract-like leaves; bracts of the involucre in several series, successively shorter outward, the outer linear- lanceolate with long-attenuate tips, mostly glabrous on the back, rarely slightly arachnoid, spine-tipped, the margins bearing many weak, yellow spines; inner bracts broader, scaberulous, most of them abruptly dilated at the tips into a lanceolate or oval, often laciniate, spine-tipped portion; corollas yellow ; achenes obovate, compressed, brownish, 4 to 5 mm. long, the pappus about 10 mm. long. Type in the U. S. National Herbarium, no 5610138, collected in the Sacramento Mountains near Cloudcroft, August 24, 1901, by E. O. Wooton. Originally this collection was determined as C. parryi, and it is nearer to that than to any other species. It differs in having fewer heads and slightly if at all arachnoid bracts with pectinately spiny margins and less dilated tips. The leaves, too, are not nearly as spiny as in that species, and only the inner- most bracts have dilated tips, while in C. parryi almost all have them. Carduus pallidus Wooton & Standley, sp. nov. Tall biennial, 1 to 2 meters high, with simple, very leafy stems sparingly branched about the inflorescence; stems stout, arachnoid above, becoming gla- brate below, striate; lower cauline leaves lanceolate, acute, uttenuate to the base, irregularly serrate-dentate, the teeth tipped with short, weak spines; upper cauline leaves narrowly oblong to triangular-lanceolate, clasping at the base, with rounded auricles, acute irregularly dentate or shallowly lobed, the 196 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. margins spine-tipped, the blades glabrous beneath, above sparingly white- villous, lanate along the midrib; inflorescence of rather few heads, these race- mose in age, crowded when young, on short, stout peduncles or sometimes ses- sile, subtended by reduced, very spiny leaves; heads campanulate, 3 cm. high and as broad or smaller; bracts linear, none of them with dilated tips, the outer arachnoid on the backs and’ margins, tipped with a long, slender spine and usually bearing 2 or more slender lateral spines just below the tip; inner bracts lanceolate, thick and firm, scaberulous, with slender, flat, weak tips; corollas greenish yellow; achenes oblong-obovate, dark brown, glabrous and shining. Type in the U. 8S. National Herbarium, no. 498699, collected in the Pecos River National Forest near Winsors Ranch, altitude 2520 meters, July 16, 1908, by Paul C. Standley (no. 4357). ADDITIONAL SPECIMENS EXAMINED: Tunitcha Mountains, 1911, Standley 7533; Chama, alt. 2400 meters, 1911, Standley 6763; mountains west of Las Vegas, 1881, Vasey; Baldy, August 14, 1910, Wooton; Gilmores Ranch, alt. 2220 meters, 1907, Wooton & Standley 3491; Gilmores Ranch, July 14, 1895, Wooton; James Canyon, August 3, 1899, Wooton. A common species in the higher mountains of New Mexico, occurring chiefly in the Transition Zone, although frequently extending farther up. It grows usually in swamps or marshes along the edges of mountain streams, sometimes in shaded thickets, frequently in open meadows. It is a tall, coarse plant, with pale yellowish stems and foliage. Commonly this has passed as Carduus parryi. Apparently no one has ever questioned this determination, yet examination of herbarium material reveals the fact that true parryi does not occur in New Mexico. That species differs from ours in its small heads, more spiny leaves, and the conspicuously dilated bracts. Carduus vinaceus Wooton & Standley, sp. nov. Tall biennial, 1 to 2 meters high, with very numerous ascending branches; stems brownish purple, striate, slender, glabrous; basal leaves glabrous, green, 30 to 50 cm. long, 20 cm. wide or less, elliptic-oblong in outline, pinnatifid nearly to the midrib, the segments overlapping, laciniately lobed, the lobes oblong-lanceolate, acute, the teeth tipped with short, slender, yellowish spines; heads very numerous, naked, campanulate; bracts in numerous series, narrowly lanceolate, with long, flat, weak, spreading tips, deep reddish purple through- out, glabrous on the back, scarcely keeled, ciliate or puberulent on the margins, tipped with short, slender, yellowish spines; inner bracts with long, slender, twisted tips; whole head 5 cm, in diameter and 4 em, high or smaller; corolla lobes long and narrow, purplish; achenes obovate, brown, glabrous, with tawny, plumose pappus 15 to 20 mm. long. Type in the U. S. National Herbarium, no. 690246, collected in the Sacra- mento Mountains near Fresnal, July 12, 1899, by E. O. Wooton. No other North American species of which we have seen either specimens or description is at all like this in the form of the involucre. Some of the Mexican species suggest our plant but not very closely. When growing it is a handsome large plant with very numerous, purplish heads and dark stems, these contrast- ing with its glabrous, bright green foliage. SUPPLEMENTARY NOTES ON AMERICAN SPECIES OF FESTUCA. By CHARLES V. PIPER. Since the publication of his monograph of the North American species of Festuca,’ the author has availed himself of opportunities to study the type specimens of several species, which has resulted in clearing up various matters of synonymy. In the course of these studies, it also became apparent that the North American grass referred to Festuca fratercula Rupr. is not that plant at all, but an unnamed species. Festuca sororia sp. nov. Festuca fratercula Contr. U. 8S. Nat. Herb. 10: 39. 1906, as to description and specimens cited. An examination of the type of Mestuca fratercula Rupr. preserved in the herbarium of the Jardin Botanique de l’Etat at Brussels shows it to be a very different species, and one not represented in the National Herbarium. Festuca sororia is a rare plant occurring in Colorado, Arizona, and New Mexico in mountain ravines. Type in the U. S. National Herbarium, no, 45866, collected in the Rincon Mountains, Arizona, altitude 225 meters, September, 1891, by G. ©. Nealley (no. 177). The species has been further collected, as follows: Cotorapo: Pagosa Peak, Baker 36, 75, 94, 177, 178; Durango, Tweedy 398a. New Mexico: Hillsboro Peak, Grant County, O. B. Metcalfe 1236. Festuca fratercula Rupr.;.Fourn. Mex. PI. 2: 124. 1881. “ Culmo fere tripedali scabro; foliis 3’’’ latis, retrorsum scabris, longis, planis, apice longe convolutis; ligula brevissima; panicula libera effusa fere pedali folium summum longe superante, radiis geminis divaricatis inaequalibus parce divisis; spiculis 3-floris cum terminali quarto abortivo; glumis inaequalibus acutis, floribus teretibus remotis, palea inferiore acuta potissimum in floribus summis breviter mucronata, glabra; squamulis lanceolatis ovarium aequantibus. “In humidis inter Pinos montis Orizabensis, 11-12000’' (Gal. n. 5778); Cumbre de Estepa, augusto (Liebm.).” To the above description the following notes may be added: Culm stout, smooth, bearing two leaves; lower leaf blade 20 cm. long, strongly nerved above, firm, scabrous on nerves and margins, attenuate-acute; ligule a fringe of short bristles: lower sheath 30 cm. long, smooth outside, scabrous within near the 1Contr. U. S. Nat. Herb. 10: 1-48. 1906. 197 198 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. top; panicle 25 cm. long; rays slender, in twos, the longest 12 cm. long, very scabrous, naked for half its length; spikelets 10 to 12 mm. long, 5-flowered: glumes firm, lanceolate, acute, minutely scabrid near the tip, obscurely veined, the lower 5 mm., the upper 6 mm. long; lemmas firm, thickish, very obscurely veined, acute, purplish near the tip, 7 mm. long; palea slightly exceeding the lemma. There are no specimens in the National Herbarium that can be referred to this species. Festuca tolucensis H. B. K. Nov. Gen. & Sp. 1: 153. 1816. In Contributions from the U. 8. National Herbarium, volume 107, the treat- ment of this species and its supposedly near allies F. multiculmis Steud? and F. aequipaleata Fourn.’, was admittedly unsatisfactory. The types of all three are in the herbarium of the Muséum d’Histoire Naturelle at Paris. Festuca tolucensis, aS shown by more abundant recently collected material, is rather variable, especially in the size of the spikelets and the development of the awns. Festuca multicuimis Steud. must be regarded as a synonym, as the type (Heller 306, Mount Toluca) is identical with the type of F. tolucensis from the same mountain. Lestuca aequipaleata Fourn. differs only in the tendency of the spikelets to be smaller and in the unawned or. very short-awned lemmas. On some specimens both awned and awnless lemmas occur, and other specimens show complete intergradation to F. folucensis. It is therefore best considered a subspecies.* Festuca texana Steud. Syn. Pl. Glum. 1: 310. 1854. The type is preserved in Steudel’s herbarium in the Muséum d’Histoire Naturelle, Paris. It was collected by Drummond (no. 387) in Texas, at San Felipe, as indicated by the specimen at Kew. The plant is Leptochloa fascicu- laris (Lam.) A. Gray. Festuca pseudoduriuscula Steud. Syn. Pl. Glum. 1: 312. 1854. The type of this, Drummond’s no. 389 from Texas, is also in Paris. It is Festuca obtusa Spreng., as is also Drummond's no. 398, indicated by Steudel as a variety but not named in publication. According to the specimens at Kew, nos. 389 and 398 are both from San Felipe, Texas. Festuca glabra Spreng. Syst. Veg. 1: 358. 1824. The original specimen of this is in the Kénigliches Botanisches Museum, Dahlem, Berlin, and is labeled “ Seacoast, Long Island, Nuttall.” It is the ordi- nary Atlantic coast form of Festuca rubra L. *Page 44. * Syn. Pl. Glum. 1:310. 1854. * Mex. Pl. 2:125. 1881. * FESTUCA TOLUCENSIS AEQUIPALEATA ( Fourn.) Festuca aequipaleata Fourn. Mex. Pl. 2: 125. 1881. The following specimens of Festuca tolucensis have been examined : Mexico (Srare): Mount Toluca, Bonpland in 1833 (type); Heller 306 (type of F. multiculmis) ; Rose & Painter 7983; E. W. Nelson 14. MoreELos: Mount Popocatapetl, Hitchcock 6001; Rose & Hay 5735, 6297, 5978a. JaLisco: Mount Nevada, Hitchcock 7167, 7158, 71653. PursLa: Mount Orizaba, Seaton 193, 228; Liebmann 510, 511 in part. Specimens of Festuca tolucensis acquipaleata have been examined as follows: PUEBLA: Mount Orizaba, Liebmann 510, 511 in part. MoRELos: Mount Popocatapetl, Rose & Hay 5697. San Luis Porosr: Specific locality not stated, Parry & Palmer 924. PIPER—-SUPPLEMENTARY NOTES ON FESTUCA. 199 Festuca delawarica Kunth, Rév. Gram. 1: 129. 1829. Poa delawarica Link, Hort. Berol. 1: 174. 1827. The type is in Berlin, labeled “ Delaware (Bernhardi)”. The specimen was apparently grown in the Botanic Garden at Berlin. It is the species American botanists call Puccinellia distans. Festuca villiflora Steud. Syn. Pl. Glum. 1: 313, 1854. The type is in Steudel’s Herbarium at Paris labeled “ Labrador, Missionary Albrecht” and consists of one small plant. It is Festuca rubra kitaibeliana (Schultes) Piper, as previously surmised from Steudel’s description. Festuca megalura Nutt. Journ. Acad. Phila. n. ser. 1: 188. 1847. This has as synonyms F. commutata Steud.’ and F. chaetantha Kunze, cited as a synonym under fF’ sciuroides by Desvaux.? Both types are in Paris. Festuca eriolepis Desv. in Gay, Fl. Chil. 6: 428, 1853. The type is in the herbarium of the Muséum d’Histoire Naturelle, Paris. It is not the plant described and figured in Contributions from the U. 8. National Herbarium volume 10, but is the same as Festuca octoflora hirtella Piper.’ The Chilean specimens have usually but 5 florets to each spikelet, a condition that also occurs in dwarfed specimens from California and Oregon. * Syn. Pl. Glum. 1:304, 1854. Contr. U. S. Nat. Herb. 10:18. 1906. “In Gay, Fl. Chil. 6:427. 1850. *Op. cit. 12. DELPHINIUM SIMPLEX AND ITS IMMEDIATE ALLIES. By CHARLES V. PIPER. Delphinium simplex Dougl. and a few related species are distin- guished among American larkspurs by the following combination of characters: Roots grumose or tuberiform; flowers short-pediceled and the inflorescence therefore narrow and strict; sepals erect or but little spreading; seeds smooth and produced into thin margins on the angles. Four species have been described as belonging to this group, namely D. simplex Dougl., D. distichum Geyer, D. strictum A. Nels., and D. burkei Greene, the last-named to me unknown. In the original description of Delphinium simplex Dougl., Hooker * gives the locality as follows: “On the subalpine range west of the Rocky Mountains near the Columbia, plentiful.” In the Herbarium at Kew are two sheets collected by Douglas labeled Delphinium sim- plex. One of these bears the legend “On the subalpine range of the Rocky Mountains near springs, plentiful.” Some one has written “type” on this sheet, a conclusion which seemingly admits of no doubt. The other Douglas sheet is from “ Grassy points of land on the banks of the Wallawallah.” On this sheet is written in Doctor Gray’s handwriting “The type of D. simplex. A. G.” In the British Mu- seum is a sheet of the “ Wallawallah ” collection, but none of the first mentioned. As the two collections represent the same species no error could have originated by choosing either as the type, but nevertheless the plant described by Doctor Gray in the Synoptical Flora is a different species from Douglas’s original. Apparently Doctor Gray understood the species aright when he examined the material at Kew, but later was misled by a Douglasian specimen in the Gray Herbarium which he supposed was true Delphinium simplex. Based on a knowledge of three species in the field and the material in the U. S. National Herbarium the following revision of the group is offered. 1H]. Bor. Amer. 1: 25. 1829. 201 202 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. KEY TO SPECIES. Flowers usually 10 to 20; leaves rather few; puberulence, at least of the inflorescence, villous in character and somewhat viscid _-.-_-_..--_------eee 1. D. eyanoreios, Flowers more numerous; leaves rather many ; puberulence not ‘ villous, more or less appressed. Leaves puberulent, usually densely so, all with narrow divisions; puberulence of inflorescence usually viscid___ 2. D. simples. Leaves glabrous or mostly so, the lower with broad divi- sions; puberulence of inflorescence usually not viscid__ 3. D. distichum. 1. Delphinium cyanoreios sp. nov. Stems strict, erect, usually simple, 40 to 99 em. high from a thick tuberiform root, at least the inflorescence densely velutinous and somewhat viscid, the jower portion of the plant puberulent or glabrate; leaves few, the basal usually glabrous, parted into 3 cuneate divisions, these again cleft into linear, obtuse lobes; upper leaves reduced and more deeply cleft into narrower segments; raceme 10 to 30-flowered, 10 to 30 em. long, the pedicels decidedly shorter than the spurs; bracts linear-lanceolate: calyx dark blue, sparsely villous without. the straight spur 12 to 14 mm. long, much longer than the sepals; upper petals whitish, veined, tinged with blue at the tips; carpels 8, 12 to 20 mm. long with- out the beak, villous, straight or slightly spreading at maturity; seeds dark, smooth, the angles produced into narrow white margins. This is the plant Dr. Gray described as D. simplex Dougl.,’ but it is not the plant of Douglas. It oceurs mainly in the Blue Mountains of Oregon. Type in the U. 8. National Herbarium, no. 529341, collected near Sled Creek Ranger Station, Wallowa County, Oregon, June 25, 1907, by Frederick V. Coville (no. 2444). Among the specimens in the U, S. National Herbarium are, in addition, the following: OREGON: Blue Mountains, Sheldon 8399: Billy Meadows, Blue Mountains, 1,500 meters alt., J. T. Jardine 812: head of Anthony Creek, Blue Moun- tains, 2,100 meters alt., Cusick 2389; Silver Creek, Harney County, Cusick 2609; eastern Oregon, Cusick 2181. 2204; Steens Mountain, Cusick 1982; Two Color Creek, Blue Mountains, Cusick 3305a; Cliff Creek, Blue Mountains, Cusick 33825; source of Rock Creek, Blue Mountains, Cusick 3175. IDAHO: Craig Mountains, June 23, 1894, Henderson ; South Clearwater, June 26, 1894, Henderson; near Sawtooth, Everman 603; Long Valley, Hen- derson 3140. Delphinium depauperatum Nutt. is distinguished from the present species almost wholly by the more elongated lower pedicels, but some specimens like Cusick’s nos. 3325 and 3175 are ambiguous in this respect. The pubescence of the two species is identical. 2. Delphinium simplex Dougl.; Hook. Fl. Bor. Amer. 1: 25. 1829. In its original locality this species grows in stream bottoms and moist meadows and has uniformly pale, dull blue flowers. Douglas’s types represent this form. Elsewhere in eastern Washington the flowers are a deeper, brighter blue, but the other characters remain the same. This brighter flowered plant matches the type of Delphinium strictum A. Nels., but it is scarcely worthy of nomenclatorial recognition. ‘Syn. Fl. 1' : 49. 1895, PIPER—DELPHINIUM SIMPLEX AND ITS ALLIES. 203 Representative examples of this species in the U. S. National Herbarium are the following: WASHINGTON: Pullman, July 21, 1894, Piper; Elmer 3; July 30, 1895, Hardwick; without definite locality, Vasey 160, 161; Wenache, Whited 1277; Tieton Basin Cotton 449; Hllensburg, Whited 523; Spokane County, Suksdorf 720; Falcon Valley, July 17, August, 1881, Suksdorf. OREGON: Pine Valley, Cusick 2440; John Day River, H. BE. Brown 72; Camas Prairie, Grigiths € Hunter 86. BritisH CotumBia: Mouth of Kootenai River, July, 1890, Macoun. 3. Delphinium distichum Geyer; A. Gray, Syn. Fl. 1°: 49. 1895. This is very closely allied to D. simplex Dougl. and seems to intergrade fully. Typically it is distinguished by the broader lobed, nearly glabrous basal leaves. The true D. simpler Dovgl. and D. distichum Geyer make up together the Delphinium distichum of the Synoptical Flora. On the whole Hooker’s original treatment of the plant as a variety of Delphinium simpler (D. simpler dis- tichiflorum Hook.)' is perhaps the wisest disposition. 1 Hook. Lond. Journ. Bot. 6: 67. 1847. ~ THE IDENTITY OF HEUCHERA CYLINDRICA. By CHARLES V. PIPER. In the original description of Heuchera cylindrica Douglas,’ the species is stated to occur “On the declivity of low hills, and on the steep banks of streams on the west side of the Rocky Mountains.” This has always been puzzling from the fact that the plant accepted by all American botanists as Heuchera cylindrica occurs only in and west of the Cascade Mountains, and grows mainly in copses about low moist meadows. An examination of the Douglasian specimens at Kew and in the British Museum clears up the discrepancy. At Kew there are two sheets, one of which is labeled exactly as quoted by Hooker. The other sheet is of the same species, and with scarcely a doubt of the same collection, but is labeled merely Heuchera cylindra- cea [sic] Dougl. The plant is not at all the one taken by American botanists for Heuchera cylindrica, but is the one recently named Heuchera columbiana by Rydberg.? Another Douglasian sheet in the British Museum has two different species mounted upon it, but each is numbered and the label reads: “In dry mountain woods near Kettle Falls. 1. H.cylindracea. 2. Heuchera.” This writing is in ink and is identified by Dr. A. B. Rendle as that of Sir W. J. Hooker. In the lower left-hand corner of the sheet written in pencil is the following: “ Heuchera cylindrica Dougl. Flor. Bor. Am.” Such pencil inscriptions in the same hand occur on many . of the Douglas sheets in the British Museum, but the handwriting is not identified. Plant no. 2, unnamed, is a young specimen of the species that all American botanists have been calling Heuchera cylin- drica, namely, Torrey and Gray,’ Brewer and Watson,‘ Wheelock,’ Howell,® Rydberg,’ Rosendahl,* and Piper.° As the plant called Heuchera cylindrica by American botanists is unnamed, it is here described as a new species. 1 Hook. Fl. Bor. Amer. 1: 236. °F. Northw. Amer. 203. 1898. 2N. Amer. Fl, 22:116. 1905. ™N. Amer. Fl. 227:115. 1905. 3H], N. Amer. 1:580. 1840. ® Bot. Jahrb. Engler 377: 80. 1905. *Bot. Calif. 1:201. 1876. *Contr. U. S. Nat. Herb. 11:322. 1906. * Bull. Torrey Club 17: 202. 1890, 205 206 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Heuchera chlorantha sp. nov. Tufted from a stout branching caudex; leaves all basal, the blades 5 to 8 cm. broad, orbicular to reniform, deeply cordate, 7 to 9-lobed, the lobes broadly rounded or faintly lobed and dentate with broad cuspidate teeth, glabrous ex- cept on the veins beneath; petioles 5 to 15 cm. long, with spreading or some what retrorse, usually brownish hairs; flowering stems scapiform, 50 to 100 em. high, villous, with spreading or somewhat retrose brownish hairs, and usually bearing one or two fimbriate leaf rudiments; inflorescence a narrow dense panicle 5 to 10 cm. long, the short branches 8 to 7-flowered the axis hir- sute and glandular; bracts ovate to lanceolate, the lower 3-cleft, fimbriate; calyx greenish, turbinate-campanulate, the lobes oblong, obtusish. the whole pruinose-puberulent, 7 to 8 mm., or in fruit 10 to 12 mm. long; petals minute or lacking. Type in the U. 8. National Herbarium, no, 690786, collected in Falcon Valley, Klickitat County, Washington, June 28, 1892, by, W. N. Suksdorf (no. 17389). This species differs from all others in the group by its leaves being mem- branaceous instead of subcoriaceous and from all except H. tenuifolia in its greenish flowers, the remainder having cream-colored flowers. The following additional specimens have been examined: WASHINGTON: Sumas Prairie, 1858, Lyall; Kitsap County, July, 1895, Piper; Seattle, Piper 2515; Tacoma, Flett 123; Montesano, Heller 4067; Falcon Valley, Suksdorf 1739 (type); Vancouver, Piper 4944; Upper Columbia, Wilkes Exped. 461 (this locality probably wrong, as many of the labels on Wilkes Expedition plants are undoubtedly misplaced). OREGON : Cascade Mountains, Cusick 2675; Cascade Mountains, latitude 44°, Newberry; Farewell Bend, Crook County, Leiberg 481; Elk Creek, Jackson County, Applegate 25738; Lake of Woods, Klamath County, Coville & Applegate; Western Oregon, June, 1880, Howell 137; Silver Creek, Hall 161. BriTisH CoLuMBIA: Skidegate. Graham Island, Queen Charlotte Island, Osgood in 1890. Heuchera cylindrica Dougl. (H. columbiana Rydb.) belongs to an exceedingly puzzling group of the genus. Rydberg, in the North American Flora, recognizes six species, mainly on the basis of leaf contour and pubescence. The characters assigned to the various species do not separate herbarium material into homogeneous units, and additional forms would require names if the scheme used were carried out fully. It would seem preferable to consider Heuchera cylindrica a polymorphous species with several intergrading subspecies. Typical examples of Heuchera cylindrica Dougl. are the following: WASHINGTON: Loon Lake, July 20, 1897, Winston; Clark Springs, Kreager 71; Wenache Mountains, Whited 1134; Clark Springs, Beattie &€ Chap- man 2019. IpaHo: Near Lewiston, Sandberg, Heller & McDougal 161. Just how the confusion as regards the identity of Heuchera cylindrica arose is difficult to say. In the Gray Herbarium is a specimen marked “ Ex Hooker,” which is not the true H. cylindrica, but the plant to which all Amerkcan botanists have applied the name. Lindley* published a colored plate of Heuchera cylindrica from specimens grown in the garden of the London Horti- cultural Society, also stating that Douglas collected it ‘‘in dry mountain woods near the Kettle Falls.” This plate evidently represents true Heuchera cylin- drica, but modified by the conditions under which it was grown, so that it might easily be mistaken for the species here named Heuchera chlorantha. * Edwards’s Bot. Reg. 23: pl. 1924. 1887. NEW OR NOTEWORTHY SPECIES OF PACIFIC COAST PLANTS. By CHARLES V. PIPER. The identity of many of the older described species of northwestern plants can only be settled by an examination of the original speci- mens. The conclusions reached from an examination of some of these are presented herewith, together with descriptions of several new species sent to me from time to time by various correspondents. Luzula arcuata major Hook. Fl. Bor. Amer. 2: 189. 1840. The type specimens of this collected by Tolmie on Mount Rainier is at Kew. It is identical with Juncoides piperi Coville. Alsine viridula sp. nov. Perennial, forming dense, prostrate, circular mats, 10 to 20 cm. broad; herb- age entirely glabrous except for a few ciliate hairs at the bases of the leaves; stems 5 to 10 cm. long, green, scarcely shining, branched; leaves ovate, plane, short-acuminate, dull, rather thin, but the veins obscure, 5 to 15 mm. long, nearly as long as the internodes, each abruptly narrowed into a short petiole, this bearing a few long hairs; flowers solitary in the axils, the pedicels mostly shorter than the leaves; sepals oblong-ovate, acutish, green, with a narrow, white, scarious margin, thin, 3-nerved, 2 to 2.5 mm. long; petals none; pods green, ovoid, not longer than the calyx; seeds dark brown, 0.5 mm. long. Type in the U. S. National Herbarium, no. 249940, collected on ridges south from Wiessners Peak, Idaho, July 28, 1895, by J. B. Leiberg (no. 1396). Growing in springy places in canyons, altitude 1,700 meters. Also collected along rivulets in woods, altitude 1,400 meters, in the Blue Mountains, Columbia County, Washington, July, 1896, by C. V. Piper (no, 2328). This species is a close ally of Alsine crispa (Cham. & Schlecht.) Holzinger,’ but that has a much looser habit, with internodes much exceeding the leaves, shining stems, less distinctly petioled leaves lacking the basal hairs, pedicels mostly exceeding the leaves, and thick, lanceolate sepals which are shorter than the capsule and when mature 3-ribbed. The seeds, too, are larger and paler. Alsine crispa also occurs in Idaho and in the Blue Mountains, but it remains perfectly true in its characters. Oreobroma longipetala sp. nov. Root stout, simple or branched, about 6 cm. long; leaves linear, 2 to 3 cm. long, about 2 mm. wide; flowering stems simple or bearing 1 to 3 erect branches, 2 Alsine crispa (Cham. & Schlecht.) Holzinger, Contr. U. S. Nat. Herb. 3: 216. 1895. Stellaria crispa Cham. & Schlecht. Linnaea 1: 51. 1826. 