fan Ce Ve bee na on eds 10 Waal wp eat oa Cues toume 1 OV Fo ote Pee ea wea ro | ‘ re Pepe an ad yeas a en ee ee ee LA tee a ee Ay eng n P Psteavyer Wy aged kt onan de eo ab ee ee eet peur ' sae bP ase, 8 tsk ame ag Hoey tere ewe Dee CC ms m4 tue i ger o lena rs eT PSAP PPT Pee ee ee ee pe > co Se TT CS A hoy SW ee en te A eel CC Oe eC ee ob Sa WT Pl ke ant Wii acl Be gage aa hea ats: fe pk bo a Ot de ie aeee ato Egan errr Sob e tern Pat eo 0 plate cn . be Pw EM me wa Ge iat Om tiaaeu al Be padn oT Wedoate pra cit ee | abe sw @ bre i tan lee LET NG UR OMA A EHO oatd Betokee teks rita i ee De a Be 8 kD SNA: beste ae we FN tee megs qe Wa PE 9 ee tp ia Re ney poe pang Wb ep euge eerie oem Me 4 Lid Hee ie me tithe e Bea bop tory ene gi wus oF Pas wa yee wnena 2nd shot eine J serif Fish vce ae UN Ot pe ‘ ehh p Age ED PO one ae Se Pe SI, sia oe a We seedorerieny kr au ud Cae ree 2 te bhd deep faa ‘ BL: ’ Poses ween ries 7 OT We ney een " Heb eb ras i 4 tard mae tee of ae BH NR Bogh come a agen. et tpite ers anise Beotday ghap gosh, goer $e rate re a Oe Se ere Ba ees Oar ire oa" Mery so ” " te ry " ” oe ” ww yey eo weno g woe nen bee si wi i 7 - Ne eran 2 eel MADRONO. A WEST AMERICAN JOURNAL OF BOTANY VOLUME XIII 1955-1956 BOARD OF EDITORS HERBERT L. Mason, University of California, Berkeley, Chairman EDGAR ANDERSON, Missouri Botanical Garden, St. Louis LyMAN BENSON, Pomona College, Claremont, California HERBERT F.. COPELAND, Sacramento College, Sacramento, California Joun F. Davipson, Department of Botany, University of Nebraska, Lincoln Ivan M. JounstTon, Arnold Arboretum, Jamaica Plain, Massachusetts Mivprep E. Marurias, Department of Botany, University of California, Los Angeles MArRIoN OWNBEY, State College of Washington, Pullman Ira L. Wiccins, Natural History Museum, Stanford University, Stanford, California SECRETARY, EDITORIAL BOARD ANNETTA M. Carter, Department of Botany, University of California, Berkeley BUSINESS MANAGER AND TREASURER RicHarpD W. Hom, Natural History Museum, Stanford University, Stanford, California (1955) Matcoro A. Noss, Carnegie Institution of Washington, Stanford, California (1956) Published quarterly by the California Botanical Society, Inc. 2004 Life Sciences Building, University of California, Berkeley Printed by Gillick Press, Berkeley, California To you, Miro SAMUEL BAKER, one-time President of the California Botanical Society, student of the genus Viola, and inspiring teacher, we dedicate this volume of Madrono. There are few teachers who can claim so great an influence upon their community as you. Of the scores of stu- dents who have gone through your classes, many have retained an interest in plants as a hobby and a goodly number have been inspired to a pro- fessional career in botany. Through your inspirational teaching you have broadened enormously the foundations for the future happiness of your students. You have advanced our knowledge of the botany of the coastal northern California region in a manner that has provided a permanent record of documented facts. We congratulate you and wish you many more years of achievement. CONTENTS Edward L. Greene and Howell’s ‘Flora of Northwest America’’— PEE TRITON ELS LUG Circemeee tee Be The hae Pea eee eM SSS LE eo ag ee el con en barca 1 Revision of the dioecious amaranths—Jonathan Sauer .........20....20.02200c0000c000v0eeeeeeeee- 5 The status of Pseudohomalomena pastoensis—M. R. Birdsey .............2222....2200-02200--- 47 | RS(SSUALERIIAS 0 kee NG a ag PA PS Pep a 48, 78, 175 Studies in North American Volvocales. I. The genus Gonium—WM. A. Pocock ...... 49 Flora of the Crested Butte Quadrangle, Colorado—Jean H. Langenheim ................ 64 PNFOtie Span SING WS stg notes Sects eli og yep netstat ati. dche tee: 80, 112, 176, 207, 268 EEnestebLowl: bapCOCK GL... SECOUINS: 3.1.05. scccsetccsaivt ewes hoon eee ee ete 81 Some considerations of the genera Echinocystis and Echinopepon in the United States and northern Mexico—Kenneth M. Stocking ...........22200..0220000022--.. 84 Mosses of California V. Pterigoneurum ovatum and P. subsessile— LOADS BEE GY SCONE! (Sr Sena eNO aE ERP aE ee 100 Chromosome numbers in Mentzelia (Loasaceae)— Henrys: Enompson ana Harlan Lewis 2... sik cee 102 Chromosome counts in the section Simiolus of the genus Mimulus (Scrophulariaceae)—Robert K.. VicREry, J7-é ..-.ccc2ccacciccsecseccnecctensceenseerenseneeccceesseees 107 Notes on Nevada Mimulus—Gabriel Edwin ..........00....ccce00ccccceeccccenecceteceeeeenneeceeeeeeeerens 110 Some taxonomic and ecological considerations of the genus Marah (Cucurbitaceae)—Kenmneth: Ms SUOGRING fcc coco ocak cnc setae bectcuectnonnsendeese sactese a3 Observations on Prasiola mexicana, a freshwater alga of unknown relationships—Herbert F. Copeland .............-20c.1.ccneccessescnnceceneeoeseeceneccensessesenecenenes 138 A new species of Bouvardia (Rubiaceae) from Baja California, WECSICO— =A CLL: COTECT ere ede aes ft cots, «aah decd eee 140 Biosystematics of Helianthus debilis—Charles B. Heiser, Jr. -..o....22000220000200--- ee se 145 A new Trifolium from Oregon—Helen M. GilRey _........c..cssccccceceecceseeceneeeeeceeeneceeeess 167 A new record of dwarf mistletoe on lodgepole and western white pine— OO OTIC E I Eas asta eee kN 288 5 ie Mera 20 0 a, ON ie, Ucn eee eee 170 Prosopis globosa Gill. in Baja California—Robert L. Dressler ............0.2.02.000000000000- 172 Studies in the Capparidaceae II. The Mexican species of Cleomella: taxonomy and evolution—-Hugh H. [tis 2.0.2.0... cece ccc ceeeceneeeeeeeeccneeteeeeeeeeeeeeee 177 Two fungi associated with a microcyclic rust, Coleosporium crowellii Cummins, on needles of Pinus edulis Engelm. in Arizona—Paul D. Keener .............2--........ 189 A natural hybrid, « Adiantum tracyi C. C. Hall—Warren H. Wagner, Jr. ............ 195 Documented chromosome numbers of plants ....20.....0.....2000-c22e0cceeeeseeeececeeeeeesseeereneeeees 205 California Botanical Society, Report of the Treasurer for 1955 0.0.0... eeieeeeeeeeeee eee 208 Notes on the Phacelia magellanica complex in the Pacific Northwest— ANd CORI EG VT 4 ON A SAR eer att PaO re ope eT Scot Ye CER EE ee RS el RRO aS See 209 The distribution of Quercus boyntoni—Cornelius H. Muller .........0..00..2...00000000000----- 221 On the generic limits of Eriophyllum (Compositae) and related genera— Sherwin Carlquist .......... Le ann eee ana ty See omit On cee aaeecet See ere eS 226 ill New species of Elatine in California—Herbert L. Mason ..00...0....0.020002000c000ccc000000----- 239 Notes on Malvaceae. VIII. Eremalche—Thomas H. Kearney ..........0..20.2c.0c200-00-000000+- 241 Muhlenbergia Brandegei, a new species from Baja California, Mexico, and its relationship to Muhlenbergia biloba—Charlotte G. Reeder .................-....00+--- 244 Phytoserology versus genealogy in Zea Mays— Joke Daevidsoniand LL GLUOMpson.. x 2 ee ins co, eee er ee ear 252 Joseph Burke in 1853—-R. Kent -Beatizec Se ae ee ved ee ee 259 A new Gossypium from Michoacan, Mexico—Howard Scott Gentry .............0..--+-- 261 A summary of the nomenclature of Douglas-fir, Pseudotsuga menziesii— Viedamidr J: Ki gina on ee ee ee cy 1 en 265 ENG OX ns ste. 8s Aaa Ss 2 Ae Sen reat SN ote eee IN ee actrees 269 ERRATA Page 46, line 35: for F. W. Stafleu read F. A. Stafleu. Page 86, line 14: for lectotype read neotype. Page 90, line 46: for Type read Lectotype. Page 92, line 1: for Type read Lectotype. Page 96, line 5: for lectotype read neotype. Page 101, line 1: for meters read feet. Page 121, line 32: for sub-glabrouse read sub-glabrous. Page 124, line 8: omit guadalupensis. Page 126, line 8: for lectotype read neotype. Page 129, fig. 5: for M. watsoni read M. watsonii. Page 130, line 20: for lectotype read neotype. Page 130, line 26: for Type read Lectotype. Page 132, line 38: for lectotype read neotype. Page 134, line 27: for Type read Lectotype. Page 137, line 17: for triangularr read triangular. Page 197, line 9: for unmistakeable read unmistakable. Page 198, line 26: for Adiantum tracyi read * Adiantum tracyi. Page 231, Table 1, lines 17-18: for Gonzales, San Benito County, read Parks Valley, San Benito County. Page 235, legend for figs. 19-21, line 3: for Jepsen read Jepson. Page 238, at end of article add: The Claremont Graduate School, and Rancho Santa Ana Botanic Garden, Claremont, California. Page 239, line 37: omit latter. Page 240, line 1: for 2—4 altae read 2-4 cm. altae. Page 240, line 41: for rleated read related. iv i. PS. fy py j wr wale 5 MADRONO VOLUME 13, NUMBER 1 JANUARY, 1955 Contents EpWARD L. GREENE AND HOWELL’s “FLORA OF NoRTH- WEST AMERICA,” Erwin F. Lange it REVISION OF THE DIOECIOUS AMARANTHS, Jonathan Sauer 5 THE STATUS OF PSEUDOHOMALOMENA PASTOENSIS, M. R. Birdsey 47 REvIEW: Richard W. Pohl, How to Know the Grasses (C. Ritchie Bell) 48 ZNNTHSON gx FEB 18 i¢55 te f [Pr l» ae g LIBRARY E A WEST AMERICAN JOURNAL OF BOTANY -UBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY MADRONO _ A WEST AMERICAN JOURNAL OF BOTANY Entered as second-class matter at the post office at Berkeley, California, January 29, 1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price $4.00 per year. Published quarterly and issued from the office of Madrofio, Herbarium, Life Sciences Building, University of California, Berkeley 4, California. BOARD OF EDITORS HERBERT L. MAson, University of California, Berkeley, Chairman EpcAR ANDERSON, Missouri Botanical Garden, St. Louis. Lyman BENSON, Pomona College, Claremont, California. HERBERT F. COPELAND, Sacramento College, Sacramento, California. Joun F. Davipson, University of Nebraska, Lincoln. IvAN M. Jounston, Arnold Arboretum, Jamaica Plain, Massachusetts. Mitprep E. Marutas, University of California, Los Angeles 24. Marion OWNBEY, State College of Washington, Pullman. IrA L. Wiccrns, Stanford University, Stanford, California. Secretary, Editorial Board — ANNETTA CARTER Department of Botany, University of California, Berkeley. Business Manager and Treasurer — RicHARD W. HoLm Natural History Museum, Stanford University, Stanford, California. CALIFORNIA BOTANICAL SOCIETY, INC. President: Lincoln Constance, Department of Botany, University of California, Berkeley, California. First Vice-president: John Thomas Howell, California Academy of Sciences, San Francisco, California. Second Vice-president: Mildred E. Mathias, ~ Department of Botany, University of California, Los Angeles, California. Recording Secretary: Mary L. Bowerman, Department of Botany, University of California, Berkeley, California. Corresponding Secretary: G. Thomas Robbins, Department of Botany, University of California, Berkeley, California. Treasurer: Richard W. Holm, Natural History Museum, Stanford University, Stanford, California. Annual membership dues of the California Botanical Society are $4.00 for one person or $5.00 for two members of the same family. Either type of membership carries with it one subscription to Madrofio and all other privileges of the Society. Dues should be remitted to the Treasurer. General correspondence, changes of address, and applications for membership should be addressed to the Secretary. EDWARD L. GREENE AND HOWELL’S “FLORA OF NORTHWEST AMERICA” ERrwWIN F. LANGE Numerous writers have referred to the fact that Thomas J. Howell, pioneer Oregon botanist, published the first flora of the Northwest under very difficult circumstances by himself setting all of the type by hand and having his friend, Martin Gorman, read the proofs. A fact which no writer has referred to is that without the help of Edward L. Greene the book would probably never have been completed. Only one writer, Alice East- wood (Erythea 6:58—60. 