61364? —13——2 207 208 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. much exceeding the leaves; bracts opposite or the upper alternate, sessile, ovate, mostly glandular-denticulate; flowers rose colored (?), long-pediceled ; sepals truncate or slightly rounded, 4 to 5 mm. long, with very shert, blunt, glandular teeth; petals 12 to 18 mm. long, Type in the U. S. National Herbarium, no. 10881, collected in the Sierra Nevada, California, in 1875, by J. G. Lemmon. Intermediate between O. pygmaea and O. oppositifolia. Arabis olympica sp. nov. Perennial from a stout, slightly branched caudex; stems erect, simple, 6 to 10 em. high, sparsely pubescent with branched hairs; basal leaves obovate to oblanceolate, obscurely crenate, obtuse, thickish, sparsely pubescent on the midrib and margins with simple hairs, the blade 5 mm. long on a shorter margined petiole; cauline leaves 3 to 8 mm. long, sessile, oblong to lanceolate, obtuse, entire or nearly so, the margins and midribs pubescent with mostly simple hairs; flowers corymbose, somewhat nodding, short-pediceled; sepals oblong, acutish, purple-tinged, 2 mm. long; petals white, as long as the sepals; pods erect, narrowly linear, obtuse, glabrous, 18 to 23 mum. long on stout, glabrous pedicels one-fourth as long; style stout, about 0.5 mm. long. Type in the U. S. National Herbarium, no. 690787, collected near Humes Glacier, Mount Olympus, Washington, August 12, 1907, by J. B. Flett. A near ally of A. furcata 8S. Wats., but distinguished by its much smaller flowers and the branched pubescence of the stem. Cheiranthus capitatus Hook. Fl. Bor. Amer, 1: 38, 1829. The type of this, collected by Douglas ‘on rocky places of the Columbia near the sea,” is in the Kew Herbarium, and 2 duplicate is in the British Museum. The plant is identical with Erysimum elatum Nutt. and has nothing to do with Hrysimum grandifiorum Nutt., to which it has commonly been referred. Astragalus glareosus, Dougl.; Hook. Fl. Bor. Amer, 1: 52, 1830. Wholly on the basis of the description the writer identified this species some- what hesitatingly with Astragalus allanaris Sheldon. In Hooker's description it was said to be “plentiful on dry gravelly banks of rivers, from the con- fluence of Lewis and Clarkes Rivers with the Columbia to the mountains.” At Kew there are two sheets of Douglas’s collecting, one labeled “On barren sandy grounds, common, a fine plant,” and the other “Abundantly on sandy dry grounds in the interior of the Columbia, 1826.” A duplicate of the second sheet is in the British Museum. All are flowering specimens of the plant later called Astragalus allanaris Sheldon. The plant of Colorado and Wyoming, heretofore referred to Astragalus glareosus, is quite different and is properly named Astragalus argophyllus Nutt. Lysimachia terrestris (1..) B. S. P. Prel. Cat. N. Y. 34. 1888. Viscum terrestre L. Sp. Pl. 1028. 1758. This species, not before recorded from west of the Rocky Mountains, has been collected at Shoalwater Bay, Washington, in ditches along a cranberry bog by Dr. Cora B. Eaton, July 1, 1910. Ligusticum scothicum I. Sp. Pl. 250. 1753. According to Hooker’ this plant was collected by Douglas at the mouth of the Columbia, and Gray’ later identified one of Cooper’s plants as this species—“ not rare along the coast at Shoalwater Bay.” What the Cooper ‘FL Bor, Amer. 1: 265. 1834. “U.S. Rep. Expl. Miss. Pacif. 12°: 62. 1860. PIPER—NEW OR NOTEWORTHY PACIFIC COAST PLANTS. 209 plant may be is uncertain, as the specimens have not been found either at Washington or in the Gray Herbarium. No Douglas specimens could be found either at Kew or in the British Museum, but in both herbaria are undoubted specimens of Ligusticum scothicum labeled “ Fort Vancouver, ex Herb. Gre- ville.’ While it would not be strange to find this plant on the Washington coast, inasmuch as it is common on the shores of Alaska, it has been collected by no recent botanist. It is quite certain that the plant does not grow at Fort Vancouver, and that the label is erroneous. as many other plants have been recorded from this place which do not occur there, Pentstemon cinereus sp. nov. Cespitose, with several to many slender erect stems from a much-branched woody base, the whole herbage densely and minutely canescent-puberulent and the inflorescence glandular; leaves all entire, subcoriaceous, the basal ones numerous, spatulate-lanceolate, acute, the blades 10 to 20 mm. long, usually a little shorter than the petioles; cauline leaves about 6 pairs, lanceolate, broad- est at the sessile base, 2 to 38 mm. long, gradually reduced upwards, the mar- gins somewhat involute; inflorescence narrow and strict, 10 to 20 em. long, the puberulence becoming glandular especially on the peduncles and flowers; peduncles erect, 3 to 7-flowered; pedicels very short; calyx 2 mm. long, the broadly ovate sepals abruptly acute, not margined; corolla dark blue, tubular, 10 to 12 mm. long, slightly bilabiate, puberulent on the outside. slightly hairy within; sterile filament bearded at the top; anthers splitting their whole length. Type in the U. S. National Herbarium, no. 690798, ccllected in dry rocky ground at Bend, Crook County, Oregon, July 4, 1907, by Kirk Whited (no. 8055a ). The species is perhaps most closely allied to P. collinus A. Nels. Plectritis congesta minor Hook. Fl. Bor. Amer. 1: 291. 1834. The original specimens of Douglas were said to be from “ near the mouth of the Columbia.” The specimens at Kew, apparently the originals, are labeled “near the ocean.” They are merely small plants of Plectritis congesta, and have nothing to do with Plectritis macrocera Torr. & Gray, under which P. congesta minor Hook. was cited as a synonym in the Synoptical Flora. Plectri- tis macrocera is not known to occur west of the Cascade Muntains. Erigeron filifolius (Hook.) Nutt. Trans. Amer. Phil. Soc. n. ser. 7: 308. 1841. Diplopappus filifolius Hook Fl. Bor. Amer. 2: 21, 1834. In the original description this plant, collected by Douglas, was said to be “Common on the Great Falls of the Columbia; and barren grounds of the interior.””’ There has been doubt as to whether this name properly applies to the plant with yellow flowers named Frigeron peucephyllus by Gray,’ or to a closely related species with violet or purple flowers. The doubt is probably traceable to Hooker’s original description, where the rays are described as “ flavescentibus.”” In the British Museum there are three sheets of Douglas’s specimens labeled Diplopappus filifolius. One of these is inscribed “ Sandy ground near the Great Falls of the Columbia, 1825.” This plant is the species described as Hrigeron filifolius in the Synoptical Flora of North America, and the sheet is doubtless the type or a duplicate type. The other two sheets are labeled ‘‘ Dry grassy plains of the Columbia, 1826.” Both are Hrigeron peuce- phyllus Gray. In the Kew Herbarium are also three Douglasian sheets of Diplopappus filifolius Hook., but all represent the same species as the “ Great Falls” plant in the British Museum. Thus it appears that Doctor Gray’s treatment in the Synoptical Flora is correct and that Hooker was misled as *Proc. Amer, Acad. 16: 89. 1810. 210 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. to the color of the rays either by their being faded, or, more likely, by the two sheets of Hrigeron peucephyllus that he associated with the “ Great Falls” plant. The question still remains open as to the identity of Diplopappus linearis Hook.’ Search for the original specimens of this, both at Kew and at the British Museum, was unsuccessful. Doctor Gray cited the name with doubt as 2 synonym of Hrigeron ochroleucus Nutt. Aster columbianus sp. nov. Stems slender, erect, 40 to 70 cm. high, somewhat puberulent, much branched above and bearing numerous crowded small heads; leaves linear, entire, acute, sessile by a broad base, 5 to 8 cm. long, glabrous, or nearly so, except the scabrous-ciliate margin; reduced leaves of the branches very numerous and more strongly ciliate; involucre turbinate, 5 mm. high; bracts linear, in about 3 series, somewhat squarrose, the outer cuspidate and ciliate, the inner acute and not ciliate, all green at the tips; rays violet, 4 to 5 mm. long; pappus sordid; achenes canescent. Type in the U. S. National Herbarium, collected at Waitsburg Washington, by R. M. Horner. In Contributions from the U. S. National Herbarium, volume 11,’ this was referred to A. amethystinus Nutt., but that species has the leaves scabrous on both surfaces. The whole aspect is that of Aster campestris Nutt., but that has the inflorescence glandular. The species has been collected as follows: WASHINGTON: Wawawai, Piper 1602; Waitsburg, Horner 559 (type), 627; Okanogan River, Sereno Watson 191. OREGON: Wallowa Mountains, Piper 2513. Aster delectus sp. nov. Stems puberulent, 10 to 30 cm. high; leaves entire, lanceolate or oblanceolate, acute, puberulent especially beneath, sessile by a narrowed base or the lower with margined petioles; inflorescence corymbiform, of 3 to 15 rather crowded small heads; involucre rather hemispheric, 5 mm. high; bracts pubescent but not ciliate, mostly obtuse, in about 3 series; rays violet, 5 mm. long; pappus sordid. Type in U. S. National Herbarium, no. 411493, collected in a moist meadow near the Sycan River, Klamath County, Oregon, August 15, 1901, by W. C. Cusick (no. 2761). A close ally of A. occidentalis Nutt. Aster paludicola sp. nov. Glabrous throughout, the rather slender stems 60 to 80 em. high; leaves linear-lanceolate, sessile by a slightly narrowed base, entire, acute, 8 to 12 em. long, 5 to 7 mm. wide, only the midnerve evident; inflorescence loosely cymose, the medium-sized heads few; involucre 6 to 8 mm. high, of 20 to 25 loose bracts in about three indistinct series; bracts glabrous, acute, green at tip, chartaceous below the middle; rays violet, 8 mm. long; pappus sordid achenes canescent. Type in the U. 8. National Herbarium, no, 620391, collected in Darlingtonia swamps, at Hight Dollar Mountain, Josephine County, Oregon, August 15, 1907, by C. V. Piper. The species is remarkable for the few bracts to the involucre. Its alliance is apparently with Aster occidentalis Nutt. ‘Fl. Bor. Amer, 2: 21, 1834, * Page 572. THE AMERICAN SPECIES OF MEIBOMIA OF THE SECTION NEPHROMERIA. Ly J. N. Rose anpD Pau C. STANDLEY. Of the several sections of the genus Meibomia, that designated as Nephromeria by Bentham’ is one of the smallest. Besides the Ameri- can species, a single one is known from Sumatra. Hitherto six American species have been recognized, one of which is but imper- fectly known. Three additional ones are described in the present paper, making a total of nine. The section Nephromeria is distinguished from the others of the genus by the few reniform or subreniform joints of the loment. In some species this reniform outline is not very pronounced, but there is always a more or less evident sinus along the dorsal suture of each joint. The species fall readily into two groups. In the first, composed of but two species, feibomia mollis and M. scopulorum, the loments are sessile, the 3 lowest joints usually do not develop, but are small and abortive, and the tips of the calyx lobes are subulate. Most important of all, the plants are low and erect. In the second group, which includes the seven other species, the loments are commonly stipitate, only the lowest joint (if any) is abortive, and the calyx lobes are flat and broad, far from subulate. Most, if not all, of the species of this group are coarse, high-climbing vines. The joints of their loments are generally much larger and broader than those of the smaller group. But few specimens of these plants, excepting only Meibomia mollis, are found in herbaria. Explorations in Central America and southern Mexico during the last few years have supplied the U. S. National Herbarium with ampler material, among which we have detected several apparently undescribed species, besides representa- tives of all but one of those previously known. *In Benth. & Hook. Gen. Pl. 1: 520, 1865, as a subsection of the section Heteroloma. 211 212 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. KEY TO THE SPECIES. Three lowest joints of the loment usually not developing, twisted; loments sessile; tips of the calyx lobes subulate. Leaflets linear-lanceolate; terminal joint of the loment semiorbicular _~-_--_-_-_ 1. A. scopulorum. Leaflets lance-ovate; terminal joint of the loment ellip- tic-oblong, abruptly narrowed at both ends_______ 2. M. mollis. Only the lowest joint, or sometimes none, abortive, the joints not twisted; loments usually stipitate; tips of calyx lobes not subulate. Flowers and fruit in a dense panicle; often more than two joints of the loment developing___.__________ 3. M. skinneri. Flowers and fruit not in a dense panicle, the inflores cence open; never more than two joints of the loment developing. Leaflets orbicular, rounded at the apex_________ 4, M. painteri. Leaflets ovate or lanceolate, acute or acutish. Leaflets thick, glabrous, and shining above______ 5. M. metallica. Leaflets thin, pubescent, and dull above. Loments subsessile; leaflets with a broad, silvery stripe along the midrib_______ 6. M. albonitens. Loments conspicuously stipitate; leaflets not striped. Leaflets apiculate; joints very large, 25 to 380 mm. long_____~~______ 7. M. lunata, Leaflets not apiculate; joints smaller, less than 20 mm. long. Joints narrowed at the base; leaf- lets lance-ovate _..._________ 8. M. angustata, Joints not narrowed at the base: leaflets ovate_______________ 9. AM. barclayi. 1. Meibomia scopulorum (S. Wats.) Rose & Standley. Puare 51, b. Desmodium scopulorum S. Wats. Proc. Amer, Acad. 24: 47. 1889. Stems erect, slender, much branched, pale, minutely uncinate-ouberulent ; leaves trifoliolate; petioles slender, 30 to 85 mm. long, striate, puberulent ; leaflets linear-lanceolate, 55 to 80 mm. long, acute, rounded at the base, glabrous except along the minutely uncinate veins; stipules lanceolate, long-acuminate, persistent; petiolules 2 mm. long; inflorescence of terminal or axillary, slender panicles or racemes, the branches minutely puberulent: flowers solitary or in twos, on slender pedicels 10 mm. long; calyx teeth ovate, with subulate tips ; corolla not seen; bracts filiform, short; loments sessile, of usually 4 joints, all except the terminal one abortive, the lowest three twisted, densely puberulent ; terminal joint semiorbicular, 10 mm. long, with scarcely any suture on the upper edge, the walls thin, finely puberulent. Type locality, “Guaymas,” Sonora, Mexico. Type collected by Dr. Edward Palmer in 1887 (no. 258). Specimens examined: Sonora: On rocky ledges in the mountains about Guaymas, 1887, Palmer 258, type collection. ROSE AND STANDLEY—MEIBOMIA, SECTION NEPHROMERIA. 213 2. Meibomia mollis (Vahl) Kuntze, Rev. Gen. Pl. 198, 1891. Puate 51, d. Hedysarum molle Vahl, Symb. Bot. 2: 83. 1791. Desmodium molle DC, Prodr. 2: 332. 1825. Stems erect, stout, herbaceous, much branched, minutely uncinate-puberulent ; petioles 35 mm. long or less, those of the uppermost leaves very short, uncinate- puberulent; leaflets 3, ovate to lanceolate, 24 to 70 mm. long, rounded or some- times subcordate at the base, acutish, of about the same color on both surfaces, sparingly puberulent above, finely soft-pubescent beneath; petiolules 2 or 3 mm, long; stipules 4 mm. long, triangular-subulate; inflorescence mostly of axillary and terminal simple racemes disposed so as to form a panicle, the racemes 25 cm. long or less; flowers usually in clusters of 3 or 4, on slender pedicels 6 mm. long; bracts short, filiform, villous, deciduous; calyx cleft almost to the base, the lobes about equal, lanceolate, puberulent, their tips subulate; corolla green- ish yellow, 3 or 4 mm. long; loments sessile, of usually 4 joints, all except the terminal one abortive, twisted, densely puberulent; terminal joint elliptic- oblong, somewhat narrowed at both ends, 9 mm. long, with a shallow sinus on the upper edge, minutely pubescent, with thin, membranous walls. Type locality, Island of St. Croix, West Indies. Specimens examined: Porro Rico: Sandy soil, Culebra, March, 1906, Britton & Wheeler 238; Juana-Diaz, “ad vias circa Escalabrado,’’ November 30, 1885, Sintenis 2900. Sr. Crorx: Jerusalem, January 8, 1896, Ricksecker 198. Cusa: Limestone hillside, Guantanamo Bay, March, 1909, Britton 1940. Costa Rica: Foréts de Nicoya, January, 1900, Tonduz, Inst. Fis. Geogr. Costa Riea 13590. Mexico: Limestone ledges near Iguala, Guerrero, October 25, 1900, Pringle 9264: Tomellin Canyon, Oaxaca, September 7, 1906, J. VN. & J. S. Rose 11327. The Mexican specimens differ in having the terminal joint glabrous except along the margin, where it is minutely puberulent ; otherwise there seems to be no essential difference from West Indian specimens. 8. Meibomia skinneri (Benth.) Kuntze, Rev. Gen. Pl, 198. 1891. PLATE 51, h. Desmodium skinneri Benth.; Hemsl. Diag. Pl. Mex. 3: 47. 1880. Stems stout, suffrutescent, usually erect, glabrous in age, the younger branches with dense, retrorse, soft pubescence, the hairs usually tawny; leaves numerous, large, trifoliolate; petioles stout, 5 em. long or less, soft- pubescent, striate; stipules lanceolate, 7 or 8 mm. long, deciduous, membrana- ceous, strigillose; leaflets ovate to oblong or rhombic-lanceolate, often unequal at the base, the terminal one largest and broadest, 12 cm. long or less, acutish to obtuse at the apex, rounded at the base, sparingly strigillose above, paler and densely velvety-pubescent beneath, the pubescence tawny along the veins, elsewhere white; petiolules very short and stout; inflorescence a dense, much branched, leafy or naked, terminal panicle, axillary racemes or panicles often present, the branches often sparingly uncinate; flowers on filiform pedicels 3 mm. long; bracts subulate, very short, hirtellous, deciduous; calyx appressed- pubescent, scarcely bilabiate, the teeth ovate and acute; corolla deep purple, G6 or 7 mm. long; loments of usually 3 joints, raised on a slender stipe 2 or 3 mm. long; constrictions very narrow; joints semiorbicular, with thin, mem- 914 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. branous walls, 11 to 18 mm. long, 7 mm. wide, with a very shallow sinus on the upper side, puberulent when young, becoming glabrous. Type locality, “Guatemala.” Type in the Kew Herbarium, collected by Skinner (no. 37). Specimens examined: GUATEMALA: 1891, J. D. Smith; Garrucha, Depart. Chimaltenango, alt. 1350 meters, March, 1892, Heyde & Lua 3287; 1892, Heyde 519; Lake Amatit- lan, February 11, 1905, W. A. Kellerman 4355. Mexico: Roadside between Mascota and San Sebastian, Jalisco, March 14, 1897, H. W. Nelson 4049; roadside between San Sebastian and the summit of the mountain known as the “Bufa de Mascota,” Jalisco, alt. 1800 ° meters, March 20, 1897, Nelson 4109. 4. Meibomia painteri Rose & Standley, sp. nov. PLATE 51, a. Stems herbaceous, climbing, stout, angled, rather sparingly uncinate-hirsute; leaves large, trifoliolate; petioles 3 to 7 em. long, stout, uncinate-hirtellous; leatiets orbicular or broadly oblong, all of about the same size, 3 to S cm. long, sparingly strigillose above and more densely so beneath, slightly paler beneath, the veins large and conspicuous; petiolules stout, 5 to 7 mm. long, hirtellous; stipules persistent, ovate; inflorescence of simple, axillary racemes, or these sometimes sparingly branched, 2 or 3 from each axil, the branches uncinate- puberulent; flowers not seen; pedicels 6 mm. long; bracts deciduous: loment of 1 or 2 joints, on a stipe 8 mm. long, the lower. joint often abortive, the terminal one sometimes smaller than the basal; constrictions very narrow; joints orbicular-reniform, 10 to 12 mm. long and almost as wide, with a shallow suture upon the upper edge, the center hard and turgid, surrounded by a rather thin wing, the whole strongly reticulate-veined, sparingly and very finely puberulent. Type in the U. S. National Herbarium, no, 690911, collected in Iguala Canyon, near Iguala, Guerrero, Mexico, September 21, 1905, by C. G. Pringle (no. 18688 ). From all other species of the group this may at once be distinguished by its orbicular leaflets and the peculiar inflorescence. The joints of the loment, too, are not like those of any other species. The species is named for Mr. J. H. Painter, formerly Assistant Curator in the Division of Plants of the U. S. National Museum, who, at the time of his death, was preparing to monograph the genus Meibomia. He had indicated this plant as a new species in the herbarium, but failed to give it a name. 5. Meibomia metallica Rose & Standley, sp. nov. PLATE 51, c. Stems stout, climbing, suffrutescent below, smooth, conspicuously uncinate- hirtellous; leaves ample, numerous, trifoliolate; petioles 25 to 60 mm. long, vather slender, uncinate-hirtellous; stipules lanceolate, attenuate, deciduous; leaflets ovate to ovate-lanceolate, rounded at the base, acute, mucronate, 7 to 11 cm. long, glabrous and shining above with an almost metallic sheen, thick and coriaceous, sometimes sparingly pubescent in the youngest leaves, densely soft- pubescent or sericeous beneath, most of the hairs whitish but those along the conspicuous veins bright yellow and longer; petiolules stout, 2 or 8 mm. long, densely pubescent with tawny hairs; inflorescence of terminal, sparingly branched panicles, or of simple, axillary racemes, the branches stout, densely uncinate-hirtellous; flowers on slender pedicels 6 mm. long; bracts soon de- ciduous, lanceolate, long-acuminate, 4 mm. long, hirsute; calyx finely puberulent, large, about 4 mm. long, almost equally 5-toothed, the teeth ovate, abruptly short-acuminate; corolla purple; loment raised on a slender stipe 3 mm. long, of usually 2 joints, the lower one often abortive; joints quadrate-orbicula#, almost straight on the upper edge but with an obtuse suture 2.5 mm. deep, about ROSE AND STANDLEY—MEIBOMIA, SECTION NEPHROMERIA. 215 20 mm. long, thin, conspicuously reticulate, finely hispidulous; constrictions very narrow. * Type in the U. 8. National Herbarium, no. 408084, collected by O. F. Cook and R. F. Griggs at Cajabon, Alta Verapaz, Guatemala, March 30, 1902 (no. 354). The specimen on this sheet is in fruit, but a flowering branch of the same collee- _ tion is mounted on sheet no, 408083. The thick leaflets, glabrous and shining on the upper surface, distinguish the plant from all the related species, but good additional characters are found in the fruit. Other specimens examined: GUATEMALA: Sehachicha, Alta Verapaz, March, 1902, H. von Tuerckheim 8367. 6. Meibomia albonitens (Lem.) Rose & Standley. Rhynchosia?.albonitens Lem, Il]. Hort. Lem. 1861: pl, 290. 1861. Desmodium skinneri albolineatum Hook. Curtis’s Bot. Mag. pl. 5452. 1864. Desmodium skinneri albonitens Hook. loc. cit. Desmodium scutatum Hemsl. Diag. Pl. Mex. 3: 46. 1880. Meibomia scutata Kuntze, Rev. Gen. Pl. 198. 1891. Type locality, “Tierras templadas” of Mexico. The species was described from plants cultivated from seeds collected by Ghiesbreght. We have seen no specimens corresponding to the plates cited and to Hemsley’s description of the species. That author describes the loments as subsessile and the joints as being one to one and one-half inches in diameter. The leaflets are jllustrated as having a lighter, silvery stripe running along the midribs. None of our specimens of the section Nephromeria agree with these requirements. Meibomia lunata seems to be very closely related and comes from about the same region where M. albonitens was probably collected, but that species has conspicuously stipitate loments and its leaflets are of the same shade of green throughout. 7%. Meibomia lunata (T. S. Brandeg.) Rose & Standley. PLATE 51, f. Desmodium lunatum T. S. Brandeg. Zoe 5: 246. 1908. Stems somewhat woody below, climbing, slender,’ the older ones glabrous, the younger uncinate-hirtellous; leaves trifoliolate; stipules deciduous; petioles slender, 2 to 5 cm. long, uncinate; leaflets all of about the same size, 55 to 95 mm. long, lanceolate, acute, mucronate, dull green above, paler beneath and densely sericeous, sparingly strigillose above, the veins few and inconspicuous; petiolules stout, 2 to 3 mm. long, villous; inflorescence of terminal or axillary simple racemes or panicles, with uncinate-puberulent branches; flowers on slender pedicels 2 or 3 mm. long; bracts deciduous, less than 4 mm. long, lance- ovate, acuminate, reddish, pilose with yellowish hairs; calyx sparingly villous, the teeth ovate, acuminate; corolla purplish, 5 or 6 mm. long; joints large, 24 mm. long or less and almost as wide, quadrate-orbicular, with an acute suture 5 mm. deep on the upper edge, conspicuously reticulate-veined, puberu- lent, usually with an acute beak at the apex; constrictions 1 to 2 mm. wide. Type locality, “In openings of forests near Zacuapan,” Vera Cruz. Mexico. Type collected in 1906 by C. A. Purpus, no. 1907. Specimens examined: Mexico: Zacuapan, Vera Cruz, 1906, Purpus 1907, type collection ; Zacuapan, January, 1907, Purpus 2961. 8. Meibomia angustata Rose & Standley, sp. nov. PLATE 51, 9. Stems slender, flexuous, herbaceous, the older ones almost glabrous, the younger very finely and rather sparingly uncinate-puberulent ; leaves numerous, 216 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. trifoliolate; petioles slender, 17 to 30 mm. long, angled, finely retrorse-pubes- cent; stipules “deciduous; leaflets ovate-lanceolate, rounded at the base, obtuse, 24 to 80 mm. long, of about the same color on both surfaces, sericeous on both sides, more densely so beneath; petiolules stout, 3 mm. long, densely pubescent; inflorescence of few, rather remote, simple racemes 17 cm. long or less, the slender branches puberulent.; flowers on slender pedicels 1 to 2 mm. long; bracts subulate or lanceolate, very small, early deciduous; calyx finely pubes- cent, the teeth triangular, acute; corolla deep purple, 5 mm. long; loment of a single joint, raised on a slender stipe 2.5 mm. long; joints semiorbicwlar, 15 mm. long, conspicuously narrowed and acute at the base, rounded at the apex, with an obtuse sinus 2 mm. deep on the upper edge, finely puberulent, the walls very thin and transparent, the persistent and very slender style 3 mm. long. Type in the U. S. National Herbarium, no. 327048, collected along the road- side between San Sebastian and Las Palmas, Jalisco, Mexico, altitude 300 to 600 meters, March 30, 1897, by E. W. Nelson (no. 4126). This plant is more closely related to M. skinneri than to any other member of the group, but it differs in the uniformly solitary joint, deep sinus, narrow base of the joint, and more scanty and open intlorescence. It is impossible to tell from our specimens whether the stems are erect or climbing. 9. Meibomia barclayi (Benth.) Rose & Standley. PLATE 51, e. Desmodium barclayi Benth. Bot. Voy. Sulph. 88. 1844. Stems herbaceous, climbing, flexuous, uncinate-hirtellous, the branches di- verging almost at right angles to the main axis; leaves trifoliolate; petioles 2 to 3 cm. long, uncinate-hirtellous; leaflets broadly ovate to rhombic-ovate, the terminal one largest, about 4 cm. long, the lateral ones 2 to 3 em. long, all puberulent on both surfaces, paler beneath; petiolules about 1 mm. long; stipules semiovate, acuminate, 4 mm. long, striate; inflorescence of terminal or axillary, sparingly branched panicles, these usually about 10 ecm. long, the branches sparingly uncinate; flowers not seen; loment of usually 2 joints connected by a very narrow isthmus, raised on a stipe 3 mm. long, the pedicels 4 or 5 mm. long; joints reniform-orbicular, 9 to 12 mm. long and almost as wide, with an acute suture 2 mm. deep upon the upper edge, the central portion bard and turgid, brown, surrounded by a broad and thin, green wing, the whole con- spicuously reticulate, glabrous except along the puberulent edges. Type locality, “Central America.” Type collected by Barclay. Hemsley: cites * a specimen of Barclay’s from Nicaragua. This is probably the type. Specimens examined: Costa Rica: Buissons 4 Nicoya, January, 1900, Tonduz, Inst. Fis. Geogr. Costa Rica, no. 13577. The specimen cited answers the original description of Desmodium barclayi very well, and is the only such plant examined. EXPLANATION OF PLATE 51.—Loment, a, of Meibomia painteri, from type specimen: b, of M. scopulorum, from a specimen of the type collection; c, of M. metallica, from type specimen; d, of M. mollis, from a specimen collected by Ricksecker (no. 198) at Jerusalem, St. Croix; e, of M. barclayi, from Nicoya, Costa Rica, Tonduz 13577; f, of M. lunata, from a specimen of the type collection; g, of ‘M. angustata, from type speci- men; h, of M. skinneri, from specimen collected by Heyde and Lux, near Garrucha, Guatemala (no. 3287). All twice natural size. * Biol. Centr. Amer, Bot. 1: 275. 1880, Contr. Nat. Herb., Vol. 16. PLATE 51. FRUITS OF EIGHT SPECIES OF MEIBOMIA. RAIMONDIA, A NEW GENUS OF ANNONACEAE FROM COLOMBIA. By W. BW. SaFForb. While studying the Annonaceae in the United States National Herbarium the writer came upon a specimen from the interior of Colombia, collected by Prof. Henry Pittier, with peculiarities ren- dering it so distinct from all recognized genera of that family as to require its segregation as a new genus. The flowers and fruit of the plant bear a superficial resemblance to those of chirimoya (Annona cherimola Mill.) and its allies; but the plant is monecious, and the inner petals instead of being minute and scale-like, as in the flowers of the chirimoya, are valvate and triangular, forming a tent-like covering to the sexual organs. The most radical difference, however, which requires this plant to be set apart from all other Annonaceae is in the peculiar form of the sta- mens. In Annona and Rollinia, which have compound fruits (syn- carpia), as well as in our own Asimina and many other genera of the family, the connective of the stamens is more or less expanded into a terminal head or hood-like covering above the two pollen sacs. In Professor Pittier’s plant the two pollen sacs are terminal without the slightest indication of the elongation or expansion of the con- nective. Indeed, the stamens differ so radically from those of typical Annonaceae that one would be inclined to separate the plant from that family were it not that its other characteristics are those of the Annonaceae, namely: Two-ranked, alternate leaves without stipules, 6-petaled flowers with the petals in 2 series, and seeds with copious, ruminate albumen and minute basal embryo. In addition to these general features, the punctate, short-petioled, feather-veined, entire leaves, and the compound fleshy fruit (synearpium) point to its alliance with the custard apples or Annonas. RAIMONDIA gen. nov. Arborescent; leaves deciduous, 2-ranked, entire, minutely punctate; flowers moncecious, nodding, closely crowded on extra-axillary branchlets often oppo- site a leaf or issuing from old bark; sepals 3, valvate, persistent, much smaller 217 218 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. than the petals, broadly ovate or triangular, cohering at the base; petals 6, hypogynous, in 2 series, coriaceous, valvate; outer petals lanceolate, concave at the base, when mature much longer than the inner; inner petals (those only of male flowers observed) ovate forming a pyramidal covering to the andre- cium; receptacle (torus) hemispherical or conoid; stamens closely crowded in a broadly ovoid or conoid mass; filaments short and thick, bearing at their extremity a pair of contiguous pollen sacs, somewhat like those of the genus Annona, but much shorter and without the characteristic annonaceous hood-like or capitate, expanded connective above them; female flower long-peduncled (when mature); carpels humerous, coherent, closely crowded on the torus; ovules solitary; fruit oblong, resembling that of an Annona, formed by the con- solidation of the carpels into a fleshy mass around the elongated torus as an axis; seeds enveloped when fresh by a thin membranous aril; testa thin, glabrous, impressed with shallow pits; albumen ruminate as in other Anno- naceae, with the minute embryo embedded in its base. Type species, Raimondia monoica. The genus is named in honor of the eminent geographer and naturalist, Prof. Antonio Raimondi, in recognition of his valuable scientific work in many fields and in grateful acknowledgment of his kindly assistance to a young botanist studying the vegetation of the shores and mountains of his adopted country.2 Raimondia monoica sp. nov. Prates 52, 53. A small tree with the younger parts densely ferrugineous-tomentose: branches at length glabrate, terete, grayish brown; leaf blades obovate to oblanceolate, 18 to 23 em. long and 5 to 12.5 em. broad, membranaceous, usually acute at the base (sometimes somewhat rounded), abruptly acuminate, feather-veined, undu- late, at first densely ferrugineous-tomentose on both sides, at length sparsely “$0 except along the prominent midrib and lateral veins beneath, these per- sistently ferrugineous-tomentose ; petioles with a similar indument, 10 to 15 mm. long, with a longitudinal sroove above, this a continuation of the impressed channel along the midrib; inflorescence densely ferrugineous-tomentose, con- sisting of several flowers closely crowded on short extra-axillary branchlets, these often issuing from the old wood or from a point opposite a leaf; peduncles 5 to 15 mm. long, densely ferrugineous-tomentose with a broad, clasping, ovate, acuminate bracteole below the middle and one at the base; flowers monecious, the pistillate flower issuing from the base of the flowering branchlet (in the specimens examined) and several staminate flowers occupying the remainder; calyx divisions broadly ovate or triangular, abruptly acuminate, 2.5 mm. long and 2.5 mm. broad at the base, ferrugineous-pubescent ; outer petals valvate, lanceolate, rounded at the apex, 15 to 20 mm. long and 7 to 8 mm. broad, with *Raimondi, Antonio. Born at Milan, 1825; died at Lima, Peru, December, 1890. An eminent geographer and naturalist. He went to Peru in 1850, and spent twenty years in traveling and collecting material for his great work on the geography and natural history of the Republic. This was to have been printed at the expense of the nation, and three preliminary volumes appeared (1874, 1876, and 1880). The edition of the fourth volume was destroyed by the Chileans in 1881, and after the war the publication was interrupted; but the materials collected by Raimondi, including his valuable herbarium, are preserved by the Peruvian Geographical Society. Included in his great work, #1 Peru, are accounts of the vegetation of various parts of the Republic. He also pub- lished Elementos de la Botanica for the use of schools (Lima, 1857), and during the latter part of his life he was professor of botany and zoology at Lima. See Amat di San Filippo, Stud. Biogr. Viagg. Ital. p. 597, 1882. Contr. Nat. Herb., Vol. 16. PLATE 52, RAIMONDIA MONOICA SAFFORD. PLaTe 53. Contr. Nat. Herb., Vol 16. a7m Seon \, Ser nay sta Serene i Syren ame mo FP Sa ML TSS EE RAIMONDIA MONOICA SAFFORD, SAFFORD—RAIMONDIA, A NEW GENUS OF ANNONACEAE. . 219 a ferrugineous, silky pubescence on the outside; inner petals triangular, con- cave, rounded at the apex, 7 to 8 mm. high and 6 to 7 mm. broad at the base, forming a cone-like covering over the sexual organs; receptacle hemispherical or conoid in the male flower, in the female flower elongating at length into a linear axis, the carpels radiating around this at right angles; male flowers without vestiges of carpels; stamens very numerous, the pollen sacs terminal contiguous, oval, dehiscing on the back by a median longitudinal slit; fila- ment stout and fleshy; female fiowers with very numerous one-ovuled carpels closely crowded and cohering, forming an oblong gyneecium, this devel- oping into a glabrous, thin-skinned compound fruit about 10 cm. long and 5 cm. in diameter, shaped somewhat like an ear of maize, depressed at the base, and rounded at the apex, borne on a peduncle 2 to 2.5 em. long; seeds oblong- obovate, flattened laterally and usually obliquely truncate at the apex, sharp- edged, enveloped when fresh in a thin membranous aril as in the genus Annona and surrounded by scant pulp; testa chestnut-colored, thin and brittle, with the surface glabrous and shining but punctate with shallow pits, rough on the inner surface to conform with the grooves of the endosperm, this ruminate as in other Annonaceae and with the minute embryo embedded in its base. The staminate flowers shrivel up and fall off after having performed their function, and the peduncles of the pistillate flower elongate and thicken as the fruit matures. The large obovate, membranaceous, undulate leaves with their short petioles, somewhat resemble those of Annona purpurea. The branches, however, are not conspicuously covered with lenticels, as in many Annonaceae, though these are present in the grayish brown bark of the older branches. Type in the United States National Herbarium, nos. 5381655 and 531656, col- lected from the same tree, at the Alto de Primicias, near Jambalé, Rio Palo pasin, Tierra Adentro, Cordillera Central of Colombia, altitude 2,600 meters, latitude about 2° 25’ north, February 5, 1906, by Prof. Henry Pittier (no. 1456). Only a single tree was observed. EXPLANATION OF PLATES 52, 53.—Plate 52, leaf and fruit. Reproduced from a field photograph taken by Cc. B. Doyle. Natural size. Plate 53, details of inflorescence and fruit. a, Male flower from which one outer petal has been removed, showing the three inner vaivate petals covering the andreecium: b, male flower from which the calyx, outer petals, and one inner petal have been removed, showing the andrecium; c, stamen composed of thick fleshy filament and a pair of pollen sacs, seen from the outside or back; d, in- florescence from the base of which a female flower has heen broken; e, young fruit with persistent calyx; f, mature fruit: cross-section showing the seeds inclosed in their mem- pranous arils; g, seeds with aril removed showing the glossy, glabrous testa. a, b, Scale about 2; ¢, scale 20; d, e, f, g, natural size. Drawings by J. M. Shall. FOUR NEW SPECIES OF GOLDENROD FROM THE EAST- ERN UNITED STATES. By E. S. STEELE. The Solidagos here described are all kindred plants, but the first seems to connect more with the American virgaurea types and the other three rather with the speciosa group. The latter would all probably be referred to S. rigidiuscula, if to any recognized species. Consulting on the one hand the original description’ and on the other the material in the National Herbarium, by 8S. rigidiuscula, we should understand a plant 50 to rarely 90 cm. high with the leaves thick, crowded throughout, none of them much’ enlarged, the inflor- escence (at least normally) narrow and dense, the heads distinctly smaller than those of S. speciosa. This would admit several Northern specimens with somewhat thinner leaves and a much less xerophytic aspect, which may eventually be regarded as distinct, but would exclude forms from east of Lake Michigan with the lower leaves more separated and enlarged. It would also exclude S. venulosa Greene,’ a plant ranging from Texas to Kansas, sharply distinguished by its larger heads. The three new species all have the heads rather small, but differ in inflorescence and foliage.* Solidago castrensis Steele, sp. nov. Stem erect, 60 to probably 75 em, high, rather slender, very smooth. glabrous nearly or quite to the inflorescence, upwards increasingly though moderately hispidulous; foliage sparse, the lower internodes 3 to 5 em. long, those above gradually somewhat shortened; about five of the lower leaves clearly petiolate. the longest petioles 5 em. long, somewhat margined but involute and appearing slender, a few of the lowest with vestiges of a hispid ciliation near the insertion ; largest leaves 12 to 14 cm. long, their blades narrowly elliptic-obovate, 25 to 35 mm. wide, acute or often apiculate, at length cuneately narrowed into the petiole, serrate for at least the upper half with strongly mucronate teeth; remaining leaves subpetiolate or at last merely attenuate at the base, more finely serrate or sometimes almost entire; all the leaves moderately thick, coriaceous, the slightly upturned margin obscurely roughened, otherwise smooth and glabrous, *Torr. & Gray, Fl. N. Amer. 2: 205. 1842. * Pittonia 5: 138. 1903. ®’The note may be here appended that the plant intended by me in Contri- butions, volume 18, page 371, under the name Solidago missouriensis was S. glaberrima Mart. 221 222 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. with two or three pairs of fairly distinct pinnate veins, the reticulation obvious but not obtrusive; bracts leaf-like and exceeding the clusters for more than half the length of the inflorescence, this loose at the base, more compact above, cylin- draceous, the racemes usually only 2 em, long, but sometimes more developed, slightly if at all compounded, rather acutely ascending; heads narrowly cam- panulate, 5.5 to 6.5 mm. deep; longer tegules? linear to linear-oblong, obtuse, especially in broader states, the tips sometimes wider than the body, sometimes with an incipient isthmus, an oval, herbaceous spot on the back extending as a line far downward; disk flowers 8 to 10; ray flowers 6 to 8; blade of rays oval, 2 mm. long, 1 to 1.2 mm. wide; achenes small, glabrous. Type in the U. S. National Herbarium, no. 670444, collected at Camp Douglas, Juneau County, Wisconsin, September 9, 1890, by Dr. E. A. Mearns (no. 96). Besides the type specimen there are from the same collector and the same locality another sheet of the same number, but dated September 13, and two others numbered 98 dated September 9. There are two sheets of Mearns 96 which belong apparently to two other species, The specimens more resemble 8. erecta than does any other Western material seen, having not only a cylindraceous inflorescence but a similar distribution of the foliage. I think, however, that the species really connects most directly with 8. sciaphila? and 8. gillmani,; a view countenanced by the location; in any case it is not remote from these. It differs from 8S. erecta in its much Jower stature, its less spreading branches, the more petiolate tendency of the lower leaves, and its smaller heads and narrower rays and tegules. It differs from S. sciaphila in its much narrower and differently shaped leaves, these also some- what thicker and firmer, and in its somewhat smaller heads with less her- baceous involucre. It is a much less heavy plant than 8S. gillmani and erect in habit, with far smaller heads. Solidago chandonnetii Steele, sp. nov. Stem 50 to 75 cm. high, of medium thickness, somewhat decumbent at the base and moderately arcuate, also irregularly somewhat flexuous, glabrous well into the inflorescence, the summit and the distal part of the branches thinly clothed with short, stout, ascending hairs; lower internodes little longer than the upper, the leaves rather numerous but not crowded; foliage habit unequally fusiform, the leaves longest about one-third the distance from base to inflores- cence; none of the leaves much lengthened, the largest about 12 cm. long, the lower all petiolate, the upper subpetiolate or at last merely with an attenuate base; blades of the lower leaves oblong, at least a few of them ovately nar- rowed or even rounder at the apex, those of the middle leaves obovate-lanceolate, acute or nearly so, the remaining leaves reducing to oblanceolate, continuing 3 cm. long up to the inflorescence; all the leaves of medium thickness, firmly coriaceous, roughened on the slightly upturned margin, otherwise smooth and glabrous, pinnately veined but the veins conspicuous only in the lower leaves, rather densely reticulate all over, the lines somewhat more prominent beneath ; inflorescence paniculate, ovoid, in weaker specimens more narrowly so, rather dense, the racemes contiguous or a little separated, the upper racemes simple, the lower moderately compounded, the heads humerous; heads narrowly cam- panulate, only 4 to 5 mm. deep; medial tegules oblong, the inner linear-oblong, both varying somewhat in width, ovately tipped or more rounded; all firm in texture, slightly herbaceous in the middle as indicated by a brown line, some- what glutinous, minutely double-keeled, the ridges tending to form a loop above, but this mostly obscure; disk flowers about 6 to 5, ray flowers about 5 to 7. ‘ Involucre bracts. * Contr. Nat. Herb. 18: 371. 1911. * Op. cit. 367. STEELE—GOLDENRODS FROM THE EASTERN UNITED STATES. 223 Type in the U. 8S. National Herbarium, no. 691211, collected at Perham, Otter- tail County, Minnesota, August 6, 1912, by Rey. Z. L. Chandonnet. The National Herbarium has a second specimen of the same collection and another from the same locality, August 10, 1911, Chandonnet 3789; also (later received) from the same collector, unnumbered, two specimens from the Otter- tail River, August 13, 1912; three from Perham, August 22, 1912; and two from Luce, Ottertail County, August 23, 1912. There is besides these also a sheet with two good plants from Zumbrota, Goodhue County, Minnesota, August, 1892, C. A. Ballard. The habitat is indicated as sometimes dry, sandy soil, sometimes prairie. This species differs from 8. rigidiuscula in its sparser, biserial foliage and its ovoid panicle. It has in its foliage a distinct look toward 8. fisheri, but it is a lower and relatively stouter plant with a larger and more dense inflorescence, smaller heads, and heavier involucres. The National Herbarium is indebted to Father Chandonnet for other im- portant communications. Solidago fisheri Steele, sp. nov. Stem 80 to 90 em. high, reducible to 60 cm., slender, smooth nearly or quite to the inflorescence, there thinly or distally more thickly, hispidulous; lower and upper stem leaves well differentiated; four or five lower internodes 3 to 5d em. long, the corresponding leaves (excluding imperfect ones at base) well petioled, the largest 11 to 17 cm, long, the blades oblong or, especially below, narrowly obovate, 20 to 42 mm. wide, obtuse or apiculate, rather finely feather- veined, reticulate in the intervals, the margin entire or commonly with a few low teeth or crenations toward the summit, the petioles always margined, rather variable in length and breadth; remaining leaves subpetiolate nearly through- out, gradually reduced in length and breadth, entire; all the leaves slightly thick and somewhat coriaceous, rather firm, smooth except the margin, appar- ently rather green when fresh; inflorescence paniculate, narrowly ovoid or cylindraceous, 15 to 21 cm. long, sometimes reduced, rather loose, the slender branches placed at an angle of about 45 degrees, only a few of the longest racemes at all compounded, the bracts of the inflorescence linear to filiform, the pedicels slender, the longer 3 to 5 mm. long; heads narrowly campanulate, 5 to 6 mm. long; lower tegules oblong, the following linear-oblong, both ovate at the tip, rather thick and firm; disk flowers 5 to 7; ray flowers 5 to 9; rays oblong-oblanceolate; achenes very small, glabrous. Type in the U. S. National Herbarium, no. 619505, collected at Michigan City, Laporte County, Indiana, August 12, 1909, by George L. Fisher (no. 107). On sand hills. The National Herbarium has three other specimens from the same source, and a smal] specimen almost surely the same from sandy bluffs at Kilbourn, Wisconsin (August 26, 1909, Steele 40). The type sheet bears two large indi- viduals. Solidago jejunifolia Steele, sp. nov. Stem about 77 cm. high, light colored, striate-angled, smooth to the inflor- escence, the summit with the branches thinly clothed with very short, coarse, ascending hairs; internodes 2 to 3 cm, long for nearly half the leafy segment, the upper moderately shorter, a few reduced to 1 cm.; longest leaves about the fourth to the sixth from the base, 7 to 8 cm. long, somewhat petiolate, the blade oblanceolate or lanceolate, 9 mm. wide, acute or nearly so, the petiole flat and rather broad, the leaves below shorter, a little broader and doubtless more blunt, the leaves above linear-oblanceolate, ovately acute at the tip, short- attenyate at the base; all the leaves rather thick, firmly coriaceous, hispidulous- 61364° —13——3 224 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, ciliate, otherwise smooth and glabrous, light colored; inflorescence about 20 cm. long, rather even in width, after pressing, 5 or 6 cm. wide, loose below, denser above, the branches slender, strongly ascending, the racemes with 7 to 12 heads, a few of the middie ones very slightly compounded, the peduncles slender, 2 to 4 mm, long; heads campanulate, 5 mm. long; involucre scarcely at all her- baceous, the tegules double-keeled but the lines not obviously looped above, the middle tegules oblong, ovately tipped; flowers about 17 to 20, of these 5 to 8 radiate, the rays oblanceolate-oblong, 2.5 mm. long; achenes glabrous. Type in the U. 8. National Herbarium, no. 619303, collected on Indian River, Cheboygan County, Michigan, August 8, 1890, by C. F. Wheeler. . Allied to 8. fisheri, but differing conspicuously in its greatly reduced foliage and in its cylindraceous instead of ovoid inflorescence; also substantially in its lower stature and relatively stouter habit and in the slightly shorter but fuller heads. There is only one specimen, but the essential characters are entirely clear. THREE NEW GENERA OF STILT PALMS (IRIARTEACKAE) FROM COLOMBIA, WITH A SYNOPTICAL REVIEW OF THE FAMILY. By O. F. Cook Aanp C. B. DoYLe. FAMILY CHARACTERS. The palms of the strictly American family Iriarteaceae are readily distinguished in nature by the fact that the trunk does not reach the ground but is supported by a cluster of large, stilt-like, aerial roots. In the smooth trunk and the long sheathing bases of the leaves the stilt palms have a superficial resemblance to the royal palms (Acris- taceae), but the floral and foliar characters show altogether different lines of specialization. The leaves of the Iriarteaceae differ from those of any other pin- nate palms in the strong development and radial position of the veins. Even when the segments are narrow and parallel so that the clusters closely resemble the pinne of other palms, there is no specialization of a definite midvein larger than the others. This lack of specialization may be considered an explanation of the peculiar irregular or premorse margin of the leaves, a feature that is par- alleled, among American palms, only in certain genera of Co- caceae, such as Tilmia and Curima. SURVEY OF THE GENERA. The classification of Triartea and the related genera has received very diverse treatment at the hands of Martius, Karsten, Wendland, and Drude. The very incomplete original description of I[riartea by Ruiz and Pavon was supplemented by Martius from his knowledge of the Brazilian species. Karsten in establishing two new Colom- bian genera, Deckeria' and Socratea,? accepted the characters as- cribed to Iriartea by Martius and thus invited criticism by Wendland for not considering the true type of the genus, the Peruvian /. deltoidea. * Linnaea 28: 258. 1856. . ? Op. cit. 28: 263. 1856. we on 226 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Finding in the female flowers of J. deltoidea the series of rudi- mentary stamens which Karsten had considered distinctive of his genus Deckeria, Wendland reunited this to Iriartea, though he ac- cepted Karsten’s second genus Socratea as well founded. Three new genera, Triartella,' Catoblastus,? and Dictyocaryum, were. pro- posed by Wendland, who also transferred to this group the genus Wettinia, previously established by Poeppig and Endlicher, but not at first recognized as a palm because of its peculiar, somewhat cycad- like inflorescence. Wendland’s association of Karsten’s Deckeria with Iriartea was later found by Drude * to have been a mistake, for the type species of Iriartea proved to have the embryo basal instead of lateral, as Wendland had supposed, and to have the first leaves divided, as well as the terminal segments of the adult leaves, which are united in Deckeria. The spadix of Deckeria has thickened branches, with the flowers inserted in pits. Thus Karsten’s genus Deckeria must evidently be retained, making seven genera which Wendland would probably have recognized if he had been acquainted with the fruits of the type species of Iriartea. In distinguishing these genera, Wendland relied largely on the fruits, but Drude has supplied ad- ditional characters by which some of the groups can be more readily separated.‘ The seven genera could be arranged in two series with reference to the distribution of the flowers. One series is characterized by the presence of functional flowers of both sexes in the same inflorescence, the other by having flowers of the two sexes in separate inflorescences, though still on the same individual. The first series, with the flowers of the two sexes together, is composed of the five genera Triartea, Dictyocaryum, Deckeria, Socratea, and Iriartella, leaving Catoblastus and Wettinia for the other series with the flowers of the two sexes separated. In Drude’s classification of the palms, in Engler and Prantl’s Natiirlichen Pflanzenfamilien,® the genera Catoblastus and Wettinia are associated with Iriartea. They differ very distinctly in the char- acters of the flowers, having the sexes separated in different in- florescences and the carpels unequally developed, only one producing a normal ovule. Of the three new forms described in the present paper, two may be considered to be intermediate between Iriarten and Wettinia, in that male flowers are still present in rudimentary form on the pistillate inflorescence. The first of these is more 'Bonplandia 8: 103. 