1898), in writing a review of the second fascicle of Howell’s Flora refers to the style of nomenclature as that centering around Greene. The correspondence of Thomas Howell to Edward L. Greene reveals an interesting story of determination on the part of the pioneer botanist, destitute and with but three months of formal schooling, to complete a badly needed flora for the only great section of the country not treated by such a work. The first reference to his book occurred early in April, 1892, when Howell wrote to Greene: ‘I have been very busy writing my Flora of Oregon, Washington, and Idaho of which I enclose a specimen page. I hope to have it published about the first of January next.” The specimen page seems to have caused a quick critical reply from Greene for later in the month (both dates are difficult to read) , Howell wrote: Thank you for your criticism of my work but it should have been stated in a former letter that it is not yet being printed. The sheet I sent you was set up from an unrevised copy and was merely intended to base calculations as to the number of pages and cost of publishing, and was not submitted to an expert proofreader for correction. It will all have to be set over agane, and I will say here there will be no bad spelling or punctuation in it... . The other corrections are such as any expert proofreader would notice. Not more than a half dozen sheets of this has been dis- tributed, so they will not do much harm. On December 12, 1892 he wrote Greene that his financial troubles ‘“‘by which I will probably loose all my savings has run me nearly crazy for the past eight months.” The next letter referring to the book was dated April 24, 1896, in which Howell mentioned that except for the slowness of the printers the whole of Part I would have been ready for distribution. The following month, May 6, 1896, he wrote: Nothing would please me more than have you pass upon every page of my pro- posed Flora before it goes to press. Can you point out the way this can be done. Or can you show me how it can get published at all? There is no one here that can do the work except under my direct supervision and then they want double price for doing it, and want their pay in advance and this I am unable to meet for I have been reduced to poverty by some unfortunate investments. Maprono, Vol 13, No. 1, pp. 1-48, Feb. 4, 1955. 2 MADRONO [Vol. 13 As to the pages already printed they will probably never be distributed in their present form, for the parties that undertook to do the printing have just gone back on their contract, and refuse to do any more of it on any terms that I can meet. This leaves me on the verge of despair for the manuscript that I have represents ten-years of the best part of my life, and to loose it now looks to me like throwing away life itself. Howell mentioned in the same letter that he was going to try to get the American Book Company to publish it but was afraid he would have to alter the book too much to be satisfactory. He sent Greene some printed pages of the book and commented on the many errors in printing. Greene wrote immediately to Howell and offered financial help of which Howell seemed very appreciative and wrote at length concerning his pres- ent experience and the prospects of his flora. He told Greene that 28 pages had been printed when the publisher gave up the contract (no reason was given but it is believed the printer was unable to read Howell’s poor hand- writing of technical terms and set the copy correctly). Howell estimated that to continue the book as the sample pages would require from 600 to 700 pages and could be published with less capital by getting out parts of about 100 pages each. He promised Greene all the proceeds from the sales and liberal pay for his services if he could work out any way to publish the book. In letters of June 11, and September 14, 1896, he thanked Greene for his offer to help with the book and promised that Greene’s wishes would be strictly adhered to. Howell reported that it would be impossible to have the American Book Company publish the Flora. The American Book Company agreed to invest $1,000 in the book if Howell could invest $2,000 for the preparation of the ‘“‘Electraplates” but being unable to secure that amount the situation seemed hopeless. If Howell could raise the $2,000 the American Book Company contemplated printing 2000 copies which would bring $3,200. With the book company taking out the first $1,000 and then half of the remainder there would be nothing left for Howell. In conclusion Howell wrote, ‘““You once made me a proposition to loan me a few hundred dollars to publish the book with, will you let me know how much you can let me have for the purpose and on what terms.” A week later, September 21, 1896, Greene answered Howell’s letter raising a number of questions. To these Howell replied October 1, 1896: I see you still have the impression that I intend to have illustrations in my pro- posed book, which I wish to assure you is not the case for I never had any intention of illustrations at all; the plates spoken of are Book plates that is electroplates of the text and not illustration plates. There are two ways of making a smaller book of it. One is to condense the descriptions and thereby make them worthless. The other is to leave out a large part of the species and make an incomplete work. There are about 3500 species of plants in the territory that I propose to cover, and nine tenths of them grow in Oregon, so you see that it will not reduce the book much to reduce the territory. I have made a careful estimate and find that I can have 1955] LANGE: HOWELL’S FLORA 3 1000 copies of 600 pages published here for $1,000. I can do this so cheap because I have a pretty fair printing outfit of my own and I have orders on hand now to assure the sale of 1000 copies in less than two years at $2.50 per copy. With $500 I could get the book out next spring, but I have no way of getting that amount now, and I would not think of letting you put one dollar into this book that you do not feel certain that you will get it back agane with interest. In a letter to Greene dated October 13, 1896, Howeil again assured him that the book would sell readily and that he did not immediately need $500 but could get started on $50 or $100. He guaranteed Greene that every page would be sent him for criticism before going to press and would be printed with his own type. With the letter he sent 28 pages of proof sheets for correction. His proof sheets were imperfect and drew some criticism from Greene but in reply Howell wrote November 26, 1896, “I am very thankful for your notes and criticisms and shall profit by them to the fullest extent.” On December 21, 1896, Howell again suggested getting the book out in parts in order to get it published with less capital. Since he was held up 16 days waiting for his proofs to arrive from Greene, he suggested sending his manuscript for criticism and correction so that the work would pro- ceed faster. Pages of manuscript and proof were sent and more were mailed on December 26, 1896. That Greene was very critical of the manuscript is evident from How- ell’s reply of January 11, 1897: Your letter with the Mss. was received yesterday, and to say that I am surprised at what you say is stating it mildly; for I certainly do not deserve the harsh things that you say about me. You seem to forget that most of my copy was written some seven years ago and was made up different from what this is, for a year ago I went over it and made it conform with Mr. Robinson’s work, not because I particularly liked it but because as a whole I liked it a little better than the way I had it. But you seme to think that I must know what you are doing in advance of its publication, for I had not read Pittonia when I sent you the copy and certainly could not have known what was in it, as my letter of a day or two later will show. Now if you had went at the copy good naturally and just changed names when you thought they were incorrect you would have saved yourself much trouble. ...I do not feel dis- posed to quarrel with you, but you will remember that my library is very small and that I live many hundred miles from any good one, and therefore work at a great disadvantage. Other references to the difficulties in completing the book were: January 16, 1897 I have been getting along very slow with the work, but have now got to where I can go faster and will have the first 100 pages ready to issue about the first of February. I am also entirely out of money. January 31, 1897 Your note of the 23d inst. with Draft for 70 dollars is at hand and I enclose note for the two as you requested. If I stated that I would have 100 pages out the first of February it was a mis- take it should have been the first of March. 4 MADRONO [Vol. 13 February 25, 1897 With this I send you another 16 pages of the book and I have a good start on the next 16 making 112 in all. With that I shal have a few hundred coppies bound and put on the market, and I expect a little revenue from it... . I shall send you manuscript in a few days for critical examination to avoid making any more blunders. I expected some money of my own to come in by this time to meet maturing bills, but it has not yet arrived, and I fear now it will not very soon and I am gitting nearly broke agane. March 11, 1897 Your letter the 5th inst, with Draft was received today. I had March 10th printed on my title page, but on account of delay at the bindry I have been unable to send out part I yet, but will do so in a day or two. I wish now that I had made the date March 15th. March 20, 1897 I have been unwell lately and unable to do much and am but little better now. I will soon begin setting the type for another part which I hope to have ready early in May. May 29, 1897 Your letter with the draft for $40 was received a short time ago. I am extremely sorry that I was unable to return it at once, but I had run some bills that I had to pay, and I had no other means of paying them now, so I had to use it. Only a few scattered letters between 1898 and 1904 have been pre- served and these contain but few references to the publication of Howell’s Flora. April 16, 1902 Mr. Gorman informs me that you lack fascicle 4 of my Flora so I send you it to you under separate cover. October 22, 1902 I am drove nearly to death at the present trying to make a living and get my Flora done at the same time. While the preceding excerpts of Howell’s letters reveal the tremendous hardships under which the first Flora of the Northwest was published, the account is not complete. The Howell-Greene correspondence also contains numerous references to problems relating to the identification, determina- tion and naming of plants to be included in the book. It is not the pur- pose of this paper to include such material. Howell’s Flora was completed in 1903 after seven separate fascicles were bound and distributed. Howell reported to Eva Emory Dye, Oregon historical writer, that a thousand copies were printed but only three hun- dred were bound. An unknown additional number were bound after Howell’s death late in 1912. The writer is indebted to Albert L. Delisle, curator of the Greene- Nieuwland Herbarium of the University of Notre Dame, for making available the Howell-Greene correspondence. Portland State Extension Center, Oregon State System of Higher Education. 1955] SAUER: AMARANTHUS 5 REVISION OF THE DIOECIOUS AMARANTHS! JONATHAN SAUER In the great genus Amaranthus", the relatively small number of dioeci- ous species, including those formerly assigned to other genera which are here united with Amaranthus, form a fairly distinct and coherent group. In contrast to the monoecious amaranths, which are represented by en- demic and pandemic species in every continent, all the dioecious ama- ranths are natives of a single continent, North America, and none has migrated very far from its original home. An endemic Peruvian species (A. Haughtii Stand]. Field Mus. Bot. Ser. 11:149. 1936) is described as dioecious, but examination of the type and all other available collections shows it is actually monoecious. The dioecious habit, which is extremely constant, is the only morpho- logical character distinguishing this group as a whole from all the monoeci- ous amaranths. However, all the dioecious species share a combination of characters which occurs in only a small minority of the monoecious species, namely pentamerous staminate flowers together with complex terminal inflorescences, often called spikes in the literature but tech- nically thyrses. It is conceivable, although by no means certain, that the group is a natural one, in the sense of having a single origin with subse- quent speciation independent of the rest of the genus. Be that as it may, no formal section or sections for the dioecious species are proposed here. Considering the lack of any widely accepted subdivisions of other parts of the genus, formal infra-generic disposition of the dioecious species does not appear urgent or useful at the present time. Hybridization is given little attention in this paper, although it is a common and widespread source of variation in the group. In fact it was natural “intergeneric” Amaranthus < Acnida hybrids that first attracted my attention to this group; the revision was initiated primarily because clearer understanding of the species involved was prerequisite to under- standing of the hybrids. The study of individual variants in the mongrel populations, which is still in progress, is being handled as a separate problem. The present paper has the limited objective of characterizing the basic entities in the group. 1 This investigation was aided by a grant from the Wisconsin Alumni Research Foundation. Publication cost cf extra pages beyond the usual MAprROoNO limit was borne by the author. 2 The English text of the International Code of Botanical Nomenclature (Lan- jouw et al., 1952, p. 44) states that Amarantus is the original spelling, in contradiction to the French text (p. 193) and the facts. In the same volume (p. 101), Amarantus is proposed for conservation over Amaranthus, while Amaranthaceae (with Amaran- thus as the type genus) is included in the nomina familiarum conservanda (p. 66). 6 MADRONO [Vol. 13 SYSTEMATIC CHANGES The taxonomic treatment presented here differs considerably from Standley’s (1917), the only other modern treatment of the group. There are ten species in place of Standley’s sixteen and all are now assigned to Amaranthus, rather than being divided among three genera: Amaranthus, Acnida, and Acanthochiton. Discussions of the revised species definitions are given below in the sections on individual species, but some explanation should now be made of the revised generic grouping. Acanthochiton is a monotypic taxon that was originally described (Tor- rey, 1853, p. 170) under the curious belief that it was a new genus of the Chenopodiaceae, where it could hardly have been included in any other genus. Torrey himself suggested that it might better be assigned to the Amaranthaceae, where later workers have invariably placed it. The pecu- liar bracts provide the one clear-cut character distinguishing this taxon from Amaranthus. This difference, although conspicuous, is only quanti- tative and seems no more fundamental than the modification of bracts to spines in the monoecious Amaranthus spinosus, for which a separate genus has never been found necessary. Moreover, the peculiar bracts de- velop only on the pistillate plants, while the staminate plants are within the range of variation of Amaranthus species in all morphological char- acters. The affinity of Acanthochiton to certain species of Amaranthus is so obvious that I believe it has been left as a distinct genus only because the plants are rather rare and have attracted little attention. Addition of this taxon to Amaranthus requires no change in the usual definition of the genus. Acnida, like Amaranthus, appeared in Linnaeus’ “Species Plantarum” of 1753 long before most of the North American amaranth species were discovered. Acnida originally included a single species, A. cannabina, which differed from all of Linnaeus’ Amaranthus species not only in va- rious quantitative characters but also in two clear-cut qualitative char- acters: dioecious habit and pistillate flowers with no perianth. Subse- quently, as more and more species became known, this discontinuity gradually blurred. Separation of the genera on the basis of dioecious or monoecious habit was abandoned when new dioecious species were dis- covered and assigned to Amaranthus because of the conspicuous perianths of their pistillate flowers. Absence of this perianth has become the sole criterion for separating Acnida from Amaranthus; no comparable key character was applicable to the staminate plants. This single remaining criterion is largely a fiction: the presence of a pistillate perianth in certain Acnida species was reported nearly sixty years ago (Uline and Bray, 1895, p. 156) but the fact has generally been ignored or glossed over in later literature. Pistillate tepals are truly absent in some species assigned to Acnida; in some they are very small and irregularly present, so that they might be dismissed as mere rudiments, but in other species they are well- developed and constantly present. Although this perianth may consist of 1955] SAUER: AMARANTHUS 7 only one or two tepals, the same is true of some monoecious Amaranthus species. In short the two “genera” can be separated only by using intricate character combinations rather than by a clear-cut dichotomy. Moreover, they are not easy to distinguish subjectively on the basis of general aspect. The absence of a clear-cut morphological discontinuity is accompanied by the absence of an absolute reproductive barrier. The existence of natural Acnida X Amaranthus hybrids has already been mentioned. This evidence is reinforced by a magnificent series of experimental crosses between certain species of Acnida and Amaranthus (Murray, 1940). These hybrids are highly sterile, but no more so than some of Murray’s crosses between two species of Amaranthus. The balance of the evidence seems to show that the affinity between Acnida and Amaranthus is too great to justify maintaining them as sep- arate genera, and the two are here united. Although neither name has priority, Amaranthus has had far more general currency and is the obvious choice for the enlarged group. This union requires that Amaranthus be described as having 0 to 5 tepals in the pistillate flowers, rather than 1 to 5 as formerly. Otherwise no change is required in the usual circumscrip- tion of the group. NOTES ON MAPS AND CITATIONS All specimens cited and mapped have been assigned to species pri- marily on the basis of microscopic flower characters. The great majority of specimens show the constantly repeating morphological patterns char- acteristic of pure species; these specimens were annotated with the usual type of label and the localities were plotted on the maps as dots. A respect- able minority of the specimens are individual variants, presumably the result of crossing between dioecious species. As is so often the case in hybrid populations found in nature, exact intermediates between the two parental species are rare; an individual usually has a preponderance of traits of one or the other species. These atypical plants were assigned to the species which they resemble most closely, annotated with special labels indicating their probable hybrid origin, and plotted on the maps with crosses. In cases where both typical and atypical specimens are known from the same locality the cross symbol is omitted, so that the total occurrence of atypical plants is somewhat greater than the maps indicate. No attempt is made to separate typical and atypical specimens in the citations since both are commonly included in a single collection. A certain class of dioecious specimens has been annotated but excluded from maps and citations. These are highly sterile plants, probably mostly Fi hybrids between monoecious and dioecious species, which cannot prop- erly be assigned to any one species. In the citations, names printed in upper case under each country refer to civil divisions of two ranks: the state, province, or department at the start of each paragraph, followed by subdivisions such as county, parish, or municipio. An asterisk preceding the name of a country or division MADRONO [Vol. 13 PISTILLATE FLOWER PARTS Left to right: UTRICLE, with seed outlined by broken line. Horizontal line across middie of utricle indicates circumscissile dehiscence. nmvnamx-COOHE TEPALS, shortest (inner) and longest (outer) respectively, if present. A. australis From C.C.Deam 21984(IND) A. floridanus A. tamariscinus From syntype, AW. Chapman = s.n. (MO) From R. Bebb 3698(OKL) A. Palmeri From syntype, J.L.Berlandier 2407 (GH) From syntype, E.Palmer 312 (US) Pee Greggii A. arenico From holotype, A.S. Hitchcock 428A (GH) 9 From holotype, J.Gregg s.n (@H) A. Acanthochiton From syntype, S.W Woodhouse s.n. (6H) Fic. 1. Pistillate flower parts of the ten dioecious species of Amaranthus. 