1860. “See Mart. Fl. Bras. 3°: 585. 1882. 2See p. 281, below. ° 2°:60, 61. 1889. 3 Bonplandia 8:106. 1860, COOK AND DOYLE-—NEW GENERA OF PALMS FROM COLOMBIA. 227 closely associated with Iriartea in having the carpels of the young flowers equally developed, but the petals are narrow and separate, while those of Iriartea are broadly imbricate. The embryo is also basal instead of being lateral or apical as in most of the species referred to Iriartea. The second of the new types, also with rudimentary male flowers on the pistillate inflorescence, has the carpels extremely unequal, even more so than in Catoblastus, and entirely lacks a style. The sharply three-lobed, star-like stigma is seated at the base of the long, subcylindrical fertile carpel. The albumen of the seed differs from that of Catoblastus in being quite uniform, instead of ruminate. The third of the new forms has the sexes entirely separated as in Wettinia and the pistillate flowers and fruits closely crowded to- gether as in that genus, but the inflorescence is branched instead of simple, and the long, narrow petals are separate instead of over- lapping at the base as in Wettinia. The fruits of the last species, though narrowed and angled at the base by mutual pressure, remain shorter than those of the other two species, which are scattered upen the longer branches and are of a regular long-oval form. In all three species the surface is rough and hairy and the outer layer or pericarp has a loose corky texture not a little suggestive of that of the fruits of Manicaria and Phytelephas. TRIBAL DIVISION. Taking these additions into account, a division of the family into three tribes seems desirable. Though Catoblastus and Wettinia seem to agree in some of the characters that separate them from Iriartea and its immediate relatives, they are very unlike in other important respects. Moreover, Catoblastus and Wettinia no longer appear as isolated genera, but rather as members of two series of related genera, like those that cluster about Triartea. SYNOPTICAL KEY TO THE TRIBES, Flowers of both sexes borne on the same spadix; a single inflo- rescence in the axil of each leaf..........-..........24.- IRIARTEAE. Flowers of the two sexes in separate inflorescences; several inflorescences from each axil. Flowers and fruits scattered on the numerous slender branches of the inflorescence; styles short or wanting. CATOBLASTEAE. Flowers and fruits densely crowded on the thickened simple or few-branched spadix; styles long and slender.... WETTINIEAE. 228 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. DESCRIPTIONS OF THE COLOMBIAN GENERA AND SPECIES. CATOBLASTEAE. SYNOPTICAL KEY TO THE GENERA. Pistillate flowers with three equal carpels at the time of flowering .......---2...-- 0... e eee eee eee eee ACROSTIGMA (p. 228). Pistillate flowers with one greatly enlarged fertile carpel at the time of flowering. Petals not imbricate; stigma sessile at the base of the long, cylindrical fertile carpel................ CaTOSTIGMA (p. 230). Petals imbricate at base; stigma rostrate or borne on a columnar Style......-..................2.... CATOBLASTUS (p. 231). ACROSTIGMA gen. nov. Trunk solitary, rather short (6 meters long) not so thick below (10 em.) as above (16 cm.) ; smooth, distinctly ringed; internodes short (10 em.) Leaves with rather short sheaths (120 cm. long) ; pinne triangular-lanceolate, not inserted on a fleshy pulvinus; petiole short (16 to 18 cm. long); segments of unequal length, united; both surfaces smooth; terminal pinns broad, united to near the end. Spathes 5 or 6, two or three of them small, basal, three large. Spadix compound, divided near the middle into several (about 5) long, taper- ing, flexuous branches; female inflorescence bearing irregularly scattered flowers each subtended by 2 minute rudiments of male flowers; female flowers with sepals rounded-triangular, broader than long, slightly connate at base; petals triangular, nearly as broad as long, less than twice as long as the sepals, dis- tinctly separated at base, soft and fleshy in texture, not ribbed on the back; staminodes 6, minute, delicate, of soft texture, less than one-fourth the length of the sepals; pistil triangular with three equal lobes, not exceeding the petals; stigmas completely sessile, forming a minute 3-pointed slit in the apex of the pistil. Fruit elliptical, slightly compressed and slightly angled on one side above the stigma, one and one-half times as long as broad. Stigmas persistent at base, close to the persistent calyx; sepals united at base, one much larger than the other two; mesocarp of rather loose, corky, granular texture, composed of coarse cellular tissue and coarse, irregular fibres, often spine-like; inner surface of mesocarp closely and ebscurely wrinkled without evident indication of a fibrous network as in Catostigma; numerous bundles of raphids imbedded in the mesocarp immediately around the inner surface; endocarp rather thick and corky, covered with a very close network of rather coarse, strongly flexuous fibres. Seed compressed-oval, smooth or very faintly impressed under the fibres of the endocarp; albumen uniform; embryo basal, extremely small, not covered by the albumen, indicated by a minute rounded prominence on the outer wall of the very small cavity. . Type species, Acrostigma equale. Distinguished from Catoblastus Wendl. in having the pistil formed of three equal carpels with minute sessile stigmas, and by the presence of rudimentary male flowers in the female inflorescences. Acrostigma equale sp. nov. PLATE 54 (A), 55, 56 (A), 57, 5s. Trunk 6 meters long by 10 cm. in diameter at 1 meter from base, becoming thicker toward the crown, there reaching » diameter of 16 cm.; internodes 10 cm, or Jess in length. COOK AND DOYLE—NEW GENERA OF PALMS FROM COLOMBIA. 229 Leaves S to 12 in a head, largest in the middle, tapering gradually toward base and tip; leaf bases 120 cm. long by 50 cm. broad at base, splitting into numerous fragile fibers; petiole short, 16 to 18 cm. long by 4 cm. in diameter ; lower surface rounded and covered with dark brownish tomentum; upper sur- face rounded and also tomentose, excepting a central strip 1 cm. wide, light green in color, free from tomentum or scurf and running the entire length of the petiole. Rachis 850 em. long by 4 cm. in diameter at base, gradually tapering toward the tip; lower surface rounded, becoming flat toward the tip, sparsely covered with greenish tomentum; upper surface with a prominently rounded central ridge 6 to 8 mm. wide at base, gradually becoming narrowed to a sharp ridge ; on either side of this median ridge a shallow groove, in this the pinne inserted ; the surfaces of these grooves, as well as of the central ridge, marked with about 12 longitudinal, slightly elevated ridges or striations, these 1 to 2 mm. apart, covered with very loosely adherent, brownish, scurfy material easily detached with the finger; toward the end of the leaf the grooves becoming shallow and disappearing ; below each pinna for a distance of about 2 cm. the rachis entirely free from scurf or tomentum, but the striations still distinct; these naked spots of much lighter color and visible at some distance from the tree; upper central ridge slightly lighter in color than the grooves, toward the end of the leaf taking on the same color as the naked spaces below the pinne; striations and naked spots also disappearing toward the end of the leaf, the last three or four pinnae thus inserted on a comparatively smooth rachis. Pinne 33 or 34 on each side, consisting of 11 to 18 segments; upper surface dark green; lower surface slightly lighter, smooth; midribs very prominent on the under side, raised 1 to 2 mm. above the surface and of much lighter color; segments of pinnze unequal, premorse, the lower segments of each pinna being the longest; lowest pinne 34 cm. long by 7 cm. wide, consisting of 7 or 8 seg- ments 1 em. or less in width, 1.5 cm. wide at point of insertion; middle pinne 105 em. long by 16 em. wide at the broadest part, 5 cm. wide at base, consisting of 15 segments 2 cm. or less in width; terminal pinne 28 to 34 cm. long by 10 to 15 em. broad, becoming 19 to 24 cm. wide at insertions, consisting of from 12 to 16 segments 2 em. or less in width. Inflorescence infrafoliar, 105 cm. long; peduncle 36 cm. long by 6 cm. wide at base, the fruiting portion 10 cm. long by 2 cm. in diameter; branches 5, the longest 59 cm. by 1.5 cm.; peduncle and bases of branches densely hirsute; flowering portion when young with a very sparse covering of long, gland-tipped hairs, these also occurring on the petals and pistils; between the hairs the sur- face appearing naked, but seen to be very minutely papillose under sufficient magnification; flowers inserted in depressions, these continued on each side to form a distinct transverse groove accommodating the rudimentary male flower. Spathes 5, the outer 12 cm. long; inner spathe densely covered, when young, with soft silky hairs 2 mm. or less in length. Fruits 40 to 50 on each branch, ovoid or slightly flattened on one side, 5 cm. long by 3 cm. in diameter or less, covered with closely adherent grayish hairs like those on the branches and peduncle; pericarp corky, 5 mm. thick, tough; kernel 8 em. long by 1.5 cm. in diameter, ovoid or slightly fiattened, the outer surface covered with a web-like coating of closely adherent fibers. Type in the U. 8S. National Herbarium, nos. 690426-690428 (all from one tree), collected in deep, marshy forests about Cordoba, Cauca, Colombia, by Cc. B. Doyle, December, 1905. Native names, “zancona” or “zancuda,” meaning “stilt’’ or “ mosquito ” palm, evidently in allusion to the long aerial roots. Also called ‘“ crespa,” perhaps with reference to the stiff leaves or to the hairy fruits. 230 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. EXPLANATION OF PLATES 54-58.—Plate 54, whole leaf and inflorescence, A, of Aerostigma equale; B, of Wettinella quinaria. From field photographs taken at Cor- doba, Cauca, Colombia, December, 1905. Plate 55, tip of leaf of Acrostigma equate. From a photograph taken in Washington. Reduced. Plate 56, bases of pinns, A, of Acrostigma equale; B, of Catostigma radiatum; C, of Wettinelia quinaria. All natural size. From photographs of dried specimens taken in Washington. Plate 57, young Spathes and female flowers of Acrostigma equale. From a field photograph taken at Cordoba, Cauca, Colombia in December, 1905. Natural size. Plate 58, fruit and female | flowers of Acrostigma equale. From field photographs taken at Cordoba, Cauca, Colom- bia, December, 1905. Natural size. All photographs taken by C. B. Doyle, CATOSTIGMA gen. nov. Trunk solitary, rather short (6 meters), slender (7 to 8 em. thick), taper- ing upward ; surface smooth, distinctly ringed; internodes rather short (16 cm.), becoming shorter above (10 ¢m.). Leaves with rather short sheaths (108 cm.), densely brown-tomentose without ; petiole rather long (34 cm.), cylindrical, smooth, and naked; pinne triangular, the lower unequally divided into 5 to 7 obliquely diverging segments, each segment inserted on a fleshy pulvinus; tips of pinne irregularly notched; both surfaces smooth; terminal pinne rather narrow, completely united, the rachis continued to margin. Inflorescence with 5 spathes, 3 short basal ones and 2 complete ones; also two or three spatheless ring-scars; spadix compound, divided below the middle into a few (about 3) long, tapering, flexuous branches bearing irregu- larly scattered flowers and fruits; surface not hirsute but very minutely granular-hispid ; female flowers inserted in shallow rounded depressions, each subtended by two rudimentary male flowers, one on either side. Female flowers with sepals subtriangular, somewhat broader than long. slightly connate at base; petals triangular, pointed, twice as long as broad, about 4 times as long as the sepals, distinctly separated at base, in texture firmly coriaceous, distinctly 5-costate on the back, the middle rib strongest ; staminodes 6, of firm texture, conic-subulate, about half as long as the sepals, strongly reflexed, opposite and alternate with the petals; pistils distinctly 3-lobed, the fertile carpel conic-cylindric, about 3 times as long as broad, greatly exceeding the others (about 7 times as long), twice as long as the petals, stigmas about twice as long as broad, triangular-conic, sharp-pointed, strongly recurved, strongly rugose-tuberculate; rudimentary male flowers accompanying the female containing a cluster of minute staminodes. Fruits narrowly oval, symmetrical, about two and one-half times as long as broad; surface even, minutely granular-tuberculate, bearing the persistent stigmas at the base, close to the persistent calyx, one lobe of this distinctly larger than the other two: mesocarp of a rather firm corky texture, composed of coarsely cellular material and stout irregular fibers; inner surface of mesocarp showing an open-meshed network of distinct, slender fibers similiar to the fibers of the endocarp, but entirely separate, often with a layer of gelati- , nous material between; endocarp delicately membranous, the fibers very delicate, not very numerous, those of the inner layer parallel at the base on the side opposite the embryo, anastomosing into a fine network on the other side. Seed narrowly oval, its surface smooth and even or with very faint impressions ; albumen uniform except for a median canal and a semielliptic cavity at the base, the cavity as broad as long, covered by a rather thick lid of albumen to which the disk-like or top-shaped embryo is attached. Type species, Catostigma radiatum. Distinguished from Catoblastus Wendl. by the narrow spathes, slender in- florescences, and large, strongly recurved, sessile stigmas, and by the presence of rudimentary male flowers on the female inflorescences. WV "A1A0G Y AOOD 31vNO|A YWOILSONOYW ‘d “ATA0Q XY HOOD VIEVNIND VITIANILLE MA Contr. Nat. Herb., Vol. 16. PLATE 54 (corrected), ine Contr. Nat. Herb., Vo. 1¢ PLaTe 55. ACROSTIGMA EQUALE Cook & DOYLE. Contr. Nat. Herb., Vo!. 16. PLATE 56. C. WETTINELLA QUINARIA COOK & DOoYLe. PLATE 57. it. Herb., Vol, > < Contr. ACROSTIGMA EQUALE COOK & DOYLE. Contr. Nat. Herb., Vol. 16 PLATE 58, ACROSTIGMA EQUALE CooK & DOYLE. Contr. Nat. Herb., Vol. 16. PLATE 59. A. FRUITS OF CATOSTIGMA RADIATUM Cook & DOYLE. B. FRuItS OF WETTINELLA QUINARIA Cook & DOYLE. PLATE 60. CATOSTIGMA RADIATUM CooK & DoYLE. COOK AND DOYLE—NEW GENERA OF PALMS FROM COLOMBIA. 231 Catostigma radiatum sp. nov. Priates 56 (B), facing p. 230, 59 (A), 60. Trunk 6 meters long, 5 meters to first inflorescence, the diameter 7 to 8 cm. at 1 meter from base, 6.5 cm. at base of first leaf; wood very hard and fibrous. Leaves 10 to 12 in a head, largest in the middle, tapering gradually toward base and tip; leaf bases 108 cm. long, splitting into numerous. fragile fibers: lower surface densely brown tomentose for about one-half its length; petiole 34 em, long by 2.5 cm. in diameter, cylindrical ; surface densely tomentose m young leaves, smooth in adult specimens; rachis 272 cm. long by 2.5 em. thick at base, tapering gradually toward the tip; lower surface rounded, becoming flat; upper surface for 50 cm. outward from base with a prominent, rounded, central ridge becoming more prominent; upper surface also marked with minute grooves; toward the tip the latter disappearing and becoming flat surfaces, the central ridge becoming more prominent; upper surface also marked with minute longitudinal raised lines running the entire length of the leaf, less prominent toward the tip. Pinne 57 to 62 on each side, simple or split into 2 to 7 divisions and consist- ing of from 1 to 10 segments ranging in width from 1 to 3 cm., the divisions standing at different angles to the rachis; tips of pinnze notched, due to unequal termination of segments, the lower segments generally the longest; upper sur- face of pinne dark green, the lower surface lighter; at the point of insertion of each pinna a fleshy cushion or pulvinus; just below the insertion of each pinna the rachis naked and lighter in color for 2 cm. or less; for 2 to 3 cm. below the insertion of each group of pinne the lower surface elad with a dense tomentum easily removed with the fingers; lowest pinns 29 em. long by 6 to 7 em. wide, consisting of 6 segments 1 cm. or less in width; lower and terminal pinne inserted singly: middle pinnz 33 to 56 cm. long by 4.5 to 138 cm. wide, consisting of 2 to 9 segments 2 cm, or less in width; terminal pinne 17 cm. long by 7 to 9 em. wide, consisting of 8 or 9 segments 2 cm, or less in width. Inflorescences infrafoliar, as many as five maturing fruit at one time; total length 105 cm.; peduncle to first branch 30 cm. long by 2.5 cm. in diameter at base and 2 em. at first branch; fruiting portion 4 cm. long by 1.5 em. in diameter; both peduncle and fruiting portion light green, smooth; branches 3 or 4, 71 em. or less in length by 2 em. in diameter. Spathes 5, the lowest 4 cm. long by 4 cm. in width at base; inner spathe 26 em. long, smooth; usually three inflorescences, two male and one female, ap- pearing in a group at the same node. Fruits densely clustered along the branches, oval 3.5 cm. long, 1.8 cm. in diameter; outer husk corky, brittle, 2 to 3 mm. thick; seed broadest below the middle, covered with a closely adherent, fibrous, fragile epidermis; texture bony; embryo basal. Type in the U. S. National Herbarium nos. 690429-690432 (all from one tree), collected along bank of stream near Cordoba, Cauca, Colombia, by C. B. Doyle, December, 1905. Native names, “ gualte,”’ “ sape.” EXPLANATION OF PLATES 59, 60.—Plate 59, A, fruits of Catostigma radiatum Cook & Doyle; B, of Wettinella quinaria Cook & Doyle. Plate 60, young spathe and tip of leaf of same species. From ficld photographs taken at Cordoba, Cauca, Colombia in 1905, the Jatter in December. Al) natural size. CATOBLASTUS Wendl. Catoblastus Wendl. Bonplandia 8: 104. 1860. Trunks slender, cespitose, supported on a cluster of short superficial roots. Leaves 3 to 4 feet long, with 24 pairs of simple pinnie. Inflorescences in verticillate clusters, 10 to 12 inches long. 232 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Spathes numerous, the outer short. Male flowers in pairs; sepals valvate, pistillodes small. Female flowers with separate sepals and petals; staminodes none ( Wend- land) ; present in the type species (Klotzsch) ; ovary trilocular. Fruits oblong-ovate; stigma scar close to the base; epicarp subrugose or puberulous; mesocarp grumous; endocarp membranous. Seeds oblong-ovate with immersed bundles rising from the base on the ventral side, running close together to the apex, then anastomosing and coming together again at the embryo; albumen ruminate (Wendland) or uniform (Klotzseh) ; embryo basal, erect. Type species, Catoblastus praemorsus. Catoblastus praemorsus ( Willd.) Wendl. Bonplandia 8: 104. 1860. PLATE 61. Oreodoxa praemorsa Willd. Mém. Acad. Sci. Berlin 1804: 26. 1809. Triartea praemorsa Klotzsch, Linnaea 20: 448, 1847. Trunks erect, cylindrical, very smooth, 12 to 15 meters high; stoloniferous. Leaves pinnate, very long, the pinnee broadly cuneiform, narrowed at base, unequally premorse-dentate at apex, alternate; color dark green. Fruit ashy gray, ovate; seed ovate, of the size of a pigeon’s egg; endosperm brown, marbled with numerous veins. Forests of the high mountain chain of Buenavista, Province of Caracas, Venezuela, growing with Orcodora acuminata, but much more rare. The natives call this palm “ pyra,” a name they apply to all the palms which have terminal buds that can be used for food. The above data are drawn from Willdenow’s original description. The species was described at greater length by Klotzsch, but it does not appear certain that he had the same palm; at least the particulars differ considerably, as may be judged from the following transcription of additional or divergent facts: Trunks about 15 from the same root, 30 to 50 ft. long, 38 inches thick, borne on a cluster of warty brown roots about as thick as the little finger. Leaves 3 to 4 feet long; rachis compressed, margined above, and with the margin pubescent; pinnie 5 to 24 pairs, 10 to 12 inches long, 14 to 24 inches broad, alternate, herbaceous, pale green, distant, nearly smooth, irregularly rhomboidal-cuneate, united at the base, repand-erose toward the apex, below with 6 to 8 parallel, prominent veins; terminal pinna flabelliform, at apex dentate-truncate, sometimes bifid, short-cuneate at base, 7 to 11 inches long, 6 to 9 inches broad. Spadices numerous (10 to 20), cylindric-fusiform, 10 to 12 inches long, 24 inches broad, thickened near the apex and obtusely pointed, narrowed toward the base. Spathes coriaceous, the interior ones complete, closed, at length opening on the ventral side, the exterior shorter, incomplete, imbricate, tubulous, open at apex. Flowers sessile, without bracts, yellowish, monmcious in separate, simply branched spathes; male flowers in pairs; calyx of three triangular, fleshy, minute sepals, valvate in the bud; petals three, acute-triangular, fleshy, 8 times as long as the sepals; stamens 9 to 15, hypogynous, the filament free, filiform, the anthers linear, attached at the base, opening by two longitudinal slits; pollen grains round, echinate; pistillode very small; female flowers soli- tary, sepals and petals triangular, fleshy, the petals twice as long; staminodes few, rudimentary; ovary with 3 carpels, the ovules basal, solitary, rarely two, anatropous; stigmas three, sessile, beak-like, connivent. Fruit ovate, somewhat wrinkled, green, at length blackish, of the size of a pigeon’s egg, 1-celled and 1-seeded; albumen uniform, horny; embryo basal. Contr, Nat. Herb., Vol. 16. PLATE 61. XXX CATOBLASTUS PUBESCENS (KARST.) WENDL. AND C. PRAEMORSUS (WILLD.) WENDL. Contr, Nat. Herb., Vol. 16. PLATE 62, Wy Li “8 / 4d FLOWERS AND FRU!IT OF CATOBLASTUS PUBESCENS (KARST.) WENDL. COOK AND DOYLE—-NEW GENERA OF PALMS FROM COLOMBIA. 2338 Flowers from June to October. Puerto Cabello, Colonia Tovar, Colombia. Moritz no. 914. EXPLANATION OF PLATES 61, 62.—Plate 61, Catoblastus pubescens (Karst.) Wendl. ; at the right a seedling of two years. Fig. 1, a pinna; fig. 2, an inflorescence; fig. 3, a male flower ; fig. 4,an anther; fig.5,a pollen cell, Figs. 14, 15, fruit of Catoblastus praemorsus (Willd.) Wendl. Reduced. Reproduced from Karst. Fl. Columb. 1: pl. 81. 1858-1861. Plate 62, flowers and fruit from the same plate at natural size. Fig. 6, tip of male in- florescence ; fig. 7, tip of female inflorescence; fig. 8, section of fruit; figs. 9 and 10, views of the seed; fig. 11, female flower; fig. 12, same after removal of calyx and corolla segments; fig. 15, section of pistil. As this species is the type of the genus Catoblastus, its characters are of special interest, but some of the most important are still in doubt. Klotzsch, and more recently Jahn, have described the carpels as distinct and with separate stigmas, and neither of these writers gives any intimation that the carpels are unequal at the time of flowering. Karsten’s Iriartca pubescens,‘ which Wendland placed as a second species of Catoblastus, has the stigmas united into a short, cylindrical style, not half as long as the very large fertile carpel. The figure of Catoblastus pubescens given by Drude in Engler and Prantl’s Pflanzenfamilien shows a long, columnar style rising above the fertile carpel and is likely to mislead regarding the principal generic character. Karsten says that the staminodia are wanting or very small and hidden under the carpels, whereas Drude’s figure shows large staminodia with anthers. To avoid further confusion from this erro- neous figure in so prominent a work of ref- FiG:41.—Ovary of Catoblastus drudei erence it may be best formally to recognize with fertile and sterile carpel, a the fact that Drude’s Catoblastus is a dif- %taminode below. At the right, same + og + : . in section. After Engler and Prantl. ferent species.” Unfortunately its origin is not indicated unless by the fact that Drude includes Peru in the range of the genus, while Venezuela is omitted. It is also evident that the true characters of Catoblastus are not to be learned from Drude’s account of the genus nor from Karsten’s beautiful figures of C. pubescens. It is quite possible that one or both of these species will eventually be removed from the genus. For the present it may be suflicient to note that C. pubescens has an obvious alliance with our new genus Catostigma. The com- pletely sessile stigmas of Acrostigma and the presence of the rudimentary male flowers on the female inflorescences afford apparent distinctions, but if they do not prove to be adequate, C. pubescens will need to be transferred to Catostigma instead of Catostigma being united with Catoblastus. The true affinities of Catoblastus, as represented by C. praemorsus, may lie with Acrostigma rather than with Catostigma, but the beak-like connivent stig- mas indicated by Klotzsch do not suggest the condition found in Acrostigma. Wendland and Drude both describe the albumen of the seeds of Catoblastus as ruminate, perhaps relying upon Karsten’s drawing, but Klotzsch says that the albumen of praemorsus is uniform and describes the seed as marbled with numerous veins, which, however, may refer to the endocarp (écorce) rather than to the endosperm itself. ?Binnaea 28: 262. 1856. * CATOBLASTUS DRUDEI. Figure 41. Oatoblastus pubescens Drude in Engl. & Prantl, Pflanzenfam. 2*: 61. f. GI, G2. 1887, not Triartea pubescens Karst. 1856, nor Catoblastus pubescens Wendl. 1860. 