1955 | SAUER: AMARANTHUS 9 STAMINATE FLOWER PARTS Left to right: SHORTEST (INNER) TEPAL. LONGEST (OUTER) TEPAL. nHIMHAMERCr-z BRACT. Stamens not shown. A. cannabinus A. australis From From F.R. Fosberg 23961(WIS) syntype, E.Palmer 462 (GH) A. arenicola A. Greggii A. Acanthochiton From isotype, From From syntype, syntype, E.Paimer 675 (US) A. S.Hitchcock 428A (MO) @.L. Fisher 608(US) S.W. Woodhouse s.n. (GH) Fic. 2. Staminate flower parts of the ten dioecious species of Amaranthus. indicates that it is outside the area covered by the map. The year given in parentheses after each major civil division name is the collection date of the oldest specimen seen from the region. Collections are listed only by collector, number (if given), and herbaria, abbreviated according to the scheme of Lanjouw and Stafleu (1954). Collections from the herbaria at the following institutions are cited: California Academy of Sciences, Carnegie Museum, Cornell University, Duke University, Gray Herbarium of Harvard University, Herbier Marie- Victorin de l'Université de Montréal, Indiana University, Iowa State Col- lege, Kansas State College, Michigan State College, Missouri Botanical Garden, Museum of Northern Arizona at Flagstaff (cited as FLAG), New Mexico College of Agriculture and Mechanic Arts, North Dakota Agri- cultural College, Oberlin College, Ohio State University, Pomona College, Rocky Mountain Herbarium of the University of Wyoming, Smithsonian Institution, South Dakota State College, Southern Methodist University, State University of Iowa, Sul Ross State College, Tulane University, University of Arizona, University of Arkansas, University of California at Berkeley, University of Cincinnati, University of Colorado, University of Florida Agricultural Experiment Station, University of Illinois, University of Michigan, University of Minnesota, University of Nebraska, Univer- 10 MADRONO [Vol. 13 sity of North Carolina, University of Oklahoma, University of Tennessee, University of Texas, University of Wisconsin, Utah State Agricultural College, West Virginia University. KEYS TO THE SPECIES The key to pistillate plants is the more reliable one, since most of the diagnostic characters are concentrated in the mature pistillate flowers. Flowering staminate plants can be identified with reasonable accuracy but recognition depends in part on familiarity with slight differences in aspect which cannot be adequately described in a key. In dealing with any of these plants it must be remembered that the frequency of hybrid- ization in the group makes identifications based solely on a key even less trustworthy than usual. PISTILLATE PLANTS A. Lamina of bract narrow, entire, scarious, not enfolding flower ; leaf margin smooth. B. Tepals completely lacking or irregularly present and rudimentary (less than 1 mm. long and without visible midveins). C. Seed 2 to 3 mm. long; utricle 214 to 4 mm. long; leaf-blade usually narrowly lanceolate to linear ‘ » 2 ds Alcannabinus CC. Seed about 1 mm. long; niricle not Sean 2 mm. in length; leaf-blade usually broadly lanceolate. D. Utricle with conspicuous, regular longitudinal ridges; bract more than 1% mm. long with stout midrib not far excurrent beyond lamina. ee eo RA OUStnalis DD. SWuice Ghee or cecal fuberculates “Sra less than 144 mm. long with slender excurrent midrib... . 3. A. tuberculatus BB. Tepals regularly present and well developed (at fence 1 mm. long and with distinct midveins) . C. Tepals 1 or 2, lanceolate to linear. D. Utricle indehiscent; leaf-blade usually linear . . . 4. A. floridanus DD. Utricle a Se leaf-blade usually oblong or lanceolate’ oa de a Se Ae tamarisemius CC. Tepals 5, at least the inner ones Spataiete: D. Utricle circumscissile, about 144 mm. long; leaf-blade membranaceous. E. Longest (outer) tepal acute or acuminate with midvein excurrent into rigid point; bract and outer tepals conspicuously longer than inner tepals ae aes ie OME Pal aeng EE. Tepals all obtuse or “HS ine iakikasizg Sa slightly or not at all; bract and outer tepals scarcely exceeding inner tepals. F. Bract with moderately heavy midrib, excurrent far beyond lamina; style branches usually 2 vt i Sly ah ale ie A aLsone FF. Bract with extremely heavy midrib, mee caren far beyond lamina; style branches usually 3... , 89 Avarenicola DD. Utricle indehiscent, about 3 mm. long; [eat Blade coriaceous. : 9. A Greggit AA. itevtiees ee Sra) exon Broad prenate Foliiccoucl enioldine and concealing flower; leaf margin crispate ..:. . ~~ 223.940: -AlAcanihochiton STAMINATE PLANTS A. Outer tepals without heavy midveins and not appreciably longer than the inner; bracts mostly with slender midribs, not over 2 mm. long. B. Bract less than 1 mm. long, the midrib scarcely excurrent. 1. A. cannabinus 1955] SAUER: AMARANTHUS i BB. Bract more than 1 mm. long or, if slightly shorter, the midrib conspicuously excurrent. C. Leaf-blade usually lanceolate, more than 1 cm. wide; inflorescence often with several branch thyrses not subtended by leaves. D. Bract with moderately heavy midrib; outer tepals with excurrent midveins ; Sie ae Dee AL QUSsial7s DD. Bract with Sender eeakeiye Pardee Of outer tepals not excurrent .. ~ «2 «= 3. A.tuberculatus CC. Leaf-blade usually iHaear fees fhen 1 cm. ide inflorescence unbranched above leaves. . 4. A. floridanus AA. Outer tepals with heavy midveins, biten deqnitely foneey than the inner; bracts with heavy midribs, mostly over 2 mm. long. B. Outer tepals acuminate, the midveins excurrent as rigid spines. C. Bract about 2 mm. long, definitely shorter than outer tepals 5. A. tamariscinus CC. Bract about 4 mm. long, usually equalling outer tepals . 6. A. Palmeri BB. Outer tepals acute or obtuse, apiculate, but the dark midveins not excurrent. C. Bract equalling tepals, often conspicuously pubescent . . 7. A. Watsoni CC. Bract shorter than tepals, glabrous. D. Leaf margin smooth. E. Leaf-blade membranaceous . . . . . ... 8. A.arenicola EE. Leaf-blade coriaceous ee e- Lote tae, Bo . 9. A. Greggit DD; Weal marginccrispate ©. 2 5. 4 . oa 10. A. Acanthochiton 1. Amaranthus cannabinus (L.) comb. nov. Acnida cannabina L. Sp. Pl. 2:1027. 1753. A. rusocarpa Michx. Fl. Bor. Am. 2:234. 1803. A. rus- cocarpa Willd. Sp. Pl. 4:768. 1805. Amaranthus macrocaulos Poir. in Lam. Encyc. Suppl. 1:314. 1810. Acnida salicifolia Raf. Am. Mo. Mag. 2:43. 1817. A.rhyssocarpa Spreng. Syst. 3:903. 1826. A. obtusifolia Raf. New FI. 1:54. 1836. A. Elliott Raf. loc. cit. 54. 1836. A. cannabina L. [ var.] a lanceolata Mog. in DC. Prodr. 13(2):277, 1849. A. cannabina L. [var.] y salicifolia (Raf.) Mog. loc. cit. 278. 1849. The Linnaean species was based on an older taxon (Cannabis foliis sim- plicibus Gronov. Fl. Virg. 192. 1739. Type: J. Clayton no. 599, Virginia, salt marshes, abundant in August). The description and place of collection make its identity reasonably certain. Michaux distinguished his species from that of Linnaeus by slight differences in leaf and utricle shape; he specified no type or locality but published a recognizable drawing of the plant. During the nineteenth century various authors attempted to main- tain a distinction between Acnida cannabina and A. rusocarpa (the latter often being given one of the “corrected” spellings suggested by Willdenow and Sprengel). This attempt was finally abandoned after Uline and Bray (1895, p. 155) pointed out, correctly I believe, that “A. cannabina and A. ruscocarpa are not distinct. We find here a complete series of inter- gradations, while the difference of age actually accounts for more than the difference called for in their original descriptions.” Amaranthus macro- caulos was first published with a completely inadequate description, but Moquin-Tandon re-examined the type (J. Bosc s.n., Carolina), citing Poiret’s species as a synonym of his own variety lanceolata, based on the same specimen. From Moquin-Tandon’s description and the place of col- lection, it is fairly certain that Bosc’s specimen should be assigned to A. 12 MADRONO [Vol. 13 cannabinus, although the material may represent one of the southern colonies which show traces of introgression from A. australis. Acnida salicifolia was described from plants growing on the seashore and in marshes and ditches in Long Island and New Jersey; no specimens are cited. The localities, together with Rafinesque’s statement that the species is intermediate between A. cannabina and A. rusocarpa, indicate that he was dealing with plants which were at the most minor variants of Amar- anthus cannabinus. Rafinesque presented Acnida obtusifolia as a sub- stitute name for the taxon which everyone had been calling A. cannabina, stating that Linnaeus had described that species as having compound leaves and was therefore dealing with an entirely different plant. This curious idea has no known basis; Linnaeus’ original description specific- ally stated that the species has simple leaves. Acnida Ellioti was de- scribed as growing on the banks of streams in Carolina and Florida. No specimens are cited, but the locations and the incomplete description sug- gest that this name, like Amaranthus macrocaulos was called forth by contact with some of the atypical plants which occur toward the southern margin of the range of A. cannabinus. Plants stout, erect, usually 1 to 3 m. tall with ascending branches; leaf-blade narrowly lanceolate to linear, faintly resembling a willow or hemp leaf (whence some of the scientific names and the common name of “water-hemp”’), flowering and fruiting entirely during summer and fall, mainly July to late October; thyrses flexible, usually 5 to 10 cm. long, the glomerules often few-flowered and widely-spaced, in the ¢ plants leafless branch thyrses often numerous and the uppermost ones often not sub- tended by leaves, the 9 thyrses either entirely terminal on leafy branches or, if a few branch thyrses present, each subtended by a leaf; bract with midrib scarcely excurrent, the ¢ bract about 1 mm. long, midrib very slender, the 2 bract about 14 mm. long, midrib moderately heavy; ¢ flowers with 5 stamens, the 5 tepals approximately equal, 2% to 3 mm. long, the inner emarginate, the outer acute, the midveins not excurrent; 2 flowers usually without perianth, rarely with 1 or 2 irregular, rudimen- tary tepals; utricle 2% to 4 mm. long, indehiscent, fleshy, with 3 to 5 prominent longitudinal ridges corresponding to the 3 to 5 style branches, often rugose and black when mature; seed 2 to 3 mm. in diameter, often obovoid, flattened with depressed endosperm, dark reddish brown. The species is almost entirely confined to the margin of tidewater in a zone where the surface is covered by salty, brackish, or fresh water at high tide. Almost every collection bears some such notation as: coastal marsh, brackish marsh, salt marsh, salt meadow, tide flats, edge of slough, bank of estuary, tidal riverbank. The plants are most common in sandy places but are also reported in mud and muck. A collection from a millpond in Delaware was the only one seen from an inland site. If the species has weedy tendencies, they are slight; there is one collection from wet, peaty clearings in the pinewoods along the Virginia coast and another from a railroad yard by the tidal Delaware River at Camden. 