234 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. There is no reason to doubt that Karsten’s species pubescens is ruminate, but he gives us no indication of the origin of the still more strongly ruminate seed that he figured as representing pracmorsus (see Pl. 59, Fig. 15). It is not to be assumed that Klotzsch would have called such a seed uniform. The most recent account of this species is by Jahn’ who gives additional data from Venezuela, probably based on palms growing near the original habitat of the species. Jahn’s description may be translated as follows: “'Phis species is very abundant throughout the subalpine region (1,500-2,000 meters) of our Caraibe cordilleras where it is given the name ‘ prapa’ by the indigenes. Its smooth cylindrical trunk reaches 6 to 8 meters in length and 10 to 15 em. in diameter. The woody part is very hard, black, and heavy (specific gravity in dry state 1.25 according to Ernst) and incloses a soft central pith. Leaves few, pinnate, 2 meters long and composed of 10 to 20 pinne on either side of the rachis or common petiole with a terminal pinna. The pinne are alternate, herbaceous, pale green, situated some distance from each other, irregularly rhomboidal, toward the tip more or less deeply incised. The 6 to 8 longitudinal veins or nerves run parallel with the pinnie, and are prominent on the under side. Pinnse 20 to 80 cm. long by 6 to 3 cm. broad. The terminal leaf is triangular or in the form of a half-open fan, cuneiform at base and truncate or gnawed at the tip. whence comes the specific name prae- morsus. ‘The aerial roots form a pyramid of radiating pillars that gives strong support to the trunk. They are distinguished by their rough surface filled with spine-like protuberances that protect them from the herb-eating animals, principally the vaquira (Dicotyles) and the tapir (Tapirus) which inhabit the same region. The inflorescences have many spathes (10 to 12), fusiform, almost cylindrical, 20 to 25 cm. long and 5 to 6 cm. thick at the largest point, which is slightly below the tip. Inflorescences springing directly from the nodes of the trunk and not (as in the Iriarteas) from the base of the crown or leaves. Every spadix has 5 or 6 spathes, three interior and complete, closed until the time of flowering, and three exterior and shorter and incomplete; that is to say, always open at the upper end like short tubes. The flowers are seated in depressions, and are yellow and without bracts, the male and female on the same tree, but in different inflorescences. The male flowers have a_ three- parted calyx and a corolla of three triangular, somewhat tleshy petals, and have 9 to 15 stamens with free filaments and with anthers which split with two iongitudinal apertures, this permitting the dispersion of the somewhat rough- ened pollen grains, The female flowers have the same number of sepals and petals as the male, with 4 to 6G staminodes. The ovary is three-celled, with the carpels very distinct, the style very short, and the three stigmas quite large, thick, with the interior somewhat channeled. The fruit is an egg-shaped berry with grumous pericarp, black, with a bluish luster, about the size of a pigeon’s egg, inclosing a single seed. The ‘ prapa’ flowers from August to October, and in the following April or May the fruits are mature.” WETTINIEAE. SYNOPTICAL KEY TO THE GENERA. Inflorescences simple; many (8 to 15) from each node..WETTINIA (p. 235). Inflorescences with 4 or 5 branches; five inflorescences from each node but only the middle one maturing. .\WeTrinELLA (p. 235). ‘Jahn, Alfredo, jr. Las Palmas de la Flora Venezolana 49-50. 1908. Contr. Nat. Herb,, Vol. 16, PLATE 63. A i ju B, WETTINELLA QUINARIA Cook & DOYLE. COOK AND DOYLE—NEW GENERA OF PALMS FROM COLOMBIA. 235 WETTINIA Poepp. & Endl. Wettinia Poepp. & Endl. Nov. Pl. Poepp, 2: 39. pls. 153, 154. 1838; * Wendl, Bonplandia 8: 104-106. 1860. Trunk solitary, 8 to 12 meters high, 15 to 20 cm. thick. Leaves with opposite simple pinn. ‘ Inflorescences simple, in verticillate clusters of 8 to 15, bearing flowers of one sex, but spathes of both sexes intermixed on the same trunk. Spathes five, three short and two much longer and complete. Male flowers with 4 or 5 minute sepals and 3 or 4 petals, valvyate in the bud; stamens 12 to 16; female flowers with 3 sepals and 3 petals; staminodes? ; ovary with 1 carpel and 1 ovule, obverse-pyramidate, villose; style filiform, rising at one side, from the base of the ovary; stigmas three, narrowly lanceo- late. Fruits obverse-pyramidate, flattened at the apex, strigose, hirsute, the endo-- carp delicately parcbment-like. Seed elliptical or obovate, surrounded with bundles rising from the base, run- ving together to the apex, then laxly anastomosing and coming together again at the embryo; albumen solid, uniform; embryo basal, erect. (PLATE 63, A.) Type species, Wettinia augusta Poepp. & Endl. loc. cit.; from the Huallaga Valley in eastern Peru. EXPLANATION OF PLATE 63.—A. Male and female inflorescences of Wettinia augusta Poepp. & Endl. B. Young female inflorescence of Wettinella quinaria Cook & Doyle. A, reduced; B, natural size. , slightly enlarged. obovate or obpyramidal, 5 to 1851-1857, by J. Triana; duplicate in De Candolle Herbarium. ‘ { | “a neath sparsely rufous-pubes- r cent; peduncles solitary, ex- tra-axillary, 17 mm. long in \ fruit, at length glabrate; flow- ter, the component carpels 7mm. long by 4 mm. broad, somewhat compressed, with a smooth hard brown testa DistRiBuTION: Known only from type locality. EXPLANATION OF PLATE 94.—Photograph of type specimen. Contr. Nat. Herb., Vol. 16. PLATE 94, ANNONA CERCOCARPA SAFFORD. Contr. Nat. Herb., Vol. 16, PLATE 95. ANNONA ECHINATA DUNAL. PLATE 96. Contr. Nat. Herb.. Vol, 16. ANNONA ECHINATA DUNAL. SAFFORD—ANNONA SERICEA AND ITS ALLIES. . 273 8. Annona echinata Dunal. PRICKLY ANNONA OF FRENCH GUIANA. Anona echinata Dunal, Monogr. Anon. 68. pl. 4. 1817. Branches divaricate, clothed when young with ferrugineous hairs, at length glabrate, blackish, rugose, and bearing many lenticels; petioles 5 mm. long, deeply grooved above, at first minutely appressed-pubescent, at length glabrescent; blades membra- naceous or subcoriaceous, thicker than those of A. sericea, pellucid-punctate, those on the upper parts of the flowering branches ovate-oblong or ovate-lanceolate, some- what acute or obtusely acuminate at the apex, rounded or cuneate at the base, gla- brous above, the midrib impressed and bordered on each side with numerous very short raised veins at right angles to it, the secondary nerves (8 to 10 on each side) sharply defined, connected by reticulating veins; beneath clothed with fine short dull grayish ferrugineous pubescence or tomentulum and reticulated between the prominent midrib and secondary nerves; lowermost leaves on flowering branches smaller and relatively broader, sometimes obtuse or retuse at the apex as in many other species of the genus; peduncle solitary, 1-flowered, 11 mm. long (in the type specimen), extra-axillary, issuing from the base of a new branchlet and apparently terminal on account of the abortion of the portion of the branchlet beyond it (as in many other Annonaceae), ferrugineous-tomentose or hirtellous and bearing a small tomentose bracteole below the middle; flowers similar in size and shape to those of A. sericea, spheroid in bud, normally 3-petaled, but sometimes in the rainy season (according to Sagot) with 3 additional inner petals alternating with the outer and closing the seams between them; calyx 3-lobed, 5 mm. in diameter, the divisions broadly triangular and obtuse, clothed on the outside like the peduncle with ferru- gineous hairs; petals broadly ovate or suborbicular, obtuse, thick, coriaceous, con- cave, 11 mm. long and 10 mm. broad (in type flower), clothed on the outside with minute ferrugineous pubescence; stamens numerous 2 to 2.5 mm. long, with a short broad filament, linear pollen sacs, and a connective expanded into a swollen head, this minutely papillose or muriculate but devoid of hairs; carpels numerous, united in a conoid gyncecium, the ovaries clothed with appressed ferrugineous hairs; fruit ovoid, small, 24 mm. long by 17 mm. broad (fruit of type possibly immature), bearing numerous recurved protuberances corresponding to the individual carpels, the sur- face clothed with fine appressed ferrugineous pubescence; seeds oblong, 6 mm. long and 3 mm. broad; peduncle at length thickened and woody, sometimes apparently terminal from the abortion of the portion of the branch beyond it. (Piates 95, 96.) Type in the Prodromus Herbarium of De Candolle at Geneva (ex Herb. Lhéritier), collected about 1795 at “‘Cayenne”’ (French Guiana) by J. B. Patris. Disrripution: Guiana and probably Brazil. SPECIMENS EXAMINED: FrENcH Gu1ana—‘‘Cayenne,” Patris, type collection, leaf, stamens, and tip of carpel; Mana, Sagot 6, leaf and stamens, from Kew Herbarium. This species is undoubtedly closely related to A. sericea Dunal, but differs conspicu- ously from that species in the character of the indument of the leaves and the absence of hairs on the swollen terminal head of the connective of the stamens. Its ovoid, echinate fruits resemble miniature soursops (A. muricata L.). The recurved carpel tips are somewhat like those of A. cercocarpa described above, but differ from them in their less length and in their much finer, appressed pubescence, the carpels of A. cercocarpa being prolonged into tail-like appendages covered with relatively coarse, strigose hairs (fig.44). The present species is also sharply distinct from the preceding in the shape and texture of its leaves, as indicated by the accompanying illustrations. EXPLANATION OF PLATES 95, 96.—PI1. 95, photograph of type specimen in De Candolle Prodromus Her- barium, Natural size. Pl. 96, drawing from type material, that of fruit reproduced from original plate; a, petal; b, stamen; c, cross section of fruit; d, tip of mature carpel; ¢, immature carpel bearing style. Figs. a andc, natural size; b and e, scale 20; d, scale 8. 85668°—voL 16, pr 10—13 2 274 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, 9. Annona acuminata Safford, sp. Nov. SMALLER WILD ANNONA OF PANAMA. Anona echinata Hemsl. Biol. Centr, Amer. Bot. 1: 19. 1879, not Dunal, 1817. A small tree 5 to 7 meters high with slender branches roughened by thickly crowded prominent reddish brown lenticels; very young branchlets clothed with minute appressed ferrugineous hairs scarcely visible even with the aid of a lens, very soon glabrate; leaves small, thin, membranaceous, glabrate, pellucid-punctulate (those of flowering branches only observed), 6.5 to 8 cm. long and 1.8 to 2.2 cm. broad, lanceolate or oblong-elliptical, gradually acuminate at the apex, the tip usually rounded, acute at the base, the blade decurrent on the short thick channeled petiole (1.5 to 3 mm. long), often conduplicate or revolute; midrib impressed above, prominent beneath, ferrugineous or cinnamon brown, and bearing minute scattered appressed hairs when young, but at length glabrous or nearly so; lateral nerves 10 to 12 on each side, not impressed above, distinct beneath and colored like the midrib, glabrous, dichotomously branching and anastomosing before reaching the margin; peduncles solitary, 1-flowered, extra-axillary, sometimes nearly opposite a leaf, at first minutely appressed -pilose, at length glabrate, 12 to 16 mm. long, remarkable in comparison with closely related species for two linear-lanceolate acuminate bracteoles 2 to 4 mm. long, one situated at the base and one at or a little above the middle; flower subglobose in bud, about 15 mm. in diameter: calyx gamosepalous, subtriangular, with three slender acuminate points projecting from the broad base, appressed-pilose on the outside and with a fringe of stiff rufous hairs within at the base of the receptacle; receptacle convex, clothed with pale yellow hairs between the bases of the stamens; stamens numerous, 2.5 mm. long, the connective expanded into a broad flat hood above the pollen sacs, its surface muriculate with short stiff points but without hairs; pollen bright orange yellow, in two vertical columns of tetrads; carpels numerous, the minutely hirtellous ovaries united into a disk-like mass and bearing club-shaped, easily detached styles 1.5 mm. long; fruit not observed. (PuaTE 97 ) Type in the Kew Herbarium (from Herbarium Hookerianum, 1867), collected at the Bojfo Station, Panama Railroad, Isthmus of Panama, J une, 1861, by Sutton Hayes (no. 142). ‘A small tree, 15 to 20 feet high.”’ EXPLANATION OF PLATE 97.—Main figure, drawing of type by A. B. Boettcher. Natural size. Fig. a, flower, showing long peduncle with acuminate bracteoles; b, receptacle, bearing a few stamens and the mass of ovaries denuded of their styles; c, carpels, composed of short hairy ovaries surmounted by club- shaped styles; d, stamens, showing linear pollen sacs, one of which has opened, displaying the pollen grains in tetrads, and the expanded, mutriculate connective heads. a, Natural size; b, scale about 2; ¢, scale about 10; d, scale 16. 10. Annona jamaicensis Sprague. WILD ANNONA OF JAMAICA. Anona jamaicensis Sprague, Bull. Herb. Boiss, II. 5: 701. 1905. Anona sericea Griseb. F1. Brit. W. Ind. 5. 1864, not Dunal, 1817. A slender tree 3 to 9 meters high; young branchlets ferrugineous-pubescent, soon glabrescent; branches grayish brown or reddish brown, bearing many inconspicuous brownish lenticels; old leaf scars prominent, lined with ferrugineous tomentum; petioles 7 to 18 mm. long, channeled above, finely appressed-pubescent at first, at length glabrescent; blades ovate or obovate to obovate-oblong, shortly and obtusely acuminate at the apex and rounded or obtusely cuneate at the base, 10 to 20 cm. long, 4.5 to 8.5 cm. broad (those near the base of young branches often considerably smaller), glabrous above, finely appressed-pubescent beneath, at length sparsely so except along the ferrugineous midrib and lateral nerves; midrib impressed above, prominent beneath; lateral nerves slightly curved, 11 to 18 on each side the midrib, PLATE 97. Contr. Nat. Herb., Vol. 16. ANNONA ACUMINATA SAFFORD. Contr, Nat. Herb., Vo PLATE 98, ANNONA JAMAICENSIS SPRAGUE. Contr. Nat. Herb., Vol, 16, PLATE 99, n Boettcher del. ANNONA JAMAICENSIS SPRAGUE, SAFFORD—ANNONA SERICEA AND ITS ALLIES. 275 not impressed above, prominent and sharply defined beneath; peduncle 8 to 12 mm. long, ferrugineous-tomentose, with a small tomentose bracteole near the middle; flowers ovoid to oblong in bud, 3-petaled; calyx 3-parted, the lobes broadly ovate, obtuse or obtusely acuminate, 3 mm. long and 3 mm. broad at the base, clothed on the outside with ferrugineous tomentum like that of the peduncle; petals ovate to oblong, obtuse, 11 to 20 mm. long and 6 to 8 mm. broad, thick and leathery, clothed on the outside with fine ferrugineous velvety tomentum, lined within except near the reddish brown base with fine grayish tomentulum; stamens numerous, 1.6 to 2 mm. long, the connective somewhat broader than the lobes of the whitish pollen sacs, minutely papillose (under the microscope); carpels numerous, closely crowded in a conoid gyneecium, the styles together with the ovaries about 1.25 mm. long, the latter clothed with ferrugineous sericeous hairs; stigmas compressed-ovoid, 0.5 mm. long, cemented together at the time of pollination by a reddish brown viscous fluid; fruit globose or somewhat oblate, more or less umbilicate at the base, 4 to 6 cm. in diameter, clothed with grayish brown pubescence, with the carpels produced into tubercles usually hooked or incurved at the tips; seeds 12 to 16 mm. long, 6 to 10 mm. broad, obovate, somewhat compressed, reddish brown or tan-colored, with a smooth thin testa more or less wrinkled by the inclosed ruminate albumen. (PLATEs 98, 99.) Type in the Kew Herbarium, collected near Bath, eastern Jamaica, by William Purdie, 1844, Cotypes, without definite locality, collected by March (nos. 4, 7, 1571) and Alexander Prior (also cited by Grisebach under ‘‘A. sericea”). Distrrsution: Known only from the island of Jamaica. SPECIMENS EXAMINED: JAMAIcCA—Without definite locality, 1849-50, Alexander Prior, in Gray Herbarium (with ovoid flower bud); roadside near Hampton, Santa Cruz Mountains, alt. 700 meters, September 4, 5, 1907, N. L. Britton 1196, in herb. New York Botanical Garden (with almost mature fruit); Sheldon Road, St. Andrew, alt. 750 meters, September 10, 1897, William Harris, 6861, in U. 8. National Herba- rium (with fully developed flower and fruit). Annona jamaicensis has been known hitherto from specimens in which the flowers were evidently immature. The petals were described by Sprague as ovate, obtuse, 11 to 12 mm. long and 8 mm. broad. The accompanying drawing (pl. 99) shows them to be longer and relatively narrower when fully developed, approaching the shape of the petals of A. cherimola and its allies, but distinguished from them in not being triquetrous or keeled on the inner face. Moreover, the connective of the sta- mens is not so much swollen as in the section to which those species belong; and the incurved tips of the mature carpels serve also to prevent the confusion of this species with A. cherimola Mill., which is sometimes cultivated in the mountains of Jamaica. The flower buds somewhat resemble those of A. sericea when immature, but the indument of the petals in the present species is more velvety and of a more reddish color, while the stamens never bear hairs on the connective terminal. In addition to these points of difference the leaves are relatively broader and are never clothed with the dark red, soft, velvety lining of the leaves of A. sericea and its close allies. EXPLANATION OF PLATES 98, 99.—P1. 98, photograph of Alexander Prior’s specimen in the Gray Herba- rium (cotype collection), with immature, ovoid, unopened flower bud. Pl. 99, main figure, drawing of specimen in the U. 8. National Herbarium (from herb. Public Garden, Jamaica), by A. B. Boettcher, showing leaves, flower, and fruit. Natural size. Fig. a, stamens; }, flower with petals and some of the stamens removed; c, seeds. Fig. a, scale 12; b, scale nearly 3; c, natural size. NOMENCLATURE OF THE SAPOTE AND THE SAPODILLA. By O. F. Coox. INTRODUCTION. The botanical investigation of tropical trees, notwithstanding their economic importance, has been greatly neglected. This is shown by the lack of any general agreement among botanists regarding the names of some of the most familiar species. In the case of the sapote the nomenclatorial complications are so numerous and intricate as to become almost interesting, and they may be useful as examples of some of the taxonomic problems that still remain to be solved. Though the method of types is now widely recognized, especially in the United States, as necessary to secure stability in the applica- tion of names, many of the consequences of its adoption have yet to be worked out. Indeed, it is evident from the codes of nomenclature proposed in recent years that our European contemporaries have continued to overlook the necessity of dealing with genera on the basis of types. The Vienna code of 1905 has a curious provision regarding the subdivision of a genus, namely, that when ‘‘one of the parts de- tached contains a great many more species than the others, the name _ is reserved for that part of it.” The framing of such a rule makes it evident that the idea of a generic name being permanently associated with a type species had not yet dawned. The regulations adopted by the Brussels congress of 1910, in seeking to establish lists of nomina conservanda and to set many different dates for beginning the nomenclature of the various groups of plants, would also intro- duce endless complications into the work of placing taxonomy on a basis of types. Restricting dates of publication does not simplify the problems of generic nomenclature unless there is also to be a con- sistent method of fixing the application of the accepted names. 1 The Method of Types in Botanical Nomenclature. Science n. ser. 12: 475. 1900. Typesand Synonyms. Science n. ser. 15: 382.1902. Types of Pre-Linnszean Genera. Science n. ser. 1'7: 350. 1903. The Nomenclature of the Royal Palms. Bull. Torrey Club 31: 349.1904. An American Code of Botanical Nomenclature. Bull. Torrey Club 34: 167. 1907. 277 278 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, Generic names must have definite relations to plants, as well as specific names. Uniformity in the application of names is one of the prime essentials of stability, and the use of types is the only method thus far suggested for supplying this deficiency in our taxonomic laws. Nomenclatorial legislation that fails to consider types can have little hope of permanence. CONFUSION OF VERNACULAR NAMES. In some cases where the scientific names of tropical economic plants have fallen into confusion, resort can be had to the vernacular names, which often have very definite applications, but with the sapotes there is no hope in this quarter. The word “‘sapodilla” has only a limited use, even in the West Indies, and is scarcely known on the Continent. There it is the chicle tree that is usually called “‘sapote,” while the tree with larger fruits is distinguished as ‘‘sapote grande” or ‘‘mammee sapota.”’ But ‘‘mammee” is also the name of another tropical fruit belonging to a different family, though often confused with the sapotes. Thus there is special need of scientific names with definite applications. The only reason for adopting ‘“‘sapote” as the English name of the larger fruit is that the name “‘sapodilla,” for the smaller fruit, has already found lodgment in our English dictionaries. In addition to serving as the common name for two important fruit trees, the word ‘‘sapote”’ is also used, with a qualifier, for many other fruits, some of them belonging to distinct families, just as we say ‘‘thorn apple,” ‘‘May apple,” ‘‘rose apple,” or ‘‘custard apple.” Sapote is supposed to have been derived from ‘‘tzapotl,” the Aztec generic name for all of the soft, sweet fruits. The Spanish name “‘sapote chico” is also thought to mean ‘‘sapote chicle,”’ or ‘‘sapote with the chicle gum,” instead of signifying ‘‘small sapote.”’ ESSENTIAL DIFFERENCES. In spite of bearing the same name, the “sapote grande” and the “sapote chico” are essentially different and would never be confused by natives or residents in the tropics who have first-hand familiarity with both trees. One of the most obvious differences is in the fruit itself, which in the sapodilla has a grayish or brownish, granular flesh like a pear, while in the sapote the fruit has a yellow flesh and a firmer and more uniform texture, not crisp like the flesh of an apple, but more like that of a cooked carrot or squash. The trees are strikingly different, the sapote with an open crown of large, lanceolate, coarsely- veined, deciduous leaves and the sapodilla with a dense covering of smooth, delicately-veined, laurel-like, evergreen foliage. The tri- angular-fusiform seeds of the sapote are very large and thick and have Contr. Nat. Herb., Vol. 16. PLATE 100 SAPOTA, Le iifdem prorfus charaéteribus infigniatur hujufce plan- za6.4 tx genus 5 ac plantx fuperioris ; 5 quia tamen tota facie, com- muni omnium Americanorum confenfu & ufu, florum , fruc« tuumgque natura ab ipfa omnino differat , ut facile ipfam in America confideranti patebit , ideo hujus genus peculiare infti- tuendum cenfui. Eft igicur Sapota plantx genus fore rafacco A, plurimis fcilicet petalis B in orbem pofitis , conftante ; ¢ cujus calyce C furgit piftillum D, quod deinde abic in‘fructum FE fe- re curbinatum » aut ovatum , mollem, carnofum ,in quo nidu- latur F unum, aut duo femina fubrotunda G, compreffa, poli- ta & roftrata. Sapotx fpecies funt. Sapota frudtu turbinato, minori. Sapota fructu ovato, majort. Sapota nomen eft Americanum vulgare. 43 Sapota 5 4 | Al we ~ PLUMIER’S DESCRIPTION AND FIGURES OF SAPOTA. COOK—-NOMENCLATURE OF SAPOTE AND SAPODILLA, 279 the whole inner face covered by an enormous hilum. The seeds of the sapodilla are many times smaller, strongly compressed, and with a relatively short, narrow hilum. These differences are really more significant than those that are usually enumerated as botanical char- acters, such as the greater number of carpels in the sapodilla and the more numerous sepals in the sapote. Such characters are subject to much variation in this group of plants. The sapodilla tree not only produces a delicious fruit (now being grown in Florida) and a valuable wood, but is the source of chicle gum, which has become an important article of commerce. The sapote is of no commercial importance, though the fruit is used exten- sively for food by the native populations of Central American regions and the West Indies. The sapote ascends into the plateau regions of Central America, while the sapodilla is largely confined to regions of low elevation. PLUMIER’S ACCOUNT OF THE SAPODILLA. The taxonomic complications in this group began with Plumier, the first botanist to attempt a formal generic description of either of the fruits in question. Plumier used the word “sapote” in its _ latinized form ‘“‘Sapota” as a generic name for the sapodilla and not for what we now call the “sapote.” As that author traveled widely in the West Indies it may be argued that he must have known both fruits, but whether so or not there at least is nothing to show that he had anything but the sapodilla in mind in preparing the description and figures on which his genus was based. (PI. 100.) The seed and fruit represented in Plumier’s plate are unmistakably those of the sapodilla. The calyx is shown with only 5 or 6 divisions, not 10 or 12 as in the sapote. The seed is of the proper size and shape, with a curved spine near the middle and with a narrow bidentate base. The fruit is a symmetrical rounded oval, as in many sapo- dillas, instead of being unsymmetrical and somewhat pointed at the end as in the true sapotes. Though not closely approaching the form of the sapote, the different kinds of sapodilla show a wide variation. Some are even narrower and more elliptical than in Plumier’s figure, while others are broadly rounded or flattened. (Pl. 101.) The most misleading feature in Plumier’s plate is the indication of 5 dissepiments in the fruit, for there are 10 or 12 carpels in the sapodilla; but, on the other hand, only a few of the divisions remain conspicuous, that is, those that contain partially developed seeds. Moreover, no such obvious radiating figure appears in the ripe fruit of the sapote, where the enormous size of the seeds results in much more extensive distortion. EXPLANATION OF PLATE 100.—Reproduction of plate 4 of Plumier’s Nova Plantarum Americanarum Genera with text (p. 43). Figures original size. EXPLANATION OF PLATE 101.