1955] SAUER: AMARANTHUS 13 UNITED STATES. CONNECTICUT (1859). Farrrretp: E. R. Drew KSC, MO, UC; E. H. Eames GH, ILL, MIN, US, WIS. Mippresex: Anonymous GH, MO; H.L. Jones OC. New Haven: E. H. Eames ILL; D.C. Eaton WIS; G. R. Kleeberger CAS; A. B. Seymour DUKE; R. H. Ward ILL. NEw Lonpon: K. P. Jansson COLO, MT, RM; W. A. Setchell UC. DELAWARE (1861). Kent: N. Hotchkiss 4765, US; E. L. Larsen 711, DUKE, GH, US, WIS, 734 GH, 762 MO, US, 770 MO. Newcastle: W. M. Canby POM, WIS; J. R. Churchill GH, MO, WIS; E. L. Larsen 665 MO; I. Tidestrom 11529 GH. DISTRICT OF COLUMBIA (1877). F. Blanchard MO, WIS; E. S. Steele MIN, NEB, OC, WIS; G. Vasey MSC; L. F. Ward GH, MO, NEB, US. FLORIDA (1880). Indefinite locality: W. W. Calkins ILL; A. H. Curtiss NCU. Duvat: A. H. Curtiss 2379 CM, CU, FLAS, GH, MIN, NEB, US, 5117 ILL, MO, US. GEORGIA (1902). CAmMpEN: R. M. Harper 1556 GH, MO, MSC, NEB, US; Wiegand & Manning 1131 CU, GH. McIntosu: Thorne & Norris 6231 CU. MAINE (1916): CuMBERLAND: Fernald, Long, & Norton 13581 GH, MT. York: A. E. Perkins GH; R. H. True 1086 MIN. MARYLAND (1887). ANNE ARUNDEL: H. H. Bartlett 1837 MICH; W. C. Muenscher 3711 CU, 3712 CU; J. W. Roller 754 TENN; J. Schneck ILL. CALVERT: A. S. Hitchcock ILL. Cectt: W. M. Benner 6165 GH. Cuartes: H. O'Neill MT. Prince GeorcEs: F. Blanchard MO. Saint Mary’s: J. E. Benedict, Jr. UARK; E. P. Killip 32194 MICH, POM, UC. Tarsort: E. C. Earle 1283 WVA, 2265 CU; F. Shreve 176 ARIZ. MASSACHUSETTS (1856). BARNSTABLE: J. M. Fogg, Jr. 3706 CU, MO; C.4H. Knowlton MO; C. Pickering GH; W. P. Rich GH, MIN; E. F. Williams GH. Essex: N.C. Fassett 16029 WIS; J. H. Sears DUKE, GH; E. F. Williams GH. MiIppLEsEx: Anonymous GH; W. J. Beal MSC; F. S. Collins MIN; G. Engelmann MO; H. L. Jones OC; G. G. Kennedy CU, POM, RM, WIS; 7. Morong ISC, MO; B. L. Rob- inson US; A.B. Seymour DUKE; W. Trelease ILL, MO. Norrork: C.E. Faxon GH; F. F. Forbes 14464 WIS. SurroLx: J. R. Churchill MO; W. P. Rich GH. NEW HAMPSHIRE (1901). Rockincuam: B.L. Robinson 779 GH; E. F. Wil- liams GH. StRAFForD: F. C. Seymour 4855 DUKE, WIS. NEW JERSEY (1841). Indefinite locality: R. C. Alexander UC; C. F. Austin GH; H.R. Bassler KSC; P. D. Knieskern CU; McMinn US. Atiantic: W. M. Ben- ner 9742 WIS; J. Bright 10941 MIN, 10942 WIS; J. B. Brinton MO. BERGEN: G. Lei- derman 63 WIS; K. K. Mackenzie 602 ARIZ; H. Moldenke 13886 CM; W. Shear WVA. CAMDEN: G. W. Bassett CM, GH, MT; G. M. Beringer MICH; C. D. Fretz UC; C. A. Gross 8 GH; A. MacElwee 1299 GH, KSC, MO, MT; I. C. Martindale GH, US, WIS; W. H. Witte NMC, RM. Carr May: E. B. Bartram MT; J. Bright CM; A. Gershoy 310 CU, GH; E. P. Killip 150 US, 316 POM; F. W. Pennell 2177 US; W. H. Witte RM. CUMBERLAND: F. W. Pennell 14867 MIN; T. Seal CM. GLoucestTer: J. B. Brinton US; R. H. True 5604A UC. Hupson: J. Carey MO; G. Thurber GH; W. M. Van Sickle US. Mippiesex: L. H. Lighthipe MSC. Monmoutu: A. P. Kelley MT. SALEM: E. L. Core 5016 WVA; E. C. Earle 679 TENN; J. M. Fogg, Jr. 7792 MIN. NEW YORK (1873). Indefinite locality: S. B. Buckley MO. AtBAny: H. D. House 24278 CU, GH, 30376 CU. Bronx: S. H. Burnham 726 GH; J. Cohn CU. CotumsiA: Muenscher & Clausen 4486 CU. GREENE: H. D. House 25167 GH; Muen- scher & Clausen 4483 CU, GH, 4484 CU, GH, US. Manuattan: E. C. Howe 2615 IA. Nassau: A. Gershoy CU; Muenscher & Curtis 6125 CU, 6126 CU. QuEENs: F.W. Hulst ILL; J. Schrenk CU. RensseLaAER: H. D. House 27059 CAS; Muenscher, Brown, & Langdon 21576 CU. Ricumonp: N. L. Britton CM; J. A. Drushel 8208 ILL, WIS, 8211 ILL, MO; A. Gershoy 796 CU. Rocxtann: Muenscher & Curtis 5722 CU. Surrotk: E. S. Miller CM, US; Muenscher & Curtis 6127 CU, 6128 CU; H. St. John 2712 CU, GH, US; H. Schrenk CU, MO, WIS; S. H. Wright MT; H.W. Young MT. Utster: Isely, Muenscher, & Winne 2414 UARK; Muenscher & Curtis 5721 CU. WESTCHESTER: E. C. Howe ILL; Muenscher & Curtis 5718 CU, GH, 5719 CU, 5720 CU, MICH, WIS; H. Schrenk MO; E. H. Walker US. 14 MADRONO [Vol. 13 MILES 100 KILOMETERS Pasi dee A. cannabinus Fic. 3. Amaranthus cannabinus: distribution map. Atypical plants indicated by crosses. 1955] SAUER: AMARANTHUS 15 NORTH CAROLINA (1885). Indefinite locality: G. McCarthy 176 NCU, US. BeEAuFrort: A. E. Radford 5370 NCU. Carteret: A. E. Radford 5018 NCU. Currt- TtucK: W. L. McAtee 1170 US. NEw HANover: Anonymous US; A. E. Radford 5039 NCU. PENNSYLVANIA (1837). Indefinite locality: E. Durand GH. Bucks: E. B. Bartram 1292 GH, MT; W. M. Benner GH; R. R. Dreisbach 3877 MICH. PHILA- DELPHIA: R. C. Alexander UC; W. M. Benner 6966 GH; I. C. Martindale GH. RHODE ISLAND (1844). ProvipeNcE: W.W. Bailey GH, MO, US; J. W. Cong- don MIN, MO, NEB; G. Thurber, GH. SOUTH CAROLINA (Before 1900). BERKELEY: K. W. Hunt 1276C CU; H.W. Ravenal GH. CuHArteston: C. P. Alexander 43 US. Horry: W. C. Coker NCU. VIRGINIA (1889). Indefinite locality: F. G. Braendle MIN. ARLINcTon: A. Chase 2674 ILL; A. S. Hitchcock ILL; T. Holm ILL, MO; A. J. Pieters MICH; E. S. Steele DUKE, KSC, MIN, MO, MSC, MT, OC. Essex: I. Tidestrom 7693 US. FairFAx: F. R. Fosberg 23960 WIS, 23961 WIS; G. H. Shull 191 GH, MO, 192 MO. James City: L. C. Artz 1134 WVA. MAtHeEws: Wherry & Pennell 12566 MO. Nor- FOLK: F. Blanchard MO, US; Fernald & Long 11026 GH, 11027 GH. NorTHAMPTON: W. M. Canby ARIZ. Prince GeorcE: Fernald & Long 10250 GH. PRINcEsS ANNE: Fernald & Griscom 2820 GH; Fernald, Long, & Fogg 4877 GH. Starrorp: F. J. Her- mann 9721 GH. Surry: Fernald & Long 9572 GH. 2. Amaranthus australis (A. Gray) comb. nov. Acnida cuspidata Bert. ex Spreng. Syst. Veg. 3:903. 1826. Non Amaranthus cuspidatus Vis. Cat. Hort. Patav. 1841. Acnida cannabina L. |var.| B cuspidata (Bert. ex Spreng.) Mog. in DC. Prodr. 13(2):277, 1849. A. australis A. Gray. Am. Nat. 10:489. 1876. A. cannabina L. [var.] australis (A. Gray) Uline & Bray, Bot. Gaz. 20:157. 1895. A. alabamensis Standl. N. Am. Flora 21:121. 1917. Although Bertero’s Acnida cus pidata is the first published name for this entity, because of the homonym rule, Gray’s name is the one available when the species is transferred to Amaranthus. Gray’s species was de- scribed from Florida collections which are excellent representatives of the species (syntypes: A. W. Chapman s.n., Apalachicola, 9 GH! here desig- nated as lectotype; E. Palmer 462, Biscayne Bay, 1874, § GH! ). Stand- ley, in erecting A. alabamensis (holotype: C. T. Mohr s.n., swampy bor- ders of tidewater streams, One-Mile Creek near Mobile, Alabama, ? US! ; isotype: ¢ US!), to care for a few collections from the Gulf Coast was recognizing characters to be expected in populations of A. australis after slight introgression from A. tuberculatus, the other dioecious species in the area, notably, shorter staminate tepals, shorter pistillate bracts and less flattened seeds. When all the available specimens of A. australis and A. alabamensis are assembled, morphological intergradation appears to be complete and there is no discontinuity which permits taxonomic separation. Plants extremely stout and large with many ascending branches, usual- ly 2 to 3 m. tall (although presumably annuals like the rest of the genus, members of this species sometimes reach surprising size, some having been reported as being 9 meters high*); leaf-blade lanceolate, often long at- 4 Prof. A. J. Sharp has sent me a photograph, taken in a Florida swamp, showing a man climbing one of these plants as if it were a tree. 16 MADRONO [Vol. 13 tenuate at the tip; flowering and fruiting mainly in summer and fall in the northern colonies, in all seasons in tropical colonies; thyrses flexible or moderately stiff, usually 5 to 10 cm. long; the terminal thyrse often ac- companied by leafless branch thyrses and the uppermost of these not sub- tended by leaves; the branch thyrses somewhat more numerous and more crowded in the ¢ than in the 9 plants; bract 114 to 2 mm. long, midrib moderately heavy in “', heavy in 9, not conspicuously excurrent in either; 3 flowers with 5 stamens, the 5 tepals approximately equal, 2% to 3 mm. long, the inner emarginate, the outer acuminate with excurrent midveins; ? flowers without perianth; utricle 114 to 2 mm. long, indehiscent, slight- ly fleshy, with 3 to 5 prominent longitudinal ridges corresponding to the 3 to 5 style branches, not rugose, stramineous; seed 1 to 114 mm. in diameter, round, somewhat flattened with depressed endosperm, dark reddish brown. The species is confined to wet places along the margins of both salty and fresh streams and water bodies. Almost every collection bears some such notation as: coastal swamp, mangrove swamp, coastal marsh, canal bank, lakeshore, riverbank, edge of lagoon, along bayou, shore of estuary, low hammock. There is no indication that the plants are every weedy invaders of artificial habitats. An isolated colony on the west coast of Mexico at the port of Manzanillo may have resulted from accidental human transport of seeds. The colony was probably ephemeral since it was reported but once and only staminate plants were found (Rose, 1895, Des o2). CUBA. HABANA (1917). Bros. Leon, Marie-Victorin, & Alain 22355 MT; Bros. Leon & Roca 7272 MT. HAITI. L’ARTIBONITE (1925). £. L. Ekman H3358 US; A. T. Sweet 70 GH, US: SUD (1927). W.J. Eyerdam 427 GH, US. JAMAICA. Indefinite locality (ca. 1840). J. Macfadyen GH. CORNWALL (1907). St. ExizasetH: W. Harris 9760 US. WESTMORELAND: Britton & Hollick 2102 US. MEXICO. *COLIMA (1891). MAnzaAnit1Lo: £. Palmer 1399 US. QUINTANA ROO (1938). CozuMEL: Lundell & Lundell 7777 US. TABASCO (1889). Centro: J. N. Rovirosa 672 US. TAMAULIPAS (1939). AttAmirA: H. LeSueur 119 ARIZ, US. *TRINIDAD (1914). W. E. Broadway US; H. Crueger GH. UNITED STATES. ALABAMA (1879). Mosite: C.T. Mohr US. FLORIDA (1867). Indefinite locality: 4. W. Chapman GH, MO, US, WIS; R. Combs 1102 CAS, NMC, RM; dH. C. Cowles ILL; J. H. Simpson KSC, US. ALACHUA: Laessle, West, & Arnold FLAS. Citrus: A. S. Hitchcock 591 NEB. Cot- LIER: L. H. MacDaniels CU. DAveE: E. A. Bessey MSC; H. N. Moldenke 531 DUKE, ILL, MO, US; E. Palmer 462 GH, MO; W. Rusby MICH; J. K. Small 10437 DUKE, FLAS, GH, TENN, US, WVA. Desoto: H. O’Neill FLAS, US. Duvat: A. Fredholm SAUER: AMARANTHUS 17 1955] ‘sasso1d Aq poyeorpul syueyd pestdAzy ‘dew uonnqiyjsip -syosjsny snyjunapup “p “OIA SusLIROTM oon 008 ool te) o& ond sTun 18 MADRONO [Vol. 13 300 US. Franxuin: A. W. Chapman GH, 151A GH, MIN, MO, NCU, US. GUuLF: W.L. McAtee 1720 US; B. F. Saurman 381 GH. Hicuianps: J. B. McFarlin 6029 MICH. Hirtssorovucu: J. D. Smith US. LAKE: G. V. Nash 868 CU, GH, MICH, MIN, MO, MSC, NEB, UC, US; J. R. Watson 77 FLAS. Lee: A. S. Hitchcock 299 CU, GH, KSC, MIN, MO, NEB, US. Manatee: J. H. Simpson MIN, MO, US, WIS. Monroe: Muenscher & Thorne 18344 CU. ORANGE: A. Fredholm 5470 GH. PALM Beacu: M. F. Baker FLAS; J. A. Harris N19335 MIN; M. J. Murray 36075 CU, 36076 CU, 38135 CU; H. O’Neill MO; Small & Carter MIN; Tisdale, Townsend, & West FLAS. Pork: J. B. McFarlin 4459 CAS, 4460 MICH. Putnam: A. M. Laessle FLAS. LOUISIANA (1886). IBERIA: Correll & Correll 9535 DUKE, GH, MO. La FourcuHe: G. Arceneaux 95 CU. OrtEAns: R. S. Cocks NO; G. L. Fisher WIS; J. F. Joor MO, US; J. L. Riddell NO; E. Wilkinson OC. PraguemineEs: Lloyd & Tracy 41 CM; Tracy & Lloyd 24 CU, GH, MIN, MO, MSC, NEB, US, 37 CM, CU, GH, MIN, MO, NEB, UC, US, WIS. St. Bernarp: J. F. Joor MO. St. TAMMany: W.T. Pen- found NO. VERMILLION: C.C. Sperry 385 US. TEXAS (1884). Harris: G. L. Fisher CAS, 27 MT, US, 165 US; J. F. Joor MIN, MO, WIS. OranceE: V. L. Cory 50863 SMU. Travis: B. C. Tharp 2864 MICH, US. *VENEZUELA. ARAGUA (1942). Killip & Lasser 37736 US; H. Pittier 14996 Us. 3. Amaranthus tuberculatus (Moq.) comb. nov. A. altissimus Rid- dell, Syn. Fl. W. States 41. 1835, nom. prov. A. miamiensis Riddell, loc. cit. 41. 1835, nom. prov. Acnida tuberculata Mog. in DC. Prodr. 13(2):277. 1849. A. altissima (Riddell) Mog. loc. cit. 278. 1849, non. nudum. A. can- nabina L. |var.| 8 concatenata Mog. loc. cit. 278. 1849. Montelia tama- riscina (Nutt.) A. Gray var. concatenata (Moq.) A. Gray, Man. ed. 2. 370. 1856. Acnida tuberculata Mogq. var. subnuda S. Wats. in A. Gray, Man. ed. 6. 429. 1889. A. tamariscina (Nutt.) Wood [var.] subnuda (S. Wats.) Coult. Mem. Torrey Club 5:145. 1894. A. tamariscina (Nutt.) Wood [var.]| tuberculata (Moq.) Uline & Bray, Bot. Gaz. 20:157. 1895. A. tamariscina (Nutt.) Wood |[var.| prostrata Uline & Bray, loc. cit. 158. 1895. A. tamariscina (Nutt.) Wood [var.| concatenata (Moq.) Uline & Bray, loc. cit. 158. 1895. A. concatenata (Moq.) Small, Fl. S.E. U.S. 393. 1903. A. tuberculata Mog. var. prostrata (Uline & Bray) B. L. Robinson, Rhodora 10:32. 1908. A. altissima (Riddell) Mog. ex Stand]. N. Am. Flora 21:122. 1917. A. subnuda (S. Wats.) Standl. loc. cit. 122. 1917. Amaranthus ambigens Stand. loc. cit. 106. 1917. Acnida altissima (Rid- dell) Mog. ex Standl. var. prostrata (Uline & Bray) Fernald, Rhodora 43:288. 1941. A. altissima (Riddell) Mog. ex Standl. var. subnuda (S. Wats.) Fernald, loc. cit. 288. 1941. This species probably first received taxonomic attention from Riddell, who proposed two new species, presumably based on his own Ohio collec- tions: Amaranthus altissimus from “‘an old prairie near Hamilton,” and A, miamiensis from “Hoffman’s prairie, Dayton.” His incomplete descrip- tions together with the localities suggest that both names refer to the taxon under consideration here. Unfortunately, Riddell added the state- ment: “TI give these as merely temporary names, until the plants shall be 1955] SAUER: AMARANTHUS 19 further investigated,” thereby invalidating his publication of the names. One of his epithets, miamiensis, has never been validly published; the other was eventually adopted by Standley in a new combination but by then it was nomenclaturally superfluous. Although Moquin-Tandon cited no specimens, his description of A. tuberculata, made from living plants in the Geneva Botanical Garden leaves little doubt as to its identity. His type specimen of var. concatenata (holotype: T. Drummond 552, pro- cumbent, New Orleans, Louisiana, 1832. Fragment.in GH! ), originally assigned to another species, although subtly different from typical A. tu- berculatus (suggesting slight introgression from A. australis) is morpho- logically well within the limits of the species as defined here. In various combinations and ranks, concatenata has generally been recognized by subsequent workers as a distinct taxon. It has ordinarily been distin- guished from A. tuberculatus by unusually large, round, widely-spaced glomerules, which give the pistillate inflorescence the look of a loose string of giant beads. Such inflorescences develop occasionally in many kinds of amaranths, either sterile hybrids or isolated pistillate p!ants which are not pollinated; in such plants seed set does not arrest vegetative growth of the inflorescence axis and flower initiation in each cyme. Watson’s variety subnuda (holotype: Oakes s.n.,;W. Vermont) is a different story. Watson’s statements on morphology and geography indicate that his conception of the species proper was based on plants which were probably mostly hy- brids between A. tuberculatus and A. tamariscinus while his variety was based on A. tuberculatus in its purest and most typical form. Uline and Bray distinguished their variety prostrata by a prostrate habit, small spatulate leaves, and a poorly developed terminal thyrse. They cited no specimens but stated “type specimens in Nat. Herb. and Mo. Bot. Gard.” They apparently marked no sheets as types, but the following specimens bear their annotations as representing this variety: G. Engelmann 257, 258, St. Louis, Missouri, August to September, 1893, ¢ 2MO!; G. En- gelmann s.n., American Bottom, Illinois, opposite St. Louis, October 7, 1867, 6 2MO! WIS!; C. A. Geyer 420, Nicollet’s Northwestern Expedi- tion, Fort Pierre (South Dakota), June 29, 1839, 2US!; E. Hall s.n., riverbanks, Athens, Illinois, 1861, 2US!; G. H. Hicks s.n., Michigan Agricultural College grounds, September 1, 1892, 2US!; J. M. Holzin- ger s.n., Winona, Minnesota, August, 1890, ¢ 2 US!. All these specimens are well within the range of A. tuberculatus as circumscribed here, al- though the Engelmann collections show traces of mixing with A. tamaris- cinus. These specimens, and many others like them, are indeed conspicu- ously different from the usual form of A. tuberculatus, but the difference may not be hereditary. The key to the dissimilarity may be found in an unheeded note by Engelmann on his American Bottom collection: “forma autumnalis, in the bottom of dried swamps, a second crop.” These plants are extremely sensitive to photoperiod, flower initiation being approxi- mately simultaneous in all individuals within an area, regardless of size. By early and late greenhouse plantings, I have been able to obtain both 20 MADRONO [Vol. 13 ordinary tuberculatus and the prostrata forms from seed of a single plant (J. D. Sauer 1592-4, 1592-4A, 1592-10, 1592-10A, WIS). Standley’s A. ambigens is based on a sheet (M.S. Bebb s.n., Fountaindale, Illinois, US!) bearing two plants, one of them an ordinary staminate plant and the other bearing abortive bisexual flowers, an anomaly among both dio- ecious and monoecious amaranths. Other sheets of this collection (CU!, MINN!, US!) bear only ordinary specimens of A. tuberculatus. Compar- able monstrosities with sterile bisexual flowers have been collected else- where within the range of A. tuberculatus (A. P. Anderson s.n., Goodhue County, Minnesota, US!; W. S. Moffat s.n., DuPage County, Illinois, ILL!; L. S. Cheney, Dane County, Wisconsin, WIS! ), as well as within the range of A. arenicola (G. E. Osterhout 1141, Logan County, Colo- rado, RM! ). Plants extremely variable in habit and size, prostrate, ascending, or erect, often very short but sometimes reaching 3 m. in height; leaf-blades extremely variable in size and shape, the smaller ones usually oblong or spatulate, the larger broadly ovate or lanceolate; flowering and fruiting entirely during summer and fall, mainly late July to early October; thyrse flexible, usually about 5 cm. long in ¢, 1 to 2 cm. long in 2; in the 2 plants several loosely arranged, leafless branch thyrses often present above the uppermost leaves; 2 thyrses either entirely terminal on leafy branches or, if leafless branch thyrses present, these crowded and each subtended by a leaf; bract 1 to 11%4 mm. long; midrib very slender in ¢, slender in 2, excurrent far beyond lamina; ¢ flowers with 5 stamens, the 5 tepals approximately equal, 2'’%4 to 3 mm. long, the inner obtuse or emarginate, outer acuminate, the midveins not excurrent; 2 flowers usu- ally without perianth, occasionally with 1 or 2 irregular, rudimentary tepals; utricle 114 to 2 mm. long, indehiscent, thin, smooth or irregularly tuberculate, sometimes with faint ridges corresponding to the 3 or 4 style branches, often reddish; seeds 34 to 1 mm. in diameter, often obovoid, lenticular, dark reddish brown. The species is at home on the margins of freshwater bodies of all sorts: rivers, creeks, lakes, ponds, marshes, and bogs. Nearly 90 per cent of all collections which bear habitat data are from such places. The plants are especially abundant in a narrow zone close to the water’s edge, where fall- ing water level has exposed a strip of bare sand or mud. Much of the vege- tative variability within the species may be traceable to variations in the growing season available in such sites. The species also occupies another quite different group of habitats: artificially disturbed places such as fields, gardens, and roadsides. Where such places are available in low ground close to its natural habitats, the species commonly moves in as a weed. Elsewhere it has had little success as a weed. CANADA. ONTARIO (1871). CARLETON: Frere Marie-Victorin MT; Frere Rolland GH, IND, MT. Essex: W.S. Cooper ARIZ. Hastincs: J. Macoun 1506 GH. Huron: J. A. Morton 1828 ILL, MIN, US. RussEtt: J. Macoun 86622 GH, 86623 GH, 86624 GH. 1955] SAUER: AMARANTHUS Zi QUEBEC (1889). Wricut: J. Macoun GH, MO, 16 GH; W. Scott CU. UNITED STATES. ALABAMA (ca. 1875). Indefinite locality: S.B. Buckley MO. ARKANSAS (1929). Cutcot: D. Demaree 14062 (in part) US. CraicHeap: D. Demaree 7095 GH, TEX, UARK, US, WIS, 27501 ISC. Mississippi: D. Demaree 7186 US. Puiturps: D. Demaree 30234 OKL, TEX. CONNECTICUT (1896). Hartrorp: C. H. Bissell GH. NEw Haven: E. B. Harger 18 GH. ILLINOIS (1840). Apams: R. Brinker 3768 ILL. Carrot: M. B. Waite ILL. CHAMPAIGN: G. P. Clinton ILL; F. Coates ILL; H. A. Gleason 76 GH; G. N. Jones 12905 IA, ILL, MIN, MO, US, 13136 ILL, 13279 ILL; A. S. Pease 13016 GH; A. B. Seymour DUKE; W. Trelease ILL. Curistian: W. FE. Andrews ILL. Cook: R. Bebb 2123 MIN, OKL, WIS; J. R. Churchill GH, RM; G. Engelmann MO; E. J. Hill 74 NUE, 276 ILL, 165 ILL; W.S. Moffatt MIN, OC, WIS, 429 ILL, 617 ILL, OC; N. L.T. Nelson MIN, UC, WIS; W. C. Ohlendorf MO, US; L. H. Pammel ISC; E. E. Sherff 1726 MO; M. W. Strahler WIS; L. M. Umbach 1168 WIS, 1333 WIS, -4880 WIS, 5931 WIS; G. S. Winterringer 1558 ILL, US. Douctas: G. S. Winter- ringer 153 ILL. Du Pace: W. S. Moffatt ILL, 267 ILL, 287 MIN, 609 WIS, 612 ILL, MO; L. M. Umbach GH, ILL, MICH, MIN, MO, MSC, OC, OKL, UC, US, WIS. Furtton: V. H. Chase 10908 ILL. Grunpy: G. S. Winterringer 81 ILL, 83 ILL. HENDERSON: H. N. Patterson GH, ILL, MO, US, WIS. Iroquots: G. S. Winterringer 84 ILL. Jackson: H. E. Ahles 5769 ILL; H. A. Gleason 1685 GH. KAneE: J. D. Sauer 1599 WIS. KANKAKEE: Anonymous ILL; C. C. Crampton 540 US. KENDALL: G. S. Winterringer 82 ILL. LAKE: O. C. Durham GH; E. J. Palmer 28331 MO; L. M. Umbach 1353 WIS. LASALLE: G. Engelmann MO; G. D. Fuller 3256 (in part) ILL; Koster & Koster 5 CU, 6 CU; L. M. Umbach WIS, 955 WIS. Lawrence: J. P. Sivert ILL. Livincston: J. D. Sauer 1601 ILL, NO, WIS. Macon: R. G. Mills ILL; G. S. Winterringer 639 ILL. Macoupin: W. E. Andrews ILL; J. D. Sauer 1608 WIS. Mason: H. E. Ahles 3423 ILL; R. J. Miller ILL. McHENry: W. A. Nason ILL; G. R. Vasey MO. McLEAn: Anonymous CM; J. D. Sauer 1606 ILL, NO, WIS. Menarp: E. Hall ILL, US. Octe: M. S. Bebb CU, MIN, US; M. B. Waite US. PreoriA: H. A. Anderson IA; F. Brendel ILL, US; F. E. McDonald ILL, NMC, UC; J.T. Stewart WIS. Pratt: A. S. Pease 14086 GH; A. B. Seymour DUKE. PIKE: J. Davis 3785 ILL; J. D. Sauer 1618 ILL, WIS. Pope: E. J. Palmer 17018 MO. Putnam: V. H. Chase 10832 ILL, 10833 ILL. RAnpotpH: H. E. Ahles 5635 ILL. Rock Istanp: A. B. Seymour DUKE; B. Shimek IA. Sarnt Crater: FE. Douglass WIS; H. Eggert CU, CM, GH, KSC, MIN, MO, UC, US, WIS; G. Engelmann MO, WIS; A.S. Hitchcock MO; J. H. Kellogg MO; J. B.S. Norton WIS; L. H. Pammel WIS; J.A. Steyermark 