—Three forms of sapodilla fruit found together in the market of Guatemai City, April, 1902. Natural size. 95623°—13——2 280 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. ACHRAS SUBSTITUTED FOR SAPOTA. The Linnwan genus Achras was based directly on Plumier’s plate of Sapota, which is the sole reference in all the editions of the Genera Plantarum. In the first edition of the Species Plantarum, Achras is represented by a single species, Achras zapota, and the first refer- ence under this is to Plumier’s mention of the ovate-fruited form of Sapota, the form shown in the plate. Thus both the genus Achras and its type species, Achras zapota, are definitely established on Plu- mier’s description and figures of the sapodilla. This feature of the case seems to have escaped the attention of Radlkofer and others who have dealt with the taxonomy of the Sapotaceae. The same specific name, though with a different initial letter, was applied to the sapote in 1760 by Jacquin! under the genus Sider- oxylum. The accompanying descriptive phrase, ‘‘Siderorylum inerme; calycibus decaphyllis,”” undoubtedly alludes to the compound calyx or involucre, which is stili used by botanists as a distinctive generic character of the sapote. Yet it can not be claimed that Jacquin had at that time any intention of separating the sapote from the sapodilla, for the latter tree is not listed in the Enumeratio. The only citation given by Jacquin under his Siderorylum sapota is of Sloane’s plate of ‘‘The Mammee Sapota tree” of Jamaica. Jacquin may have borrowed his specific name from the first edition of the Species Plantarum, where Sloane was cited, as well as Plumier. CONFUSION OF SPECIES BY LINNZAUS. When the second edition of the Species Plantarum was prepared, Linneus had become aware of the existence of more than one kind of sapote in the West Indies and undertook to distinguish between them in the works of his predecessors. The sapote received a new specific name, Achras mammosa, for Jacquin’s Stderorylum sapota, if not already a synonym of the Linnean Achras zapota, would have become a homonym if transferred to the genus Achras. References to Jacquin’s species and to Sloane’s plate were the real basis of the new Linnean species, but a reference to Plumier was also included, and this has been the occasion of much confusion. Plumier had mentioned two forms of fruit, one turbinate and the other ovate, under his genus Sapota, and Linneus, assuming that these were the two fruits that were to be assigned to different species, took Plumier’s ovate fruit to be the same as Sloane’s ‘‘mammee sapota.” As a result of this mistake the same reference to Plumier’s ovate sapodilla that had been given under Achras zapota in the first edition of the Species Plantarum was transferred in the second edition to the new species, Achras mammosa, while Plumier’s reference to the 1 Enum. Pl. Carib. 15. Contr, Nat. Herb., Vol. 16. PLaTe 101. THREE FORMS OF SAPODILLA FRUITS. COOK—NOMENCLATURE OF SAPOTE AND SAPODILLA. 281 turbinate form of the sapodilla was cited under what was now called Achras sapota, the initial letter of the specific name being changed from the zapota of the first edition. From the standpoint of that time this was a natural attempt on the part of Linneus to improve the form and limit the application of his name Achras zapota of the first edition, while establishing a new species for the sapote, but we no longer admit the right to alter a name or change its original application. ADJUSTMENT OF LINNAAN NAMES. As Plumier’s drawing shows an ovate sapodilla fruit, this must be considered as the type of the genus Achras, and of the species Achras zapota, as established in the first edition of the species Plantarum. If Linnzeus had been correct in placing Plumier’s ovate (sapodilla) fruit with Achras mammosa, he would have been incorrect in apply- ing a new name to this species, for it was the ovate sapodilla that had already received the name Achras zapota in his first edition. Thus, if the treatment of the second edition were taken at its face value, Achras zapota would be the name of the sapote, Achras mam- mosa would be a synonym, the sapodilla would be left without a Linnean name, and the question of finding a post-Linnean substi- tute would arise. But if, as seems certain, Linneus was mistaken in supposing that Plumier’s ovate fruit was a sapote, there is no occasion for following this line of reasoning any further. We have only to reject the reference to Plumier under Achras mammosa as a piece of erroneous synonymy. The exclusion of the reference is also justified by the fact that Linneus gives a specific description of Achras mammosa which mentions the cunciform-lanceolate leaves of the true sapote, in contrast with ovate-oblong leaves ascribed to the sapodilla. With the misplaced reference to Plumier excluded, the name mammosa can be supported by the references to Jacquin and Sloane, and remains available as a specific designation for the sapote. TWO VARIETIES OF SAPODILLA NAMED BY JACQUIN. In a later work! Jacquin recognized the fact that the two forms mentioned by Plumier were varieties of the same species and avoided the mistake made by Linneus in associating Plumier’s ovate fruit with the sapote. For the ovate-fruited form Jacquin adopted the correct Linnean name Achras zapota and gave the subspecific name zapotilla to include the form with the broader, turbinate fruit. The sapote, though treated as a distinct species, was not designated by a binomial name, but was called Achras zapota major. 1 Stirp. Amer. 56. 1763. 282 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. The name zapotilla has been given specific rank by Coville and may come into use when a classification of the many different forms of the sapodilla is attempted. Though a reference to Plumier’s small, turbinate fruit is included by Jacquin under the subspecies zapotilla, it is preceded by references to varieties described from Jamaica by Brown and Sloane, in the latter case to Sloane’s plate no. 230. This shows a variety with rather small, round fruits, which may be taken to represent the type of the species or subspecies 2apo- tilla. Sloane’s account, published in 1725, indicates that the tree had been introduced into Jamaica rather recently from the region of Campeche. THE NAME SAPOTA NOT TO BE REVIVED. These complications regarding the sapote do not affect the sapo- dilla, which remains in undisputed possession of the generic name Achras, with Achras zapota as the specific designation and Sapota of Plumier as a generic synonym. Later use of Plumier’s name Sapota, as in Miller's Gardener’s Dictionary, does not suffice to re- establish it as a valid generic name for either the sapote or the sapo- dilla. As Achras represented a direct substitution of a name without change of generic content, the only way to restore Sapota would be to return directly to Plumier’s name. Some writers would take this course and deny the right of Linnzus to transfer the Greek name of the wild pear tree to a tropical genus. Though such a policy would not be approved at present, a future revival of interest in Greek civilization might easily lead to a restora- tion of the ancient plant names to their proper uses. In any event we should be taking entirely unwarranted liberties in transferring Plumier’s name from the sapodilla to the sapote. To do this it would have to be argued that Miller’s use of the name Sapota served to reestablish it under the binomial system, and then the principle of elimination would need to be invoked as a reason for applying the name to a different genus, because Miller included the sapote with the sapodilla. This method of selecting generic types by elimina- tion, though sometimes defended by zoologists, is very poorly adapted to botanical purposes. Pre-Linnwan botany reached a much higher development than pre- Linnean zoology and can not be wholly disregarded in the later development of the science. Nevertheless, some of the older names were discarded in the reform of botanical nomenclature by Linneus, and these names are not available for further use. Unless the names that Linneus rejected are to be allowed to rest in oblivion there is no good reason for following the proposals of Linneus. Certainly 1 Contr. U. S. Nat. Herb. 9: 369. 1905. COOK—NOMENCLATURE OF SAPOTE AND SAPODILLA. 283 nothing is gained by admitting Achras and other Linnean substitutes for pre-Linnean names like Sapota and then allowing these same pre-Linnean names to be brought back into the system because they happened to be misapplied by some of the early post-Linnean writers. If the Linnezan genus Achras had not been based on the same type, the placing of Plumier’s name Sapota as a synonym would not stand in the way of restoration by a later author who used it in the original application, but it is certainly not in the interest of nomen- clatorial stability to revive discarded pre-Linnzan names that have been replaced by direct substitutes, as in the case of Sapota and Achras. It is a rule of botanical nomenclature that the substitution of a new name does not alter the type of a genus. Still less should the type be changed by the casual use of a name for species not con- generic with the original type. Pre-Linnzan genera used~by post- Linnean authors should not be treated as having been adopted under the binomial system unless the pre-Linnzan type was included. Ap- plication of this rule to the present case requires us to seek a post- Linnean generic name for the sapote. LUCUMA AND VITELLARIA NOT APPLICABLE TO THE SAPOTE. The name usually given to the sapote in post-Linnzan literature is Incuma mammosa, but this generic assignment seems not to be cor- rect. The genus Lucuma was established by Molina in 1782 on a Chilian tree not closely related to the sapote, and the tendency of recent writers has been in the direction of separating the sapote from Lucuma. Radlkofer, Pierre, Engler, and Urban are in agreement in this respect, though differing in their applications of generic names. Radlkofer proposed to revive the name of Gaertner’s problematical genus Vitellaria and apply it to the sapote, but other writers have not followed this suggestion. As long as the type of Vitellaria remains unidentified the application of the name can not be determined. Many other genera have been segregated from Lucuma or are treated as synonyms, but none of them appears to have been based on the sapote or its closer relatives. Thus the sapote appears to have had no generic name of its own until 1890, when Pierre established a new genus, Calospermum, with Achras mammosa L. as the type species. CALOSPERMUM AND CALOCARPUM AS HOMONYMS. The name Calospermum was changed by Pierre to Calocarpum in 1897. The reason for this substitution was not stated, but may be found in the fact that Pfeiffer’s nomenclator credits Rafinesque with having given the name Calospermum to a genus of alge in 1814. Yet the name Calocarpum is open to even more serious objection, 284 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. for there was a still older name Callicarpa that had been applied by Linnzus himself to another genus of flowering plants from the same regions as the sapote. It is true that the forms of the name used by Linnzus and Pierre, ‘“Callicarpa” and ‘“Calocarpum,”’ are capable of bibliographic discrimination, but essentially they are merely variants of the same word. Such names do not differ as words, but merely as combinations of letters. Instead of aiding in the recognition of plants these ambiguous designations serve rather to confuse them. Some writers have proposed to admit variations of spelling, or even typographical errors, as constituting distinct names. The danger of this tendency finds a striking illustration in the present instance. There would be no need to stop with a few variations like Callicarpa and Calocarpum, for Kallikarpon, Kalokarpus, and many others are possible. Indeed, this name is capable of no less than 64 variations of spelling, to say nothing of the possibilities of developing some genuine typographical errors. Each family of plants might have its Calikarpum or Kalocarpon, or a whole family might be provided with generic names based on the same flexible combination. A NEW GENERIC NAME FOR THE SAPOTE. Hence, it appears that no satisfactory generic designation is avail- able in literature for the sapote, notwithstanding the many names that have been applied to it during the past two centuries. As no combinations of the word Achras are known to have been used hitherto, the name ‘‘Achradelpha,” recently proposed,! may avoid the danger of homonymy. Allusion to the sapote as the ‘‘sister of Achras”’ is warranted by the fact of similarity which has been the occasion of so much confusion. The type species of Achradelpha is Achradelpha mammosa, based on Achrasmammosa L. Some might hold that the specific name pre- viously used by Jacquin in the binomial Siderorylum sapota should be revived under the new genus, instead of adopting the Linnean name mammosa. This is one of the cases where the rule ‘‘Once a homonym, always a synonym” would find a useful application. Though it may not seem likely in the present case that the sapote will ever be referred back to Achras and thus cause a direct conflict of homonyms, if the alternative combination were adopted, there are other cases of more closely related genera where confusion would be created if the names of species were to be altered with each change of generic assignment. This course becomes necessary unless names that have been subject to rejection as homonyms are permanently discarded. Moreover, it is undesirable to have the same specific ? An advance summary of the results of this study has been published in the Journal of the Washington Academy of Sciences, 3: 158, March 19, 1913. COOK—NOMENCLATURE OF SAPOTE AND SAPODILLA. 285 name applied to two important economic species in two closely related genera. In this particular case there is another reason for refusing to base a new combination on Jacquin’s Sideroxylum sapota. There is nothing to show that this was an original name proposed for the sapote as distinct from the sapodilla. Jacquin, like Linneus, at first did not discriminate between the sapote and sapodilla. Instead of distin- guishing the two fruits and giving them different names, Jacquin’s first treatment may be interpreted as a mere transfer of the composite Linnean species from Achras to Sideroxylon. In his next work Jacquin accepted the genus Achras and used the Linnean binomial Achras zapota in its correct application to the sapodilla, with a citation of the original place of publication in Species Plantarum. Thus it was not until Linnzus proposed the name mammosa, in the second edition of Species Plantarum, that the sapote can be said to have received a specific designation. SUMMARY OF PRINCIPAL SYNONYMS. The results of this study of the nomenclature of the sapote and the sapodilla are summarized in the following lists of the principal synonyms: . Achradelpha mammosa (L.) Cook. THE SAPOTE. Malus persica maxima foliis magnis, etc. Sloane, Voy. Jam. 2: 124. pl. 218. 1725. Siderorylum sapota Jacq. Enum. Pl. Carib. 15. 1760. Achras mammosa L. Sp. Pl. ed. 2. 469. 1762. Achras zapota major Jacq. Stirp. Amer. 56. 1763. Sapota mammosa Mill. Gard. Dict. ed. 8. no. 2. 1768. Lucuma mammosum Gaertn. f. Fruct. & Sem. 3: 130. pl. 203-4. 1807. Lucuma mammosa DC. Prodr. 8: 169. 1844. Vitellaria mammosa Radlk. Sitzungsb. Math.-Phys. Akad. Miinchen 12: 325. 1882. Calospermum mammosum Pierre, Notes Bot. Sapot. 11. 1890. Calocarpum mammosum Pierre in Urban, Symb. Antill. 5: 98. 1904. Achradelpha mammosa Cook, Journ. Washington Acad. Sci. 3: 160. 1913. Achras zapota L. THE sAPOoDILLA. Sapota fructu ovato, majort Plum. Nov. Pl. Amer. 43. pl. 4. 1703. Achras zapota L. Sp. Pl. 1190. 1753; Jacq. Stirp. Amer. 57. 1763. Achras sapota L. Sp. Pl. ed. 2. 1. 470. 1762. Sapota achras Mill. Gard. Dict. ed. 8. no. 1. 1768. Sapota zapotilla Coville, Contr. U. 8. Nat. Herb. 9: 369. 1905. A MONOGRAPH OF THE HAUYEAE AND GONGYLOCAR- PEAE, TRIBES OF THE ONAGRACEAE, By Joun Donne. Samira and J. N. Ross. INTRODUCTION. A recent study of the genus Hauya, native of the mountains of Mexico and Central America, has led to the discovery of a new species and has shown that an outlying species, Hauya arborea, from the desert of Lower California, must be excluded and made the type of a new genus. The differences between Hauya and the new genus, here named Xylonagra, are so numerous that one feels the need of calling attention less to these than to the points they have in com- mon which are requisite to keep them in the same tribe. Herbarium work on the group has been supplemented by field work on the part of Dr. Rose, who studied Xylonagra on Cedros Island and brought back a large number of herbarium specimens. During his visit to Lower California Dr. Rose collected material of another species of the Onagraceae, which has been referred wrongly to the genus Gongylocarpus. This, too, proves to be of a new generic type, here named Burragea, which with Gongylocarpus forms a new tribe. SYSTEMATIC TREATMENT. Tribe HAUYEAE Raim.! Calyx tube long-produced above the ovary, cylindrical, deciduous, the segments 4, elongated, reflexed in anthesis; petals 4, inserted at the apex of the calyx tube, sessile; stamens 8, inserted with the petals; filaments subulate or filiform, the anthers linear, versatile; ovary 4-locular, the ovules few or very numerous, affixed to the interior angle of the cells, ascending; capsule oblong, ligneous, loculicidal, the valves 4, septiferous in the middle; seeds few or very numerous, ascending, imbricate, obliquely produced upward into a subcartilaginous wing. Trees or shrubs; leaves alternate, petiolate, simple, entire; flowers solitary or 12 leafy racemes. Hauya was made the type of the tribe Hauyeae by Raimann, it being the only genus. As treated here, the tribe contains Hauya and Xylonagra. KEY TO THE GENERA. Arborescent; leaves large; flowers large, few, axillary; seeds biseriate.........----- 2-22 e eee eee eee eee ee eee eee 1. Havya (p. 288). A dwarf shrub; leaves small; flowers minute, numerous, in a terminal, leafy-bracted raceme; seeds uniseriate..... 2. XYLONAGRA (p. 294). 1 In Engl. & Prantl, Pflanzenfam. 37: 211. 1893. 987 8 288 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 1. HAUYA DC. Prodr. 3: 36. 1828. Frutescent or arboreous; flowers large, axillary, solitary, sessile or pedunculate, from white to pinkish; calyx tube cylindrical, elongated, the segments narrow; stamens more or less exserted; filaments filiform; anthers about as long as filaments, awned at apex, reticulate, at length twisted; stigma globose or ellipsoidal; ovules most numerous, in 2 rows to the cell; capsule linear-oblong or ellipsoidal; seeds most numerous, biseriate, the testa coriaceous, the wing longer than the testa, unilaterally auriculate and incrassate; cotyledons oblong, compressed, flat; radicle very short. The genus Hauya wasdescribed in 1828 by A. P De Candolle in the third volume of the Prodromus, and a second description, probably prepared before the first, was published the next year. The genus contained a single species, H. elegans, based upon a drawing which was published along with the second description. This illustration was one of the few of the Mocino & Sessé drawings which De Candolle was able to reproduce, the others being known only from the tracings which were afterward distributed to a fewinstitutions. Nothing more was learned of the genus until 1877, when M. Barcena, a Mexican, described and figured as H. elegans a plant which he had collected in the State of Hidalgo. His illustration is not very good and no specimens of the plant were preserved. Mr. W. Botting Hemsley in 1878 described two additional species and in 1880 he amplified his descriptions and redescribed H. elegans, publishing with the text two plates to illustrate the three species. The first of his new species, H. barcenae, is based on Andrieux’s no. 391 from Oaxaca, Mexico; the second, H. cornuta, upon two speci- mens, one collected by Salvin and one by Savage, in Guatemala. Neither of these species has been re-collected. In 1883 Dr. Sereno Watson transferred Oenothera arborea Kellogg to Hauya, and was followed by Mrs. M. K. Curran, Dr. E. L. Greene, and others, but, as will be shown under Xylonagra, it isa very different plant from true Hauya. In 1893 Mr. Smith described two new species of Hauya from Guatemala and in 1898 he assigned one of these, H. rodriguezti, to Costa Rica, publishing a new description, while in 1909 he and Doctor Rose described 5 new species from Central America. This in brief is the taxonomic history of the genus. There are a few more references to it in literature, but they are compiled from the papers mentioned above. The genus has its center of distribution in Guatemala; in fact, it is largely Guate- malan, for, of the 11 species described below, 7 occur in that country. One species is known from Costa Rica. Three species are known only from Mexico and one of the Guatemalan species has been found just over the border. The Mexican species are widely scattered, one each coming from the States of Hidalgo, Oaxaca, Guerrero, and Chiapas. This genus was named for Abbé René Just Haiiy, celebrated as a mineralogist, being the discoverer of the true laws of crystallization, who was in early life a student of botany. He was born in 1743 and died in Paris in 1822. He was a friend of the elder De Candolle, who published the genus Hauya six years after Haiiy’s death. KEY TO THE SPECIES OF HAUYA. Flowers distinctly peduncled. Peduncles much longer than the ovary.............-....... 1. H. heydeana. Peduncles shorter than, or at most equaling, the ovary. Calyx segments not appendaged........................ 2. H. barcenae. Calyx segments appendaged. Calyx segments shorter than the tube............... 3. H. lucida. Calyx segments longer than the tube............... 4. H. rusbyi. SMITH AND ROSE—THE HAUYEAE AND GONGYLOCARPEAE. 289 Flowers sessile. Calyx segments not appendaged.............-...-.-.-.--.-- 5. H. elegans. Calyx segments appendaged. Appendages 3 to 4 mm. long. Valves of the capsule with a central dorsal ridge.... 6. H. cornuta. Valves of the capsule without a central dorsal ridge. 7. H. microcerata. Appendages 10 to 15 mm. long. Calyx segments about equaling the tube........... 8. H. rodriguezit. Calyx segments much shorter than the tube. Capsules 3.5 cm. long...............-- Le eeeeee 9. H. quercetorum. Capsules 6 to 8 cm. long. Leaves orbicular to cordate-oval........... 10. H. ruacophila. Leaves oblong-ovate to oblong-obovate..... 11. H. lemnophila. 1. Hauya heydeana Donn. Smith, Bot. Gaz. 18: 3. 1893. Figure 465. Shrub or small tree, at most only 6 meters high, glabrous throughout; young branches purplish; leaf blades lanceolate, 6 to 12 cm. long, narrowed at the base, gland-tipped; petioles 2 cm. or less long; peduncles 12 to 35 mm. long; flowers small for the genus, 6 to 7 cm. long; calyx segments narrowly linear, longer than the tube, with short, blunt tips; petals as long as and twice as broad as the calyx segments, purplish; cap- sule 2.5 cm. long, the valves 4 mm. broad, plane on the back. Type LocaLity: Along the road from San Cristébal, Department of Alta Verapaz, to Beleji, Department of Quiché, Guatemala. SPECIMENS EXAMINED: GUATEMALA: Beleji, May, 1892, Heyde & Luz (J. D. Smith 2935). Coban, von Tiirckheim (J. D. Smith 8210) in June, 1902, 604 in 1907. This species is well marked in both foliage and flowers and especially by the long peduncles. 2. Hauya barcenae Hemsl. Diag. Pl. Mex. 1: 13. 1878. Ficure 46. Tree 13 meters or more high; young branches puberulent; leaves ovate to rotund, 5 to 6 cm. long, acute, rounded at the base, puberulent, becoming glabrate, sub- coriaceous; petiole 25 mm. long; peduncle short but distinct; flowers 6 to 7 cm. long; calyx tube about 3.5 cm. long, the segments about as long as the tube, not appendaged ; capsules 5 cm. long, the valves plane on the back; seeds small, 5 mm. long. Type Locauity: ‘‘Huauapan” (Huajuapan), Oaxaca, Mexico. SPECIMENS EXAMINED: Mexico: Huajuapan, Oaxaca, Andrieux 391. InLustraTions: Hemsl. Biol. Centr. Amer. Bot. 5: pl. 29. f. 2; Engl. & Prantl, Pflanzenfam. 37: f. 87. 3. Hauya lucida Donn. Smith & Rose, Bot. Gaz. 52: 48, 1911. Figure 47. Hauya donnelismithit Loes. Repert. Nov. Sp. Fedde 12: 236. 19138. Tree 10 to 13 meters high, with a globose or spreading top; very young leaves pubes- cent on the nerves beneath; mature leaves glabrous and shining, obovate to elliptical, cuspidate, narrowed at the base, 8 to 13 cm. long, 4.5 to 6 cm. broad, with 8 or 9 lateral nerves; petioles 1.5 to 2.5 cm. long; peduncle very short, not over 7 mm. long; flower 8 to 10 cm. long; calyx tube 4 to 6 cm. long, the segments 3 to 4 cm. long, with append- ages 3 to 4 mm. long; petals 3 cm. long; filaments 17 to 19 mm. long, or after flowering becoming 20 to 23 mm.; ovary 9 to 12 mm. long; stigma exserted beyond the petals; capsule 3 to 4.5 cm. long, the valves plane on the back; seeds oblong, 11 to 12 mm. long, obtuse; embryo oblong-obovate. TYPE LocaLity: Rio Torres, San Francisco de Guadalupe, near San José, Costa Rica. SPECIMENS EXAMINED: Costa Rica: Near San José, June, 1893, Tonduz 8005 (type); same locality, April, 1894, J. D. Smith 4801; June 24, 1896, Tonduz 10118; October, 1898, Tonduz 7445, 12719. 94559°—13——2 290 CONTRIB “SMITH AND ROSE—THE HAUYEAE AND GONGYLOCARPEAE. 291 4, Hauya rusbyi Donn. Smith & Rose, sp. nov. Fieure 48. Small tree; young branches greenish, pubescent with soft, spreading hairs; leaves broadly oblong-ovate, 8 to 12 cm. long, sometimes abruptly acuminate, rounded at the base, somewhat pubescent above, paler beneath and with dense soft pubescence; petiole 12 to 20 mm. long; peduncle 8 mm. long; flowers 7 to 8 cm. long; calyx tube 3 cm. long, the segments narrow, 4 to 5 cm. long, appendaged at the tip, very pubescent without, within greenish above, reflexed after flowering and purplish below; petals not seen; style purple; capsule 3.5 to 4 cm. long, plane on the back; seeds 8 mm. long. Typeinthe United States National Herbarium, no. 574646, collected on Liméa Mountain, Guerrero, Mexico, altitude 1,350 meters, July 28, 1910, by H. H. Rusby (no. 157). Also . collected in the mountains near Iguala, Mexico, October 24, 1909, by C. G. Pringle (no. 9258). Hauya rusbyi is probably nearest H. barcenae, differing from it in the shape of the leaves, in its somewhat shorter petioles, in having the calyx tube shorter than the segments and the segments appendaged, and in its somewhat shorter capsule. 5. Hauya elegans DC. Prodr. 