843 MO; L. F. Ward US; W. Welsch ILL. SANGAMon: G. D. Fuller 7911 ILL. ScHuUYLER: J. D. Sauer 1618 ILL, WIS. Stark: V. H. Chase 714 MO, 4375 WIS. STEPHENSON: C. F. Johnson US. TAzEwE.t: V. H. Chase 3223 MO, 7914 ILL, 7915 ILL, 10900 ILL; A. B. Seymour DUKE. Wasasu: J. Schneck GH, ILL, 97 ILL; H. Shearer ILL. WinneEBAGO: R. B. Anthony WIS; M. S. Bebb IA, ILL, MICH, MO, OKL, WIS; S. C. Wadmond MIN. INDIANA (1854). BracKFrorpb: C.C. Deam 343 US. Crarxk: C. Mohr US. DEAr- BORN: C. C. Deam 50790 WIS, 55886 WVA. Evxuart: J. S. Brooks 1447 IND. Howarp: C. M. Ek CAS, 153 US. Jackson: C. C. Deam 38062 IND. Jay: C.C. Deam 59138 IND. JouHnson: H. M. Clarke WIS. Koscrusxo: C. C. Deam 21931 IND, 21984 IND; Yuncker & Welch 10666 GH, NO. LaxeE: C. C. Deam 1760 IND; J. M. Greenman 3192 MO; P. C. Standley 57472 IND, 57476 IND; L. M. Umbach US, 1376 WIS, 1381 WIS, 4678 WIS. Lawrence: R. M. Kriebel 1354 IND. Marion: R. C. Friesner 18125 GH, OC, WVA, 18126 OKL; W. Rhoades TENN; J. S. Wright NMC. Marsuatt: C. C. Deam 21010 IND; Scovell & Clark 1364 US. MONTGOMERY: E. W. Olive SDC. Morcan: P. Weatherwax IND. Newton: C.C. Deam 57326 IND. Nose: C. C. Deam 54495 IND. Owen: C.C. Deam 23928 IND; Haas & Welch 4971 ay) MADRONO [Vol. 13 IND, 4972 IND, 4973 IND; S. C. Hood 4025 FLAS. Porter: C.C. Deam 26526 IND; E.J. Hill 119 ILL; L. M. Umbach 5134 WIS. Posey: C. C. Deam 10062 IND, 22292 IND, 22297 IND, 22298 MIN, 22304 CAS, 24293 MT, 24303 MT. Putaskxt: C. C. Deam 46363 IND. RAnpotpH: C. C. Deam 15442 IND, MIN. Riretey: C. C. Deam 55894 IND. St. Josepu: J. A. Nieuwwland MT, US, 11511 MO. VANvDERBURG: H. M. Zelner IND. WARREN: C. C. Deam 51294 IND. WELLs: C.C. Deam IND, 467 IND. Waite: C. C. Deam 51252 IND. WuitteEy: C. C. Deam 59164 IND, OKL. IOWA (1873). Indefinite locality: J. C. Arthur 65 MO. ALLAMAKEE: W. L. Tolstead ISC. Benton: J. J. Davis WIS. BLAcKHAWK: M. Burk 581 MO, 874 ILL, 892 ILL, MO. Boone: G. M. Lummis ISC. BrEMEr: B. Shimek IA. Butter: J. D. Sauer 1696 WIS. Des Mornes: P. Bartsch IA. Dickinson: B. Shimek IA. DUBUQUE: A. Horr GH; W. A. Weber 1937 COLO. Emmet: R. I. Cratty IA, OC; F. W. Paige ISC; B. Shimek IA. Fayette: B. Fink GH, ISC, MIN, OC, US. Hamirton: B. Shimek TA. Hancock: B. Shimek IA. Harpin: L. H. Pammel ISC; B. Shimek IA. Jounson: A. S. Hitchcock IA, MO; B. Shimek IA; M. P. Somes 3910 US, 4400 ISC, 4428 ISC; 4429 ISC. Ler: J. L. Fults 1624 Gn part) ISC; 2. H. Pammel isC Pe: Rolfs ISC; B. Shimek IA. MuscaTinE: H. W. Clark US; F. Reppert IA; M. P. Somes 3722 MO. Story: Anonymous ISC; C. E. Bessey GH; A. Hayden 319 ISC; A. S. Hitchcock KSC, MO, MSC. Ware tito: L. H. Pammel ISC. Warren: L. H. Pam- mel ISC. WEBSTER: C. H. Churchill ISC. Winnesaco: L. H. Pammel ARIZ; B. Shimek IA. WINNESHIEK: H. Goddard ISC; W. L. Tolstead ISC. KENTUCKY (1840). Indefinite locality: C. W. Short CINC, MO. BrecKENn- RIDGE: McFarland, Plymale, & Schacklette 14 MT. Kenton: M. G. Williams WIS. LOUISIANA (1832). Indefinite locality: W. M. Carpenter MO, US, WIS. Avoy- ELLES: G. Ware WIS. Orteans: R.S. Cocks NO; T. Drummond 552 GH; J. F. Joor MIN, MO. RaApipes: J. Hale US. St. CHartes: J. Howard NO. *MAINE (1899). Oxrorp: W. H. Allen GH. *MASSACHUSETTS (1899). Hamppen: L. Andrews 6 GH. Mippresex: M. L. Loomis GH; E. F. Williams GH. WorcEsTER: Anonymous GH. MICHIGAN (1838). Indefinite locality: E. J. Cole MIN; C. F. Wheeler MIN, MSC. Berrien: C. Billington MICH, MSC. Cass: H. S. Pepoon 73 MSC. GENESEE: D. Clarke 23 CM, 466 (in part) MSC, 2023 MSC. Gratiot: C. A. Davis GH, ILL, MICH, MIN, MO, MSC, OC, POM, RM, TENN, UC, WIS. Incuam: L. H. Bailey, Jr. GH; C. F. Baker POM; W. J. Beal MSC; G. H. Hicks US; H.C. Skeels MSC; G. D. Sones MO; J. W. Toumey ARIZ, NDA; C. F. Wheeler MSC, WIS. Ionta: C.F. Wheeler POM, US. Kent: C. W. Bazuin 711 MSC; W. Boott GH; H.C. Skeels MSC. LEnAwWEE: C. H. Stocking MICH. MAcoms: D. Cooley MSC. Monroe: M. E. Day OC. Musxecon: C. D. McLouth MSC. Oaxtanp: C. Billington MICH. Ottawa: E. J. Cole MSC. Satnt Cratr: C. K. Dodge GH, MICH, MIN, MO, MSC, OC, RM, TENN, TEX, US, WIS, 266 US. Van Buren: L. M. Umbach WIS. WASHTENAW: C. Billington MICH; J. H. Ehlers 1452 MICH, 5503 MICH; F.J. Hermann 9178 MICH, MO, US; A.J. Pieters MICH. MINNESOTA (1848). Indefinite locality: C. C. Parry US. Aitkin: J. H. Sand- berg 767 MIN, MSC, US. Bettrami: Butters & Rosendahl 6578 MIN; P. Jones 422 MIN, 423 MIN. Cass: E. L. Nielson 1799 MIN. Cuippewa: L. R. Moyer 1410 MIN; J.B. Moyle 3255 MIN, 3505 MIN. Cray: H. F. Bergman NDA. CLEARWATER: M.L. Grant 3139 MIN. FREEBorRN: J. B. Moyle 4001 MIN. GoopHuE: A. P. Ander- son GH, UC, US, WIS, 772 MIN, 830 MIN; C. O. Rosendahl 6792 MIN; J. H. Sand- berg MIN. Hennepin: F. H. Burglehaus MIN; Butters & Rosendahl 3410 CU, MIN, 3413 MIN, 3414 MIN; E. A. Mearns 837 US; T. S. Roberts MIN. Houston: W. A. Wheeler 522 MIN, 547 MIN, 598 MIN. LEsueEur: J. B. Moyle 2901 MIN. Ramsey: E. Mearns 839 US; Moore & Moore 13631 GH, ISC, MIN. Rice: Moore & Moore 10283 MIN, 10284 IA, ILL, MIN, MT, UC. Stearns: F. W. Dewart MO. WABASHA: S. M. Manning MIN; L. H. Pammel ISC. Wasuincton: J. W. Moore 16048 MIN; SAUER: AMARANTHUS 23 1955] ‘sassora Aq poyeotpur syued yeodAyy ‘deur uornnqnysip :snqnjnssaqny snyjunamy *s “org v BLA] 2? ae | TT if See Le gas 1 = noche vc ry : ~ ye ! pie ne i ! / Ga ie H 1 a a 7 e (7 ' i f Zz @>. 622 4 ] 1 SS ae eS oe CL. aa SS SS ee | ' ye } & ' ! wn ee” NZ. a i) = J “ ) > Py na — 7 . Pies >< ] as Cc Fag e % Benn eee te De es { L7~NL wv ee | SUZLERONMNN & eens ee ooe 00% oor ° ; 4 e oy y \ i t,t J re - ate) nef 1 ” - x sid erik : \ el(dpw uypw sp eups epg) ---| fe D o° e2° é | : pret, a > y Rannen ai : eve ive e | Ary e cm vs Idd1881881" " e “leo g $ e: ad ! Re pha ne A cand ee ed UYUaMO1 | ee | bt) p x, e Sd Ase ae a | eg ee |e e° uf x 5 1? l ! e x my \ © nx { « \ % °, * ) ox eo *x SO as <@ ) ) oy { ~—wNr e wy Sak spe e x \ Ves aS SS —-~— 0 1 mers 5 t e x a ! ! e Pee x | é OF jpn as r ’ ey tact aa e -@ % e SSS fhe Me aes 0 &. 4 ‘ . Ae He \ x snyprnosegns -y > \ * Nag ; ! Dae me, ( e if ZR — —saee 24 MADRONO [Vol. 13 A. floridanus Fic. 6. Amaranthus floridanus: distribution map. Moore & Moore 10434 MIN, TEX, 10630 MIN. Winona: J. M. Holzinger CU, ISC, MIN, RM, US, UTC. MISSOURI (1832). Jackson: B. F. Bush 278 KSC, MO, US, WIS. PIKE: W. Trelease 816 MO. Sant Cuartes: E. Douglass MO. Sant Louis: T. Drummond GH, 38 GH, 311 (in part) GH; G. Engelmann GH, 255 MO, 257 MO, 258 MO; L. O. Overholts MO; J. A. Steyermark 8971 MO; W. Trelease MO, WIS. NEBRASKA (1853). Indefinite locality: F. V. Hayden MO; P. A. Rydberg NEB. Cepar: T. A. Bruhin WIS. CuHerry: J. M. Bates NEB; R. Thomson 83 US. Grant: Rydberg & Tulen 1644 NEB, US, 1674 (in part) GH, 1778 NEB, US. HooxeEr: R. Thomson 265 US. Pirrrce: N. F. Petersen NEB. NEW YORK (1900). Cavuca: M. R. Garner 19389 CU. OnonpDéAGA: Muenscher & Brown 22061 CU, GH; K. M. Wiegand CU. Ricumonp: A. Hollick US. Sant LAwRENCE: O. P. Phelps 418 CU, GH, US, 1152 CU, GH, 1153 CU, GH. Saratoca: Muenscher & Lindsey 3239 CU. Tompkins: F. P. Metcalf 6399 CU. WASHINGTON: S. H. Burnham GH. NORTH DAKOTA (1860). BENson: J. Lunell 408 MIN, 585 MIN. Cass: H. F. Bergman MO, MIN, NDA; O. A. Stevens NDA, 1399 NDA, WIS. O1iver: F. V. Hay- den MO. Ransom: Fieldstand 1097 NDA. Warp: O. Lakela 504 MIN. OHIO (1833). Indefinite locality: J. R. Paddock ILL; R. Peter MICH. Av- GLAIZE: A. Wetzstein OS. BUTLER: D. Demaree 27354 ISC. CHampaicn: W. A. Keller- 1955] SAUER: AMARANTHUS 25 man OS. CLermontT: E. L. Braun CINC; D. L. James OS; James & James OS. Cosuocton: F. B. Selby 209B OS. Erte: E. L. Moseley CM, US; W. Whitney OC. FAIRFIELD: W. Goslin OS; Goslin & Goslin OS. FRANKLIN: M. J. Murray 38140 CU; Schaffner & Brown 4 OS; W. C. Werner 110 (in part) GH, 111 GH. GREENE: M. Mohr 577 CINC. Hamitton: Anonymous CINC; E. L. Braun CINC; R. Buchanan CINC; 7. H. Kearney, Jr. OS; C. G. Lloyd US, 2376 GH, MICH, MSC; C. W. Short GH; M. G. Williams WIS. Hotmes: W. A. Kellerman OS. Locan: Kellerman & Beattie OS. Lucas: E. L. Moseley MICH. Metcs: W. A. Kellerman OS. MERCER: C. E. Thorne 1794 OS. Montcomery: B. Frank MO; S. E. Horlacher OS. OrTawa: E. L. Moseley GH, OS, US. Pickaway: Bartley & Pontius OS. Ross: Bartley & Pon- tius OS. SHELBY: Kellerman & Beattie OS; E. S. Thomas OS. Stark: Mrs. T. W. Case OS. WarreEN: E. B. Harger 8180 GH. Wayne: Bontrager & Greene OS. SOUTH DAKOTA (1839). Cray: W. H. Over 5136 US. STANLEY: C. A. Geyer 420 US. TENNESSEE (1877). DAvipson: A. Gattinger US; W. H. Seaman US; L. F. Ward MO, US. Montcomery: A. Clebsch 664 TENN. SHeEtsy: Norris & Sharp 16323 TENN, 16327 TENN. VERMONT (1882). Appison: E. Brainerd GH; C. E. Faxon GH. RuTLanp: D.L. Dutton CM, CU, DUKE, KSC, MO. WISCONSIN (c. 1845). ApAMs: WN. C. Fassett 22735 WIS. Brown: J. H. Schu- ette CAS, GH, IA, MICH, MIN, OKL, POM, UC, US, WIS. Burrato: R. H. Dennis- ton WIS. Dane: R. Burton 40 WIS; L. S. Cheney WIS; J. R. Churchill GH, MO, WIS; J.J. Davis GH, WIS; N. C. Fassett 22733 WIS, 22734 WIS; T.J. Hale WIS; J.R. Heddle 1430 WIS; G. N. Jones 17694 ILL; J. D. Sauer 1590 ARIZ, CAS, COLO, FLAS, MT, NCU, NDA, NMC, NO, SDC, TENN, UARK, 1592 WIS; L. H. Shin- ners 1460 WIS; E. E. Terrell 2422 WIS; W. Trelease MO, WIS; R. H. True WIS; J. H. Zimmerman 3426 WIS, 3608 WIS. Dopce: L. H. Shinners 1458 WIS. Fonp Du Lac: J. J. Davis WIS; Waters & Pammel ISC. Grant: R. B. Anthony MIN, WIS; J.J. Davis UC, WIS; N. C. Fassett 12620 WIS, 13649 WIS, 13656 WIS, 14168 WIS, 14170 WIS, 14201 WIS; H. H. Smith 7738 WIS, 7739 WIS. Green LAKE: Fas- sett & Sperry 18389 WIS; L. H. Shinners WIS. Iowa: J. J. Davis WIS; G. N. Jones 17633 ILL; C. T. Mason, Jr. 1489 WIS. LA Crosse: J. J. Davis WIS; D. S. Pammel ISC; L. H. Pammel ISC, UTC. Laravette: L. S. Cheney WIS. MARQUETTE: N. C. Fassett 15327 WIS. MitwavukeEe: I. A. Lapham GH, WIS; Morgan & Morgan 635 IA; W. W. Oppel C2 GH; R. Pohl 1343 MIN; L. H. Shinners 1457 WIS. Ovuta- GAMIE: P. O. Schallert 658 DUKE. Pepin: E. A. Baird WIS. Pierce: N. C. Fassett 5314 WIS. Ractne: J. J. Davis DUKE, ILL, WIS; J. R. Heddle 300 WIS; S. C. Wadmond MIN. RicuraAnp: E. K. Jones 406 WIS. Rock: G. B. Olds WIS; G. D. Swezey CAS. Saux: J. J. Davis WIS; Levi & Rose WIS; H. F. Luders US; A. B. Seymour DUKE. Sawyer: Gilbert & Fassett 8228 WIS. SHEBOYGAN: C. Goessl WIS. TREMPEALEAU: E. A. Baird WIS; N. C. Fassett 5311 WIS, 5312 GH, DUKE, WIS. WatwortH: G. R. Kleeberger CAS; S. C. Wadmond MIN. WavuKeEsHA: G. H. Corn- well WIS; I. Cull 315 WIS; H. C. Greene WIS. WaAupaca: Fassett & Rhodes 13169 WIS. 4. Amaranthus floridanus (S. Wats.) comb. nov. Acnida floridana S. Wats. Proc. Am. Acad. 17:376. 1882. This species was described from a homogeneous group of Florida col- lections (syntypes; J. L. Blodgett s.n., Key West, ¢ GH!, here designated as lectotype; A.W. Chapman s.n., sandy coast at North Clear Water Pass, 1875, 2 MO!; A. P. Garber s.n., South Florida, 1876, 2? GH!). Watson also mentioned a Curtiss collection without identifying it (probably A. H. Curtiss 115, Florida, 1880, 2 GH!). 