3: 36. SS 1828. Shrub or tree; young branches vel- vety-pubescent; leaves lanceolate to ovate, 27 to 62 cm. long, rounded at the base, acuminate, tomentose above, velvety-pubescent beneath; flowers sessile, 12.5 cm. long; calyx tube longer than the calyx segments, these not appendaged; petals nearly orbicular, rose-colored; capsule 3.7 cm. long, the valves plane on the back. TYPE LocaLity: Mexico. The taxonomic history of this, the type species of the genus, has already been given with that of the genus. There is only to add that in 1880 Mr. Hemsley referred here Coulter’s no. 172 from Zimapan, Mexico. 6. Hauya cornuta Hemsl. Diag. Pl. Mex. 1: 13. 1878. Ficure 49, Shrub or small tree; young branches pubescent; leaves rather small for the genus, oblong-lanceolate to elliptical, 5 to 7.5 cm. long, short-acuminate, puberulent above, canescent-tomentose beneath; calyx tube 3 cm. long, puberulent, the segments about half the length of the tube; capsule 22 mm. long, the valves with a short ridge on the back. TYPE LocALITy: Volc4n de Fuego, Guatemala. Only the type specimen has been seen, which was collected in Guatemala, August 6, 1873, by Osbert Salvin. The material was lent us by the director of the Kew Gardens. Fia. 48.— Hauya rusbyi. Rusby’s 157. Natural size. 292 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 7. Hauya microcerata Donn. Smith & Rose, Bot. Gaz, 52: 46.1911. Ficure 50. Probably a small tree; young branches and buds with appressed, canescent pubes- cence; leaves long-petiolate, obovate to oblong-obovate, 7 to 11 cm. long, 4 to 6 cm. broad, shortly cuspidate, glabrate above, tomentose beneath; flowers sessile; calyx tube 8.5 to 10 cm. long, the segments 3.5 to 4 cm. long, with appendages 3 to 4 mm. long; petals oval, 33 mm. long; ovary velvety-pubescent, 11 mm. long; capsule 5 cm. long, the valves plane on the back; seeds unknown. TyPE Locauity: Santa Rosa, Department of Baja Verapaz, Guatemala. SPECIMENS EXAMINED: GuaTemaLa: Santa Rosa, alt. 1,500 meters, September, 1888, von Tiirckheim (J.D, Smith 1423). Cuesta de Quililhd, near Purulhd, alt. 1,400 meters, April, 1905, Pittier 155. Mexico: Canjob, Chiapas, May 2, 1904, Goldman 923. The specimen from Mexico is referred here with some doubt. wy & ‘ YS WL Fig. 49.— Hauya cornuta. Fic. 50.— Hauya microce- Type. Natural size. rata. Pittier’s 156. Natural size. 8. Hauya rodriguezii Donn. Smith, Bot. Gaz. 18: 3. 1893. Figure 51. Tree, 10 meters high, with a stout trunk; young branches slightly pubescent; leaves pubescent when young, soon glabrate except the angles, rhombic-oval to elliptical, 11 to 12 cm. long, abruptly acuminate, narrowed or rounded at base, with 8 or 9 pairs of nerves; petioles 1 to 3 cm. long; flowers sessile, 13 to 15 cm. long; calyx tube 6.25 to 7.5 cm. long, the segments of about the same length as the tube, with slender ap- pendages; petals oval, two-thirds the length of the calyx segments; ovary canescent ; capsule 5 to 5.25 cm. long. TYPE LocaLity: Acatepeque, Guatemala. SPECIMENS EXAMINED: GuaTEMALA: The type specimen, collected at Acatepeque, March, 1892, by J. D. Smith (no. 2529), 9. Hauya quercetorum Donn. Smith & Rose, Bot. Gaz. 52: 47.1911. Ficure 52. Probably a small tree; old branches glabrescent, the growing parts somewhat hirsute; leaves large, orbicular to lanceolate, obtuse to acuminate, glabrous or nearly so above, SMITH AND ROSE—THE HAUYEAE AND GONGYLOCARPEAE. 293 glaucous and glabrous beneath except for the hairs along the midrib and lateral veins; petioles 2 to 3 cm. long; flowers sessile; calyx tube 7 to 9 cm. long, pubescent, the seg- ments 3.5 to 5 cm. long, with appendages 5 to 6 mm. long; ovary very pubescent, 10 to 11 mm. long; capsule woody, 3.5 cm. long, the valves plane on the back; seeds lanceolate, 15 mm. long, 5 mm. broad, acute. TYPE LocALITy: Volcano of Juamaytepeque, Department of Santa Rosa, Guatemala. Known only from the type locality; collected by Heyde and Lux in March, 1883. On the same volcano and at the same altitude Heyde and Lux collected additional material of a Hauya which may or may not belong here, the leaves being of somewhat Fig. 51.—Hauya rodriguezii. Type. Fia. 52.— Hauya quercetorum. Type. Natural size. Natural size. different shape, the ovary more hirsute, and the flower somewhat different in several respects. 10. Hauya ruacophila Donn. Smith & Rose, Bot. Gaz. 52: 47.1911. Fiaure 53. Hauya longicornuta var. a ovalifolia Loes. Repert. Nov. Sp. Fedde 12: 237. 1913. Small tree; young branches velvety-pubescent; leaves orbicular to ovate, 5 to 7 cm. long, 3 to 6 cm. broad, somewhat pointed, cordate or rounded at the base, glabrous above, densely cinereous beneath; flowers sessile; calyx tube 9.5 to 10 cm. long, pubescent, the segments 5 to 5.5 cm. long, with appendages 10 to 12 mm. long; petals elliptical, 4.5 cm. long, 2.5 cm. broad; ovary velvety; capsule woody, 6 cm. long, the valves plane on the back; seeds oblong, 13 mm, long. 294 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. TYPE Locauity: Volcin Acatenango, Guatemala. SPECIMENS EXAMINED: GuateMALa: Volcén Acatenango, Department Zacatepéquez, alt. 1,700 meters, March, 1892, J. D. Smith 2528 (type). Alotenango, Department Zacatepé- quez, alt. 1,300 meters, March, 1892, J. D. Smith 2527. . 11. Hauya lemnophila Donn. Smith & Rose, Bot. Gaz. 62: 48.1911. Fiaure 54. Hauya longicornuta var. b. oblongifolia Loes. Repert. Nov. Sp. Fedde 12: 237. 1913, Probably a small tree; young branches cinereous-hirsute; leaves oblong-ovate to oblong-obovate, 9 to 15 cm. long, 4.5 to 7.5 cm. broad, rounded at the base, some- what pointed, glabrate above, hirsute beneath; flowers sessile, calyx tube 7.5 to 9cm. long, the segments 4.5 to 5 cm. long, with appendages 12 mm. long; petals white, 33 to 35 mm. long; ovary hirsute, 13 tol4 mm, long; capsule linear-oblong, 7.5 to 8 cm, long, the valves plane on the back. TYPE LocaLity: Laguna de Carrazal, Depart- ment of Santa Rosa, Guatemala, altitude 1,500 meters, Known from the type material, collected in May, 1892, by Heyde and Lux (J. D. Smtth 2936), and also from Heyde’s no. 516, which may be a part of the type collection. 2. XYLONAGRA Donn. Smith & Rose, gen. nov. A low, bushy shrub with numerous short, ascending branches; leaves alternate, small, sol- itary or in fascicles, with glandular tips; inflo- rescence of elongated terminal racemes; flowers pediceled; calyx scarlet, the proper tube short, abruptly enlarging into a long, funnel-shaped throat, the segments 4, short, ovate to triangular, about one-third the length of the tube; petals 4, scarlet, shorter than the calyx segments; sta- mens 8, exserted a little beyond the petals, the Fic. 53.—Hauya ruacophila. Type. filaments short, the anthers mucronate; style Natural size. exserted beyond the stamens; stigma capitate; capsule small, 4-celled, loculicidal; seeds few to the cell, in one row, with a small subterminal, acuminate wing; embryo oblanceolate, the cotyledons elliptical, thrice longer than radicle. This genus differs from Hauya in its bushy habit and small leaves, in the character of the inflorescence, in its much smaller flowers, differently shaped calyx tube, highly colored petals, short filaments, merely mucronate and not reticulate anthers, much smaller capsules of different texture with cells containing one row of few seeds instead of two rows of very numerous ones, and in the very small seeds with a different kind of wing. Ithas, moreover, a very different range and habitat. 1. Xylonagra arborea (Kellogg) Donn. Smith & Rose. Oenothera arborea Kellogg, Proc. Calif. Acad. 2: 32. pl. 1859. Hauya californica 8. Wats. Proc. Amer. Acad. 20: 366, 1885. Hauya arborea Curran, Proc. Calif. Acad. II. 1: 253. 1888, Low shrub, 60 to 120 cm. high (originally described as 6 to 8 feet), with slender woody branches, these when young red and covered with short, appressed, cinereous pubescence; leaves petioled, small, the blade lanceolate, 8 to 15 mm. long, entire, obtuse, tipped by a large orange yellow, sometimes red, gland, scantily pubescent on SMITH AND ROSE—THE HAUYEAE AND GONGYLOCARPEAE. 295 \ NN Fia. 54.—Hauya lemnophila. Heyde’s 516. Natural size, 296 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. both surfaces; inflorescence a terminal, leafy raceme, sometimes consisting of only 1 to3 flowers, sometimes much elongated (in one case measuring 50 cm. long) and bearing 100 or more flowers; pedicels 5 mm. long, becoming stout; flowers scarlet, 2 to 2.5 cm. long; petals erect, 5 mm. long; stamens only slightly longer than the petals; style slender, scarlet, pubescent, exserted; capsule 10 to 12 mm. long; seeds 5 to 8 in each cell, in one row, each slightly winged on one side and at the summit, the terminal portion as long and broad as the body itself. The above description differs somewhat from Doctor Kellogg’s rather extravagant one. He characterizes it as a tree primrose with a stem 5 to 8 cm. in diameter, the floral branches bearing glandular hairs, the leaves sessile, and the flowers in dense spikes. The specific name arborea is unfortunate, as there is not the least suggestion of a tree in either the size or the habit of the plant. TYPE Locauity: Cedros Island, Lower California. The range and habitat are Cedros Island and the west coast of Lower California on the open desert hills and low mountains, along with Agaves and junipers. SPECIMENS EXAMINED: Lower Cauirornia: Cedros Island, 1896, Anthony 46; March, 1911, Rose 16133. San Andrés, September 21, 1905, Nelson & Goldman 7157. San Bartolome Bay, April 29, 1889, Pond. Rosalia Bay, 1896, Anthony 46a. This species was collected in considerable abundance on Cedros Island by Dr. Rose during his recent trip with the Albatross. He found it only after an all-day climb. Unfortunately, only a single flower was seen, but fruiting specimens with leaves were taken, and these will doubtless be a welcome addition to many herbaria. The plant is low, usually 60 to 90 cm. high, a bush with many slender upright branches, often forming thickets along the dry ravines and stony hillsides. The leaves, as seen, were always small, and no evidence was met that they are ever large, so that doubtless the record of their being sometimes 2 inches long is due to a slip of some kind. Each leaf is tipped with a large gland, which suggests those found on the tips of leaflets in ant-inhabited Acacias. Whether these glandular bodies serve any special use in the economy of this plant we do not know. The inflorescence is some- what curious, and it is not easy to decide whether the flowers should be said to be axillary or, taken together, to form a leafy raceme. The flowering branch seems to die after the fruit matures, although it may persist for years. In one case, however, a branch bearing perhaps a hundred capsules had stopped flowering and been con- tinued as a leafy shoot. This was probably abnormal and doubtless due to its desert surroundings. This plant, like so many desert plants, must have a severe struggle all the time. The warm sunshine urges it into leaf and flower, while the dry soil withholds what little water it has. One day it puts forth its leaves; the next it must drop them or die. A day later a shower stimulates it to a new effort. Thus are produced these abnormal forms—stunted branches an inch long, which are 20 years old, and elongated branches measured in feet which have grown in a season. Tribe GONGYLOCARPEAE Donn. Smith & Rose. Calyx tube produced above the ovary, filiform, nearly closed at the apex by an annular disk; calyx segments 4, much shorter than the tube; petals 4, inserted at the base of the disk; stamens 8, inserted with the petals, inappendiculate, four longer than the othérs, all fertile, the anthers ovoid; ovary immersed; stigma capitate exindusiate; fruit immersed, subdrupaceous, rhorboidal, bilocular, 2-seeded; puta- men ligneous. Leaves alternate; flowers axillary, solitary. The tribe Gongylocarpeae, here described as new, seems to be abundantly distinct from the Gaureae by its immersed ovary and 2-celled fruit. SMITH AND ROSE—THE HAUYEAE AND GONGYLOCARPEAE. 297 KEY TO THE GENERA. Herbaceous; leaves not crowded, petiolate, denticulate; disk glandular; ovary concrete with the stem and petiole, 2 or 3-celled, 2 or 3-ovuled; fruits remote. 1. GONGYLOCARPUS (p. 297). Fruticose; leaves crowded, subsessile, entire; ovary immersed in the stem, 2-celled, 2-ovuled; fruits approximate in the thickened stems............. 2. BurraGeEa (p. 297). 1. GONGYLOCARPUS Schlecht. & Cham. Linnaea 5: 557. 1830. An annual herb, glabrous, the stem becoming reddish; leaves petiolate, ovate- lanceolate, denticulate-ciliate; flowers remote; calyx tube adnate to both branch and ‘petiole, the segments spreading, linear, cucullate; petals inserted on the glandu- lar disk of the calyx, obovate-cuneate, shorter than the calyx segments, entire; ovary 2 or 3-celled, the style filiform and short, the ovules solitary in each cell; fruit con- crete with branch and petiole. Only a single species of Gongylocarpus (G. rubricaulis) has been described, which is extremely rare in collections. The type came from near Jalapa, Vera Cruz, Mexico, and the species has been reported also from the States of Jalisco and Chihuahua, widely separated localities. There is a possibility that more than one species is represented by this material. Owing to the weedy appearance of the plant and to its inconspicuous flowers and seemingly abnormal fruits, it is likely to be neglected by collectors. 1. Gongylocarpus rubricaulis Schlecht. & Cham. Linnaea 5: 558. 1830. Stem 45 cm. long, erect; branches remote, shorter than the stem; leaves acuminate or acute, narrowed into the petiole, remotely and sharply denticulate, shortly or obsoletely ciliate, the cauline 6 cm. long and 2.5 cm. broad, petiole 1.25 em. long; calyx tube above ovary 4 to 10 mm. long; segments 4 mm. long; petals deciduous; stamens and style equaling or exceeding the calyx segments. Tyre LocaLity: Jalapa, Mexico. SPECIMENS EXAMINED: Mexico: Near Jalapa, Vera Cruz, Schiede & Deppe. Naolinco, Vera Cruz, August, 1912, C. A. Purpus 6155. Mountains near Chapala, Jalisco, Novem- ber 18, 1892, C. G. Pringle 5344. Southwestern Chihuahua, August to November, 1885, Dr. E. Palmer 34. 2. BURRAGEA Donn. Smith & Rose, gen. nov. A perennial shrub; leaves alternate, closely set, subsessile, thickish, entire; sepals 4, in the bud free at the tip, highly colored, oblong, becoming reflexed, all similar; calyx tube very slender, much longer than the segments, partly closed at the apex by an annular disk; filaments 8, 4 longer; style slender; stigma capitate, at length bipartite; ovary imbedded in the branch, 2-celled; fruit somewhat diamorid-shaped, 2-celled, 2-seeded, imbedded in the flowering shoot and tardily breaking away. Type species, Gaura fruticulosa Benth. This remarkable genus is dedicated to Commander Guy H. Burrage, United States Navy, who was in charge of the U. S. steamer Albatross during the spring of 1911, when Dr. Rose made his expedition to Lower California. We thus render a slight token of our appreciation of the facilities so richly supplied during this expedition. The genus Burragea is to be associated with Gongylocarpus, but in its bushy, perennial habit, large, showy flowers, and elongated flowering branches it must be regarded as quite distinct. Asin that genus the ovary is sunk in the flowering branch; but in Gongylocarpus the fruits are single and suggest little nuts, while in Burragea they represent a persistent collective fruit. 298 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Mr. Bentham, who was the first to study the species, took the specimens for a de- formed Gaura, and commented on them as follows: ! “In all specimens there is the same semiarticulation and thickening of the flowering part of the branches so as to enclose the ovaria, which Chamisso and Schlechtendal observed in an allied Mexican species, and which appeared to them to justify the constituting a distinct genus under the name of Gongylocarpus. It is, however, much more probably the effect of some disease or parasite. I could not indeed, any more than the above quoted authors, discover any traces of fungus or insect, but the distortion of the ovaries, as well as of the more enlarged capsules and seeds which may be found still remaining in their hardened state in the old woody branches, show that this is not the healthy natural form of the plant. The flowers appear in all other respects perfect, and are evidently showy.”’ KEY TO SPECIES. Glandular-pubescent; leaves linear..............2-.0..ceeeeeeeeee 1. B. fruticulosa. Glabrous; leaves oblanceolate...............00 0.0. c cee cececeecee 2. B. frutescens. 1. Burragea fruticulosa (Benth.) Donn. Smith & Rose. Gaura fruticulosa Benth. Bot. Voy. Sulph. 15. 1844. Gongylocarpus fruticulosus T. S. Brandeg. Proc. Calif. Acad. II. 2: 158. 1889. A low, bushy plant, 30 to 60 cm. high; young branches purple, covered with short, spreading, glandular (?) hairs; leaf branches often short and stunted, showing numerous closely set leaf scars; leaves numerous, often closely set, on vigorous branches more remote, linear, 2 mm. broad, entire, acute, slightly narrowed at base, not perceptibly veined, glandular-pubescent; calyx tube slender, 15 mm. long, the segments oblong, pointed, 7 mm. long, becoming reflexed; petals obovate, about 7 mm. long. TYPE LocaLity: Magdalena Bay, Lower California. Collected on Santa Margarita Island, Lower California, March 19, 1911, by J. N. Rose (no. 16284). Bentham evidently had both species of Burragea in his Gaura fruticulosa, for he says that some of the specimens are glabrous and others are pubescent. The pubescent form has been selected to serve as a type for Bentham’s species. Both species grow about Magdalena Bay, but, so far as observed, not together. . 2. Burragea frutescens (Curran) Donn. Smith & Rose. Gongylocarpus frutescens Curran, Proc. Calif, Acad. II. 1: 231. 1888. A low, widely spreading shrub, rarely over 30 to 60 cm. high, glabrous throughout, with numerous short branches, the young ones dark purple; leaves alternate, narrowly oblanceolate, often 7 mm. broad, 3 cm. long, thickish, the midvein indistinct; calyx tube 2 to 2.5 cm. long, petals 12 mm. long; fruiting branches thick, 1 to 10 cm. long, turgid, purple, at first leafy but becoming naked. TYPE LOCALITY: Magdalena Bay, Lower California. Collected also on the shore of Santa Maria Bay, Magdalena Island, Lower California, March 18, 1911, by J. N. Rose (no. 16263). NOTE. While reading the final proof of this paper, our ajtention was called to two species, one with two varieties, just described by T. Loesencr in Fedde, Repertorium Spe- cierum Novarum Regni Vegetabilis, volume 12. These are based on the same collec- tions which we have studied and, being clearly synonyms, are so referred. Two names, H. hemsleyana and H. pedicellata, also are mentioned, but doubtless have not been published. 1 Bot. Voy. Sulph. 15. 1844. BOTRYCHIUM VIRGINIANUM AND JTS FORMS. By Ivar Tipesrrom. BOTRYCHIUM VIRGINIANUM. In 1905,! an account was given by me of a peculiar form of our Virginia plant. It showed a forking sporophyll, which phenomenon does not appear to be so unusual, for I have observed a number of similar cases since that time. The unusual, however, about the plant was the presence of two fertile pinne on one of the sterile segments (branches). The plant was collected in the woods along Chesapeake Bay, some 30 miles east of Washington. Along with typical specimens grew some very small plants which were referred to the small form of the species or B. gracile Pursh. In 1907, while botanizing near Chevy Chase, Maryland, I came upon a rather large colony of the species and found also a large num- ber of smaller plants growing under the shade, so to speak, of plants of normal size. The smallest fruiting specimen which I could find meas- ured 10 cm. in height. The branches of the sterile frond measured about 2 cm. in length while the fertile frond somewhat exceeded this measure. The spore-bearing part alone (i. e., exclusive of the stalk) measured only 5 mm. The largest specimen of B. virgimanum which I have collected in Maryland, measures about 70 cm. in height, while the sterile branches are 20 cm. or more and the pinne 6 em. more or less in length. Much larger specimens may exist, but I give measurements only of those which I have collected and which are pre- served in my own herbarium. The size of the plant seems to depend largely on its age and the perennial root appears to last many years. The variation, therefore, of the species in Maryland and Virginia (fruiting specimens alone being considered at this time) appears to lie between 10 and 70 cm.,so far as the height of the plant is concerned. It is apparent that B. gracile Pursh (B. virginianum var. gracile Presl) is only a young fruiting plant of our typical B. virginianum. One of the earliest records of our Virginia plant is found in the works of Morison ? where it is under the name: Lunaria botrytis elatior Virginiana pinnulis tenuissimis, etc. 1 Torreya 5: 160. f. 1. 2 Pl. Hist. 3: 595, sect, 14. pl. 4. f. 5. 1799. 299 300 © CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. The plant was sent to England by the pioneer of Virginian botany, John Banister. Gronovius and Linneus referred it to the genus Osmunda in Flora Virginica. The range given for the species by Professor Underwood ' is: “In woods, Nova Scotia to British Columbia, Florida and Arizona. Also in Europe and Asia.”’ It is comparatively easy to clear up the history of a species when we are in a position to study the living individuals. If, on the other hand, we are limited to fragmentary material, as was often the case with the old botanists, it is not at all surprising that such simple, diminutive forms as the early fruiting fronds of B. virginianum were given a distinct name or referred to the species as a variety. If we examine material from other parts of America as, for instance, the region from Mexico southward into Peru, we are confronted with the same difficulty as the earlier botanists, for the material from the Latin-American countries, besides being scant, shows a much greater range of variation than do the plants from the United States. With an insufficiency of material, therefore, and with practically no knowledge of the living plants, we can do little except to describe the herbarium material at hand and to attempt to refer it to forms already known. The record given by Linneus? of his Osmunda virginiana (Botrychium virginianum) reads: Osmunda scapo caulino solitario, fronde supra decomposito. Osmunda fronde pinnata caulina, fructificationibus spicatis. Gron. virg. 196. Osmunda asphodeli radice. Plum. fil. 136, t. 159. Pet. fil. 168. t. 9. f. 2. Habitat in America. From the above it is apparent that the concept of the species as held by Linnaeus has not changed. It is doubtful also if he ever saw Plumier’s plant. The account of this fern points clearly to B. virginianum or some close ally, but the figure is not of our plant in all its details. It should be stated also that the botanical artists of two centuries ago often disregarded the details and a striking result of this license is seen in the earliest representations of our handsome . Adiantum pedatum. We cheerfully make allowance for the artist’s fancy when we read Cornut’s description of the plant, for in the description we find that very element, the treating of plants ,as living beings, which is necessarily of the most importance in the make up of any book on botany. Plumier® gives an account of the plant of which the following is a part: “Je trouvay cette Plante dans les forests de l’Isle Saint Domingue, 6u j’en ay veu une autre espéce tres semblable, mais dont les feuilles étoient plus émoussées, un peu moins découpées & bordées d’une dentelure tres-delicate. 'Small, Fl. Southeast. U.S. 3. 1903. 2 Sp. Pl. 1064. 1753. § Trait. Foug. 136. pl. 159.1705. (Parallel columns of French and Latin text.) TIDESTROM—BOTRYCHIUM VIRGINIANUM AND ITS FORMS. 301 “Monsieur Sarrazin tres-habile Medecin, scavant Anatomiste & Botaniste du Rey dans le Canada, envoya de ce méme Pais, ces deux mesmes especes & Monsieur Vaillant aussi Botaniste du Roy, & tres-expert Anatomiste. II luy manda en méme temps que les Sauvages appelloient ces deux mémes plantes I’ Herbe aux Serpens, y ayant recours d’abord qu’ils en ont esté mordus, pour remedier & leur morsure par Vappli- cation de cette Herbe.”’ It should be noted that Clayton! gave as the vernacular name in Virginia for this plant “PWern-Rattle-Snake-root,” a translation probably of the Indian name for which the French version is ‘‘l’Herbe aux Serpens ”’ as given above. Professor Underwood and Mr. Benedict? record three species of Botrychium from the West Indies: B. virginianum, B. jenmam, and B. underwoodianum. No definite locality is given for the first species, so we may conclude that it is not a local plant even in the West Indies. The second species is known only from Jamaica. The third is also known from Jamaica and Haiti (7). In Plumier’s illustration ® there is a portion of a frond which appears to represent a second species of which he speaks in the text. The latter plant he describes as having ‘‘les feuilles plus émoussGes, un peu moins découpées & bordées d’une dentelure tres-delicate””—a description which suggests B. underwoodianum or some close ally. His Osmunda Asphodelt radice received the name Osmunda cicutaria in 1798.4 It is repre- sented as having the stalk of the fertile segment inserted nearly midway between the root system and the lowermost sterile pinne (or branches), a condition which probably led the earlier authors to regard the illustration as being representative of a distinct species. Swartz * accepted the species upon the authority of Savigny and also cites Plumier’s illustration, showing the low insertion of the fer- tile segment. That such a condition actually existed in Plumier’s plant from Santo Domingo is probable, from the fact that I have observed a single instance of this sort in a Jamaican specimen. Prantl® in his review of the Ophioglossaceae refers B. cicutarvum Swartz to B. virginianum. His statement with respect to the latter species, ‘‘Pedunculus e basi laminae vel rarius e petiolo oriundus,” might lead us to believe that he had also observed instances wherein the condition represented in Plumier’s plate existed, but we have no direct proof that such was the case. BOTRYCHIUM DICHRONUM. Professor Underwood described this species’ from material col- lected by Mr. W. N. Clute at Morces Gap, J amaica, at an altitude of 1,500 meters. The description of the species fits also B. virginianum in all particulars except two. From the latter B. dichronum is said 1 Gronov. Fl. Virg. 196. 1739. 5 Syn. Fil. 171. 1806. 2N. Amer. Fl. 16!: 3-10. 1909. 6 Ber. Deutsch. Bot. Gesell. 1: 350. 1883. 3 Trait. Foug. 136. pl. 159. 7 Bull. Torrey Club 30: 45. 1903. 4 Savigny in Lam. Encycl. 4: 650. 7637°—13——2 302 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. to differ in its peculiar short panicle and ‘‘especially in its persistent sterile leaf which remains fresh until the new one is fully developed, the plant thus having two growing leaves at the time of maturity, to which allusion is made in the specific name.” All the West Indian material accessible to me has shown the above characteristics, and I have not seen any specimens referable to typical B. virgumanum. BOTRYCHIUM BRACHYSTACHYS. In describing this species ' Kunze makes the following observation: Die Kiirze der fruchtbaren Fieder, welche nur an einem von 12 untersuchten Exemplaren etwa zur Hiilfte iiber das sterile Laub hervorragte, an allen tibrigen kiirzer war, ist allerdings das auffallendste Unterscheidungszeichen; méchte aber fiir sich allein nicht zur Begriindung einer eigenen Art hinreichen. From the above it is apparent that the concepts of B. dichronum and B. brachystachys agree as to the comparative length of the fertile segment. Although the author of the latter species did not. consider this character as sufficient in itself to base a species upon, neverthe- less in all specimens which I have examined it appears constant, and all? except one specimen show the persistent leaf. All the West Indian and the Central American records for the plant show that it belongs to the Temperate Zone. The Panama material collected by Mr. William R. Maxon is by far the best and the most interesting. All the specimens show the persistent sterile segment. In one (no. 675552), collected ‘on moist forested slopes of Cerro do Lino, above El Boquete, Chiriqui, Panama, altitude 1,300 to 1,560 meters,” the fertile segment is inserted at the base of the sterile branches, measuring 20 cm. in length (the stalk included) and exceeding the sterile branches by 2 to 3 cm. only. The entire plant measures 44 cm. Another specimen (no. 675968) shows a forking fertile segment equaling in length the central branch of the sterile part, which is 22 cm. long. The entire plant measures 71 em., while the persisting frond measures 60 em. in length. The most interesting of Mr. Maxon’s specimens is one (no. 675988) presenting several peculiarities, as shown in plate 102. Only the upper portion of the plant was collected. The fertile segment emerges from the common stalk about 3 cm. below the branching of the sterile segment, thus proving in a measure the possibility of the condition represented in Plumier’s plate. It measures 41 em. in length, the distance from the point of insertion to the lowermost pinne being 17 cm. These are bipinnate and measure 17 cm. in length. The 1 Linnaea 18: 305. 1844. *U. 8. National Herbarium nos. 826306, 830774 (Guatemala, alt. 1380-1550 meters, coll. von Tiirckheim); no. 830770 (Guatemala, alt. 1,800 meters, coll. J. D. Smith); and from western Panama, no. 677413, alt. 1,000-1,300 meters, coll. Pittier; nos. 675552, 675610, 675968, 675969, and 675988, alt. 1,300-1,700 meters, coll. Maxon. Contr. Nat. Herb., Vol, 16. PLATE 102. BOTRYCHIUM CICUTARIUM (SAVIGNY) SWARTZ. TIDESTROM—BOTRYCHIUM VIRGINIANUM AND ITS FORMS. 303 upper portion (16 cm.), with the exception of a few pinnules, is sterile, a condition which I have never before observed. (PLATE 102.) EXPLANATION OF PLATE 102.—A specimen collected in humid forest along the upper Caldera River, near “Camp I,” Holcomb’s trail, above El Boquete, Chiriqui, Panama, altitude 1,450 to 1,650 meters, March 22 to 24, 1911, by William R. Maxon, no. 5569 (U. 8. Nat. Herb. 675988). Scale slightly less than 4. Mr. C. G. Pringle collected a specimen ' of nearly typical B. virgim- anum in the State of Hidalgo, Mexico. In this plant the fertile segment (including the stalk) is 31 cm. long, while the middle branch of the sterile segment is only half as long, a condition normal (at least in older plants) in typical virginianum. There is no trace of a persisting sterile leaf. The South American and Old World material in our collections in Washington is too scant and fragmentary to permit much dis- cussion. In one specimen? from Ecuador, however, the fertile seg- ment is inserted about 1 cm. below the lowermost sterile pinne, and in another specimen on the same sheet the position of the fertile segment is normal. A further discussion of extra-North American plants belongs properly to botanists who are able to avail themselves of collections rich in European and Asiatic material and to those who have had the opportunity to botanize in South America. CONCLUSION. The foregoing review of North American material seems to justify the recognition of two species, which may be distinguished as follows: Plants with persistent leaves and with fertile segments equaling or somewhat exceed- ing the sterile segment. Botrychium cicutarium (Savigny) Swartz, Syn. Fil. 171. 1806. Osmunda cicutaria Savigny in Lam. Encycl. 4: 650. 1797. Botrychium virginicum B mexicanum Hook. Bot. Misc. 3: 223. 1833. Botrychium brachystachys Kunze, Linnaea 18: 305. 1844. Botrychium dichronum Underw. Bull. Torrey Club 30: 45. 1903. Plants without persistent leaves and with long-exserted sporophyll (in older plants). Botrychium virginianum (L.) Swartz, Journ. Bot. Schrad. 18002: 111. 1801. Osmunda virginiana L. Sp. Pl. 1064. 1753. Botrychium gracile Pursh, Fl. Amer. Sept. 656. 1814. The excellent material collected by Mr. Maxon gives us a good insight into the very variable B. virginvanum group and helps us to distinguish the two forms proposed by botanists long ago. It would save much confusion and add much to the credit of botany if only such material could be considered in plant descriptions. 1U.5S. Nat. Herb. no. 461872. 2U.S. Nat. Herb. no. 50754. SPHENOCLEA ZEYLANICA AND CAPERONIA PALUSTRIS IN THE SOUTHERN UNITED STATES. By Ivar TIDESTROM. SPHENOCLEA ZEYLANICA. In 1903, a plant was sent to the United States Department of Agriculture for determination from Gueydan, Louisiana, with the intimation that it was a threatening weed in rice fields. It proved to be Sphenoclea zeylanica Gaert. and since then has been reported from time to time. This species was described under its native name, ‘pongati,” and also iliustrated by Rheede! as early as 1692 and the illustration alone is almost sufficient for the identification of the plant. Adanson * mentions it under this vernacular name, citing at the same time the figure by Rheede. In 1788, Gaertner 3 diagnosed the genus and gave a description of the only species under the name here accepted. The illustration by Gaertner shows a pentamerous flower and an ovary with 2 cells and a central placenta. The vernacular name “tembulwaenna” is also given. Jussieu‘ characterized the genus the following year under the name ‘ ‘Pongatium.” This author cites Rheede’s figure. He places the genus among Plantae incertae sedis and in addition to the generic diagnosis he observes as follows: “ Herba aquatica (Rheede Mal. II. t. 24); folia alterna; flores dense spicati terminales; horum tubus staminifer mox deciduus. Caracter ex sicco. Habitus Phytolaccae junioris. An affinis Samolo p. 97, aut Portulaceis?”’ In 1790, Loureiro® diagnosed the genus Rapinia and described one species, R. herbacea, which has been referred by nearly all the great authors to Sphenoclea zeylanica. Loureiro evidently referred the bract and two bractlets which subtend the flowers to the calyx, hence his diagnosis of the calyx: ‘“Perianthium 8-partitum, inferum: laciniis subrotundis, concavis: bino ordine, exteriori breviore.”” In all other respects the account given of Rapinia agrees well with the characters of Sphenoclea. Loureiro appears to have been the first to notice the flowers, for he placed the plant in the class Pentandria and in the order Monogynia of Linneus’s simple, provisional, but very excellent method for determining plants. In 1791, Retzius® described the plant under the name Gaertnera pangati. He distinguished the subtending bracts, which he describes 1 Hort. Malabar. 11: 47. pl. 24. 4Gen. Pl. 423. 1789. 2 Hist. Nat. Sénég. 83. 1757. 5 Fl, Cochinch. 1: 127. 1790. 8 Fruct. & Sem. 1: 118. pl. 24. f. 5. ® Obs. Bot. 6: 24. 1791. 305 806 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. as “spatulate and distinct,” from the inferior 5-lobed calyx. Mar- tius* placed the genus in an order Sphenocleaceae by itself, but just preceding the Campanulaceae. Lindley? made a suborder Sphenocleaceae, but expressing a doubt as follows: “This remarkable plant is very much like a campanulaceous genus in structure; but its exalbuminous seeds, the absence of collecting hairs from its styles, and the round subsessile anthers, seem to indicate the type of a different order; and the peculiar habit of the only known species seems to confirm the propriety of the separation.”’ An excellent account of the species has been given by Wight. In his work we find a series of illustrations from the unopened flower to the characteristic spongy, pendulous placenta, the small, oblong seed 0.5 mm. long, and the minute embryo. Figure 10 of plate 138 shows the capsule after the dehiscence has taken place. In specimens at hand this feature is very conspicuous and the remains of the capsule as well as of the placenta can be seen almost with the naked eye. Both Gaertner and Retzius had noted the circumscissile capsule. The following is a description of the species: Sphenoclea zeylanica Gaert. Fruct. & Sem. 1: 113. pl. 24. f. §. 1788; Schoenl. in Engl. & Prantl, Pflanzenfam. 4°: 61. 1889. Pongati Rheede, Hort. Malabar. 11: 47. pl. 24. 1692; Adans. Hist. Nat. Sénég. 83. 1757. Rapinia herbacea Lour. Fl. Cochinch. 1: 127. 1790. Gaertnera pangati Retz. Obs. Bot. 6: 24. 1791. Pongatium indicum Lam. Tabl. Encycl. 1: 444. 1791. Sphenoclea pongatium DC. Prodr. 7: 548. 1839; Wight, Illustr. Ind. Bot. 2: 115. pl. 138. 1850. Pongati zeylanica Kuntze, Rev. Gen. Pl. 2: 381. 1891. Herbaceous plant 1 meter high or more; stem branching, from a thick cluster of roots, 1.5 cm. in diameter at the base; leaves mostly oblanceolate, thin, light green, 9 cm. in length or less, tapering to a petiole 1 cm. long; flowers in terminal spikes, 3 to 6 cm. long, the flower subtended by one bract and 2 bractlets (?), these broad, rhombic towards the apex; calyx 5-cleft, the lobes rounded, persistent and inclosing the mature fruit; corolla minute, whitish, 5-lobed; stamens 5, sessile in the sinuses with the corolla lobes alternating; anthers 2-locular, round, dehiscing longitudinally; ovary 2-celled, the style short, the stigmas 2; ovules very numerous on a spongy, central placenta; mature capsule circumscissile, about 3 mm. in diameter; seed oblong, minute, of a light brown color. The following North American specimens have been examined: Louisiana: Gueydan (in the rice fields), Pipes; Crowley, Aldrich; southwest Lou- isiana, Dodson; Markville, W. L. MecAtee. T have also seen specimens from the Lower Orinoco, collected by Doctor Rusby and from Porto Rico, collected by Heller, by Underwood and Griggs, and by Sintenis. Recent authors do not seem to have taken cognizance of this weed, although it was already established in Louisiana and other Southern States in Doctor Gray’s time. This author gave an account of the plant in the Synoptical Flora of North America.‘ 1Consp. Regn. Veg. 31. 1835, Tlustr. Ind. Bot. 2: 115. pl. 188. 1850. * Nat. Syst. ed. 2. 238. 1836. 421:10. 1886. Contr. Nat. Herb., Vol. 16. PLATE 103. CAPERONIA PALUSTRIS ST. HIL. TIDESTROM—CAPERONIA PALUSTRIS IN THE SOUTHERN STATES. 307 CAPERONIA PALUSTRIS. Caperonia palustris is another plant which has not been noticed very often, but in this country it appears to be of recent introduc- tion. It was described and illustrated by Martyn’ prior to 1737 (probably in 1735 or 1736). Seeds were introduced into England in 1731, when they were sent to the Chelsea Garden by Houstoun. lts first mention in literature, therefore, precedes the appearance of Linneus’s Hortus Cliffortianus (1737). Martyn’s description is suf- ficient for the identification of the plant; his illustration of it is unmistakably the species in question. The description given by Martyn reads: “ Ricinoides palustre, foliis oblongis serratis; fructu hispido. Houstoun. “Caulis huic viridis, pilis albicantibus hirsutus, striatus, concavus, foliis vestitus oblongis, serratis, quatuor uncias longis, tres uncias latis, & nervis donatis conspicuis, a costa media ad latera tendentibus, & in serris foliorum singulis desinentibus. Ex alis foliorum prodeunt pediculi, Flores masculinos longa serie gestantes, exiguos pentapetalos, candidos; infra quos conspiciuntur foemininii, quibus succedunt Fructus hispidi.”’ The plate is reproduced herewith to illustrate the excellent work which was done in the dawn of modern botany. According to recent authors? the species appears to inhabit Cuba, Haiti, Guadeloupe, Martinique, the territory from Mexico to Guiana, and tropical Africa. Its characters may be given as follows: Caperonia palustris St. Hil. Hist. Pl. Brés, 245. 1824. PiatE 103. Ricinoides palustre, foliis oblongis serratis fructu hispido. Mart. Hist. Pl. Rar. 173. pl. 88. 1728; DC. Prodr. 157: 755. 1866. Croton palustre L. Sp. Pl. 1004. 1753. Caperonia castanaefolia auct., not St. Hil. loc. cit. Androphoranthus glandulosus Karst. Fl, Columb. 2: 15. pl. 101. 1862. Plants annual, moneecious, of a light green aspect, about 50 cm. high; stem branch- ing, striate, more or less beset with whitish acicular hairs; leaves on petioles 1 cm. or more in length, varying in outline from oval to linear-lanceolate, prominently nerved, sharply serrate or serrulate, the blades somet imes 15 cm. in length, sparingly pilose; peduncles pubescent, sometimes glandular, axillary, about 10 cm. in length; flowers remotely spicate, the staminate uppermost, bracted; bracts ovate, about 1 mm. long; staminate flowers minute; sepals 5, ovate-acute, 2 mm. long; petals obovate, clawed, slightly exceeding the calyx; stamens 10 or less in number, included; pistil- late flowers somewhat larger; calyx cleft nearly to the base; lobes unequal, glandular- ciliate; petals minute or none; ovary sessile, 3-locular; styles short, cleft into numer- ous filiform lobes; mature capsule hispid with glandular hairs, 1 cm. or less in diameter; seeds subglobose (1 in each cell), about $3 mm. in diameter, minutely alveolate, of a light brown color. The only North American specimens seen are from Gueydan, Louisiana.’ EXPLANATION oF PLATE 103.—Reproduction of Martyn’s plate 38, as cited. 1 Hist. Pl. Rar. 38. pl. 38. The title page of Martyn’s work bears the date of 1728, the time when the first fascicle was published. 2 Urban, Symb. Antiil. 4: 344. 1903. Contr. Nat. Herb., Vol. 16, Part 14. FRONTISPIECE. | 207 “ ON Pdos Sant 4 Pescavere \ ‘ Cojo neito' ‘ 1 aban oe oe 3 ® Mirafvores @ndho San Bernardo ewan ta Anita \ H : Jacterom on DS wet Cae ‘“ Caba"de Jan Lucos H1s° Wa 113° Wee ine 110° Map OF LOWER CALIFORNIA, SHOWING THE ROUTE OF THE EXPEDITION. PLANT RECORDS OF AN EXPEDITION TO LOWER CALIFORNIA. By Epwarp A. GOLDMAN. INTRODUCTION. The list which is here published is based on a collection of plants made by Mr. E. W. Nelson and the author in the course of general exploration in the service of the Bureau of Biological Survey of the United States Department of Agriculture. The expedition upon which the collection was obtained occupied the period from April, 1905, to February, 1906, during which the entire length of Lower California was traversed. Lack of time and of transportation facili- ties limited our collection to the more conspicuous and important species, mainly trees and shrubs. Herbarium specimens gathered by others and published records have been used to some extent, but no effort has been made to complete the list of Peninsular species, which would necessarily be very long, owing to diversified conditions of cli- mate and topography. The larger and economically more important native species are included as far as possible, the entries being accom- panied by data of collection and distribution. Although botanical collecting in Lower California began with the visit of H. M. S. Sulphur in 1839, comparatively little was known of the flora of the Peninsula until Dr. Edward Palmer visited the coast in 1887. 'T.S. Brandegee, landing at Magdalena Bay two years later, took up the work and prosecuted it at intervals in the field and in the herbarium during more than 13 years, greatly advancing our knowl- edge of the Peninsula flora, especially of the interior. The investi- gations of the various collectors have been largely restricted to par- ticular areas which, owing to accessibility, have been repeatedly revisited, while much of the mountainous interior, especially of the central section, remains entirely unknown. The work of Brandegee was centered in the Cape District from Magdalena Bay southward, including the Victoria Mountains, but he made an overland journey from Magdalena Bay to San Quintin and visited the high mountains of the northern part of the Peninsula. The other principal collect- ors—R. B. Hinds, of H. M. S. Sulphur; L. J. Xantus, Dr. J. A. Veatch, 309 310 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. Dr. Edward Palmer, Lieut. C. F. Pond, and Dr. J. N. Rose—have confined their attention mainly to thé coasts. Several important mountain ranges along the backbone of the Peninsula—the Sierra de Calamahué, Sierra de Santa Lucfa, Sierra de San Francisco, and the Sierra de la Giganta—remain unexplored, and they doubtless bear on their upper slopes many new and interesting plants. New light on distribution is also to be expected, as a number of species now known only from near the summits of the Victoria Mountains will probably be found to range farther north in the Sierra de la Giganta. Lower California was not included in the botanical treatment of Mexico in the Biologia Centrali-Americana and has been given no place in any general flora of North America, except the as yet incom- plete North American Flora. The most important papers dealing with the flora of the Peninsula are those published by Brandegee, mainly in the Proceedings of the California Academy of Sciences and Zoe; by Bentham in the Botany of the Voyage of the Sulphur; and, as scattered articles, by Rose, Gray, Watson, Greene, and Hitchcock and Chase. The flora of the Peninsula is readily separable into two main divi- sions: One, identical with that of southern California, which entering from the north occupies the northwest coast and the Sierra del Pinal and San Pedro Mértir mountain regions, comprising species which disappear rapidly to the southward, a few reappearing on the summits of the high mountains in the Cape District south of La Paz; the other, a more austral flora, derived from or related to that of the adjacent mainland coast of Mexico, occupying the entire southern part of the Peninsula except the summits of the higher mountains and extending northward in a narrowing strip east of the San Pedro Martir Moun- tains. Brandegee* states that the greatest change in the flora takes place in about latitude 28°. While a rough division may be made in the vicinity of this parallel, many austral species reach much farther northward along the coast of the Gulf of California. The region as a whole is of unusual interest, owing in part to its configuration and to the inclusion within its borders of these widely differing floral areas. The higher mountains are crowned by familiar appearing forests of oak and pine. In the more arid desert sections a number of species in adapting themselves to their environment have developed into monstrous forms which so prevail as to give the landscape an aspect of unreality. Several remarkable genera seem to be peculiar to the Peninsula and numerous species belonging to genera ranging widely in tropical America are here rather narrowly restricted in range. * Southern Extension of California Flora. Yoe 4: 199-210. 1893. GOLDMAN——PLANT RECORDS OF LOWER CALIFORNIA. 811 The subjoined list includes 3 species described in the present paper and 19 already described from material collected by this expedition: Page. Agave nelsoni Trel. Rep. Mo. Bot. Gard. 22: 61. pl. 65-67. 1911. Type from San Fernando............2-2- 222 ee ee eee eee eee ee eee eee eee eee eee 318 Agave cerulata Trel. Rep. Mo. Bot. Gard. 22: 55. pl. 45-47. 1911. Type from Calmallf.............---- 2-2-2202 ee eee eee ee eee eee eee ee 318 Agave promontorii Trel. Rep. Mo. Bot. Gard. 22: 50. pl. 85-37. 1911. Type from Sierra de la Laguna........--..-2.- 22-02-22 202 eee ee eee eee eee eee 318 Agave vexans Trel. Rep. Mo. Bot. Gard. 22: 62. pl. 70-72. 1911. Type from near El Potrero, about 30 miles southwest of Mulegé..................++ 319 Agave goldmaniana Trel. Rep. Mo. Bot. Gard. 22: 49. pl. 29-31. 1911. Type from Yubay.....--.--------- 2-2-2 e eee eee eee ee eee eee eee eee eee 319 Quercus brandegei sp. nov. Type from Rancho El Paraiso, at northwest base of Victoria Mountains............-------- 2-2-0 eee eee eee eee cence eee 321 Quercus idonea sp. nov. Type from near Rancho San Bernardo, 13 miles west of Miraflores, southern slope of Victoria Mountains...............-..- 321 Quercus devia sp. nov. Type from near La Chuparosa, a spring at 1,500 meters altitude in Sierra de Ja Laguna.........-.---------- 2-0-2222 e eee 322 Cassia goldmani Rose, Contr. U. 8. Nat. Herb. 10: 98. 1906. Type from near El Potrero, about 30 miles southwest of Mulegé......--.----------- 335 Brongniartia peninsularis Rose, Contr. U. 8. Nat. Herb. 12: 268. 1909. Type from near El Potrero, about 30 miles southwest of Mulegé........-.- 337 Elaphrium epinnatum Rose, N. Amer. Fl. 25: 243. 1911. Type from Cape San Lucas.......-.--- 2-2. + 2-222 eee eee eee ee ee eee erence eee 339 Elaphrium goldmani Rose, N. Amer. Fl. 25: 256.1911. Type irom between Matancita and La Cruz.......----------- 2-20 ee eee eee eee eee ee ee eee ee 340 Manihot chlorosticta Standl. & Goldm. Contr. U.S. Nat. Herb. 18: 375.1911. Type from San José del Cabo... .....----2----- 2+ +222 eee eee eee seer tenes 343 Schmaltzia ribifolia Greene, Leaflets 2: 156. 1911. Type from San Matias Pass (altitude 1,140 meters) at north end of San Pedro M&rtir Mountains... 345 Ceanothus goldmanii Rose, Contr. U. 8. Nat. Herb. 12: 284.1909. Type from La Huerta, at west base of Sierra del Pinal...........-.------------- 347 Ceanothus submontanus Rose, Contr. U. S. Nat. Herb. 12: 284. 1909. Type from near Alamo.......-.------ +--+ -+2-- 2222s cece renee eee e eens 347 Arbutus peninsularis Rose & Goldm. Contr. U. 8. Nat. Herb. 13: 312. 1911. Type from near La Chuparosa (altitude 1,350 meters), Sierra de la Laguna.......--- 2-2-2222 eee eee eee eee ee eter eee eee eens 358 Mesosphaerum insulare Standl. & Goldm. Contr. U.S. Nat. Herb. 13: 375. 1911. Type from Espiritu Santo Island....-..........------------------- 362 Franseria carduacea Greene, Leaflets 2: 156. 1911. Type from Tinaja de Santana, 35 miles north of San Ignacio...............-----------+--------- 369 Porophyllum confertum Greene, Leaflets 2: 155. 1911. Type from Cerralvo Island .........----- 2-2-2 eee ee eee eee eee eee eee eee 370 Senecio goldmanii Greene, Leaflets 2: 156. 1911. Type from Rosarito, near the west coast....--.--.2-2..---- 2-2-2 ee ee ee eee eee eee 370 Viguiera chenopodina Greene, Leaflets 2: 154. 1911. Type from between Santo Domingo and -Matancita...............0.- 2-2 e eee nee eee eee eee eee 371 For aid in the determination of species, I am especially indebted to Mr. T. S. Brandegee, the well-known authority on the botany of the region, his special field of study. Others who have given the 812 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM, benefit of their knowledge of special groups are Dr. J. N. Rose, Mrs. Katherine Brandegee, Mr. Frederick V. Coville, Dr. William Tre- lease, Mr. George R. Shaw, Mr. Paul ©. Standley, Mr. O. F. Cook, Mr. Carleton R. Ball, Dr. E. L. Greene, and Miss Alice Eastwood. ANNOTATED LIST OF SPECIES. ROCCELLACEAE. Orchilla Family. Rocella sp. ORCHILLA. The native name of this lichen is “‘orchilla.”” It was formerly the source of a large and profitable industry in the vicinity of Magdalena Bay, on the west coast of the Peninsula. It was shipped to England, where from the raw material fast dyes of several colors were extracted. When cheaper aniline dyes came into general use, however, the business became unprofitable. A recent demand for vegetable dyes may result in a revival of the industry. The plant seems to be confined mainly to a narrow strip of shore line, extending from near Magdalena Bay north for about 150 miles, Orchilla grows in thick drooping, mosslike fringes along the branches of desert shrubs and trees near the coast. PINACEAE. Pine Family. Abies concolor (Gord.) Parry. WHITE FIR. Occurs rather sparingly on the cooler slopes above 2,250 meters in the San Pedro Martir Mountains. Collected at Vallecitos, July 15. Cupressus guadalupensis S. Wats.(?). GUADALUPE CYPRESS. A single cypress tree was found by us in a notch at about 2,700 meters altitude on the crest of the San Pedro Martir Mountains several miles east of Vallecitos, July 15, and from it a branch bearing ripe cones was collected. From near the spot, which was reached on horseback, we had a splendid view of Santa Catalina Peak, the high- est of the San Pedro Martir Mountains, across a canyon to the southeast, and at our fect the east slope of the range broke away abruptly to the desert over 2,100 meters below. In the absence of specimens for comparison we assume ours to be the same as a species collected in the San Pedro Martir Mountains by Townsend and Anthony, regarded by Dr. C. 8. Sargent as somewhat different from the type of guadalupensis, but not separable from it.! | It seems very unlike C, goveniana, a widely ranging species, which approaches the Lower California boundary, and may prove to differ from guadalupensis, which grows abundantly at lower elevations on the more humid, fog-enshrouded slopes of Guadalupe Island. Juniperus californica Carr. CALIFORNIA JUNIPER. CEDRO. Abundant over many slopes along our route, beginning at about 450 meters alti- tude a few miles east of Ensenada, thence up through San Rafael Valley to the sum- mit of the Sierra del Pinal. It was also noted near El Alamo and in Trinidad V. alley, and farther south at 990 meters along the road from Rancho Santo Tomés to San Antonio, on the west slope of the San Pedro Martir Mountains.