26 MADRONO [Vol. 13 Plants slender, erect, with ascending branches, usually about 1 m. tall; leaf-blade very small, linear to narrowly oblong; flowering and fruiting from late spring through early fall; thyrses flexible, usually 10 to 20 cm. long, either all terminal on leafy branches or, if leafless branch thyrses present, these widely spaced and each subtended by a leaf; bract 1 to 114 mm. long, the midrib moderately heavy in ¢, heavy in 2, not con- spicuously excurrent in either; ¢ flowers with 5 stamens, the 5 tepals approximately equal, 2 to 214 mm. long, the inner obtuse or emarginate, the outer obtuse or acute, with midveins excurrent; 2 flowers usually with 1 or 2 tepals, the shorter tepal rudimentary, the longer 1 to 114 mm. long, narrowly lanceolate, acuminate, with excurrent midvein; utricle about 114 mm. long, indehiscent, thin or slightly fleshy, irregularly ru- gose, sometimes with faint ridges corresponding to the 3 style branches, often reddish; seeds 34 to 1 mm. in diameter, round, lenticular, dark red- dish brown. | The inadequate habitat data indicate that the species is a native of coastal dunes and beaches, but has become a local weed in gardens and fields near the coast. UNITED STATES. FLORIDA (ca. 1850). Indefinite locality: 4. W. Chapman MO, US; A. H. Curtiss MIN, 115 GH; A. P. Garber GH; J. B. McFarlin TEX; J. H. Simpson US. AtacHua: E. West FLAS. Brevarp: A. H. Curtiss 5775 CU, FLAS, GH, ILL, ISC, KSC, MIN, MO, MSC, NEB, POM, UC, US, WIS, 57754 GH. Duvat: A. H. Curtiss US. LEE: S. M. Tracy 7621 CU, CM, GH, ISC, MIN, MO, MSC, NEB, UC, US, WIS. Manatee: A. Cuthbert FLAS; J. B. McFarlin 6158 MICH; J. H. Simpson 43 MO. Monroe: J. L. Blodgett GH. Pinertas: A. W. Chapman MO. SARASOTA: A. W. Chapman GH, MO; A. H. Curtiss 2373 GH, MIN, US; J. B. Mc- Farlin 6184 CAS, 6185 UC. 5. AMARANTHUS TAMARISCINUS Nutt. Trans. Am. Phil. Soc., new ser. 5:165. 1837. Montelia tamariscina (Nutt.) A. Gray, Man. ed. 2. 370. 1856 (pro parte). Acnida tamariscina (Nutt.) Wood, Am. Bot. and FI. 289. 1870. Nuttall described this species as abundant on the sand beaches of the Arkansas and Grand (Neosho) rivers in what is now Oklahoma. His speci- mens, which I have not seen, are reported to be so immature that they show few diagnostic characters (Uline and Bray, 1895, p. 157; Gray, 1876, p. 489). However, the partial description and the locality make it reasonably certain that Nuttall was dealing with the species under con- sideration here. He could hardly have chosen a place closer to the heart of the range of this species, nor one more typical of its habitat. Plants usually stout and erect, with ascending branches, | to 2 m. high; leaf-blade usually oblong to lance-oblong; flowering and fruiting entirely during summer and fall, mainly July through October; thyrses stiff, usu- ally 10 to 20 cm. long, either all terminal on leafy branches or, if leafless branch thyrses present, these loosely arranged and each subtended by a leaf (by a leaf scar late in the season); bract 11% to 2 mm. long, with moderately heavy, excurrent midrib in 6, about 2 mm. long with heavy, 1955] SAUER: AMARANTHUS 27. excurrent midrib in 2; ¢ flowers with 5 stamens and 5 tepals, the inner tepals about 214 mm. long, obtuse or emarginate, outer tepals about 3 mm. long, acuminate, with conspicuous, excurrent midveins; 2 flowers with 1 or 2 tepals, the shorter tepal rudimentary, the longer about 2 mm. long, narrowly lanceolate, acuminate, with moderately heavy, sometimes branched excurrent midvein; utricle about 11% mm. long, circumscissile, thin, rugose, sometimes with faint ridges corresponding to the 3 or 4 style branches, often reddish; seeds about 1 mm. in diameter, round, lenticular, dark reddish brown. The species is at home on the margins of inland water bodies: river floodplains, streambanks, sandbars, muddy lakeshores, the edges of ponds, marshes. About two-thirds of all collections with habitat data are from such apparently natural sites. However, the species has very definite weedy tendencies and about one-third of the collections is from artificial habitats: roadsides, railroad rights-of-way, fields and gardens. Even in the heart of its range the species is a common weed where fields and ditches have invaded its native riverbottoms. The migrants along the margins of the main range and the isolated waifs which turn up sporadi- cally far outside the main range are almost invariably weeds of places disturbed by man. UNITED STATES. ALABAMA (1893). Mosite: C. T. Mohr US. ARKANSAS (1894). Indefinite locality: B. F. Bush 479 MO, 480 WIS. BENTON: B.F. Bush 15778 MO, NEB; D. M. Moore 29020 UARK. Cuicot: D. Demaree 14062 (in part) CAS, MIN, MO. Faurxkner: F. A. Haas 1064 US, 1065 US. INDEPENDENCE: W. Trelease MO. Jacxson: D. Demaree 20365 ISC, MO, 20368 WIS. JEFFERSON: D. Demaree 8768 CAS, US, 8775 MO, OC, US, 18589 MO. Lawrence: P. H. Rolfs ISC. Marion: D. Demaree 20602 MO. Miter: G. Ware WIS. Prairie: D. Demaree 18534 MO. Putasxt: D. Demaree 8325 CM, GH, MO, 8326 US, 8327 CM, 8356 DUKE, 8360 GH, WIS, 8428 CAS, MO; G. M. Merrill 659 DUKE, 999 UARK. *DELAWARE (1896). NEwcastie: A. Commons GH, MO: E. Tatnall US. *IDAHO (1897). PAYETTE: L. F. Henderson 2981 GH. ILLINOIS (1860). Indefinite locality: W. E. Andrews ILL. ApaAms: R. Brinker 22989 ILL. CHAMPAIGN: H. M. Franklin ILL; G. N. Jones 13129 ILL, 17498 ILL. CHRISTIAN: Winter & Sauer 1621 WIS. Cuinton: H. E. Ahles 5715 ILL. Coox: O. E. Lansing, Jr. 2632 GH; W.S. Moffatt 276 ILL, 483 ILL; W. C. Ohlendorf OC. Douc- LAS: G. S. Winterringer 12 ILL, 19 ILL. Grunpy: G. N. Jones 17760 ILL. KAne: J.D. Sauer 1600 WIS. La Sate: G. D. Fuller 3256 (in part) ILL. Pike: J. D. Sauer 1619 WIS. Sartnt Cratr: E. Douglass MO; H. Eggert MO; G. Engelmann MO, WIS; J. B.S. Norton WIS. Tazewetr: V. H. Chase 8589 ILL. Vermition: G. N. Jones 18793 TLL. INDIANA (1896). Indefinite locality: E. R. Drew UC. Jounson: H. M. Clarke IND, WIS. LAxe: W.S. Moffatt 468 WIS, 504 ILL. L. M. Umbach WIS, 857 WIS, 977 US, 1553 WIS. IOWA (1877). ApAms: M. J. Fay 4315 IA, WIS. Aupuson: B. Shimek IA. BoonE: Anonymous ISC; L. H. Pammel ISC, TENN. Buena Vista: W. F. Couch 103 ISC; J. D. Sauer 1693 WIS; B. Shimek IA. Butter: J. D. Sauer 1697 WIS. CaLyoun: B. Shimek IA. Carrotit: M. J. Fay 3997 IA, 4007 IA, WIS, 5346 IA. Cass: M. J. Fay 3436 IA. Cerro Gorpo: Pammel & McNider 1093 ISC. Cray: A. Hayden 4007 GH, ISC, MIN, MO, US, 4008 ISC, US, 7180 ISC, 7469 ISC; B. Shimek TA. Decatur: J. P. Anderson ISC, MO, RM. Dickinson: H. S. Conard TENN; 28 MADRONO [Vol. 13 R.I. Cratty ISC; B. Shimek IA. Emmet: F. W. Paige ISC; B. O. Wolden 727 ISC. Fremont: M. J. Fay 4132 IA, 4140 IA, WIS, 4570 IA, 4576 IA, 5554 IA. GREENE: J.D. Sauer 1677 WIS; B. Shimek 1A. Guturte: M. J. Fay 2117 IA, 4185 IA, WIS, 5326 IA. HAamitton: A. Hayden 10329 GH, ISC; L. H. Pammel ISC. Harrison: M. J. Fay 3707 IA, WIS, 3721A IA; B. Shimek IA. Jerrerson: McDonald & Gilly 2122 ISC. Jounson: M. P. Somes 4401 ISC. Kossutu: J. C. Blumer 4421 ISC; R. I. Cratty ISC; Pammel & Cratty ISC. Lee: J. L. Fults 1624 (in part) ISC; L. H. Pammel MICH; B. Shimek IA. Lyon: L. H. Pammel ISC; B. Shimek IA, ISC. Manpison: Blosser & Blosser 162 ISC; M. J. Fay 4999 IA. MAnaAsKa: D.W. Augustine 437 ISC, 465 ISC, 466 ISC. Marswatt: F. C. Stewart ISC. Mis: M. J. Fay 3571 IA. O’Brien: B. Shimek IA. Pace: M. J. Fay 3794 1A, 4179 IA, WIS; L. H. Pammel ISC. Pato Arto: A. Hayden 4009 GH, MO, 4010 ISC, 7181 ISC, 7182 ISC. PLlyMouTH: M. E. Jones POM. Pocanontas: J. D. Sauer 1695 WIS. Pork: E. Anderson WIS; L. H. Pammel ISC; Pammel, Frankel, & Rieman 1001 GH,ISC; B. Shimek IA. Potta- WATTAMIE: W. Cleburne NEB; F. Eastman 442 NEB. Sac: B. Shimek IA, ISC. Scott: M.P. Somes 36601 US. Story: C. £. Bessey FLAS, ISC; UEC AR. Burgess ISC; G. W. Carver ISC, MO; R. I. Cratty ISC; A. Hayden 462 ISC; D. Isely 3932 SMU; C.C. Lorensberry ISC; C. E. Maxwell ISC; L. H. Pammel ISC; J. W. Parsons ISC; J. D. Sauer 1675 WIS. Tama: S. Rouse ISC. Taytor: M. J. Fay 3789 IA. Van Buren: J. Fults ISC; M. McDonald 1170 ISC, 1194 ISC; L. H. Pammel ISC. Wa- PELLO: L. H. Pammel ISC. Warren: L. H. Pammel ISC. WEBSTER: D. W. Augustine 708 ISC; J. C. Blumer 4454 ISC. WINNEBAGO: B. Shimek IA. WricHt: R. B. Moor- man ISC; B. Shimek IA. KANSAS (1847). Indefinite locality: G. R. Kleeberger CAS; E. A. Popinoe US. ALLEN: A. S. Hitchcock KSC. Anprerson: A. S. Hitchcock KSC. Atcuison: A. S. Hitchcock KSC. Bourson: A. S. Hitchcock KSC. Brown: Clothier & Whitford KSC. Cuase: F. E. Bray KSC. Cuautavgua: A. S. Hitchcock KSC. CHEROKEE: Clothier & Whitford KSC; A. A. Jacobs KSC. Cray: W. A. Kellerman KSC. Croup: S. V. Fraser KSC, 674 KSC. Correy: Clothier & Whitford KSC. Cowtrey: A. S. Hitchcock KSC. Crawrorp: Clothier & Whitford KSC. Decatur: J. D. Sauer 1686 WIS. Dicx- Inson: J. B.S. Norton KSC. DonipHAn: Clothier & Whitford KSC. Douctas: W. A. Kellerman KSC; R. L. McGregor 562 KSC, 732 KSC; A. M. Philips DUKE; W. C. Stevens US. Epwarops: A. Finch KSC. Erk: Clothier & Whitford KSC. ELLSworTH: M. Becker KSC. FRANKLIN: A.S. Hitchcock KSC; A. M. Philips TEX. Geary: F.C. Gates 18695 MO, 19978 KSC, NEB. GrEENwoop: A. S. Hitchcock KSC. Harvey: Clothier & Whitford KSC. JAckson: M. Reed KSC. JEFFERSON: G. L. Clothier KSC. Jounson: H. L. Pellet KSC. Kincman: A.S. Hitchcock KSC. LABETTE: A. S. Hitch- cock KSC. Lincoin: A. S. Hitchcock KSC. Linn: A. S. Hitchcock KSC. Lyon: A. M. Philips TEX, US. Marton: A. S. Hitchcock KSC. MaArsHALL: A. S. Hitchcock KSC. McPue_erson: J. E. Bodin MIN; A. S. Hitchcock KSC. Montcomery: A. S. Hitch- cock KSC. Morris: A. Fendler 737 GH, MO; A. S. Hitchcock KSC. NEMaAna~: A. S. Hitchcock KSC. NeosHo: F. Broadbent KSC. Ness: A. S. Hitchcock KSC. Norton: J.D. Sauer 1685 WIS. Osace: Z. D. Brown KSC. OssorneE: C. L. Shear 211 (in part) NDA, RM. Pottawatomise: F. C. Gates 16301 KSC, MO; H. H. Laude MO; Norton, Clothier & Pond KSC. Repustic: D. K. Thomas KSC. Rice: A. S. Hitchcock KSC. Ritey: S. J. Adams KSC; W.T. Allen US; H. F. Bergman NDA; F. C. Gates 12749 MT, 13656 OC; A. S. Hitchcock KSC, 349 GH; W. A. Kellerman KSC, MO; J. B.S. Norton KSC, MO, WIS, 428 GH, KSC, MO, NMC, RM, US; D. Otis KSC; A. W. Staver KSC; J. Walquist KSC. Russety: A. S. Hitchcock KSC. Satine: J. Hancin 617 KSC, 696 KSC, 700 KSC, 2197 MO, 2199 WIS, 2200 WIS, 2203 MO, 2205 MO, 2211 MO; J. B. S. Norton KSC; M. Reed KSC; H. W. Ryding 10 US. SEDGWICcK: T. L. Andrews ISC; A. S. Hitchcock KSC; S. F. Poole 16 GH, 17 GH. SHAWNEE: Harper & Harper CU; J. Lockhart KSC; R. B. Smyth 1403 KSC. SuMNeER: Clothier & Whitford KSC; A. S. Hitchcock KSC. WasaunsEE: Norton & Clothier KSC. WasuinctTon: F. E. Gwin KSC. Witson: W. H. Haller KSC. Woopson: Clothier & Whitford KSC. WyanpvotTTeE: K. K. Mackenzie KSC. 29 AMARANTHUS SAUER: 1955] “——— = ec e@ ve WN yaeN maa ~ aon” © ° = $s .~) x 8 & u ~ x Fic. 7. Amaranthus tamariscinus; distribution map. Atypical plants indicated by crosses. 30 MADRONO [Vol. 13 LOUISIANA (1952). Cappo: G. Ware WIS. NATcuiTocHEs: G. Ware NO, TEX, WIS. *MAINE (1906). PEnosscot: O. W. Knight 5257 GH, US. *MASSACHUSETTS (1882). MippLesex: F. S. Collins GH, MIN, MO. Sur- FOLK: C. E. Perkins MIN. MINNESOTA (1943). Winona: Moore & Neva 16160 MIN. MISSOURI (1841). Atcuison: B. F. Bush MO. Bates: F. P. Metcalf 986 US. Crark: F. Drouet 1723 GH; S. B. Mead ILL; L. H. Pammel ISC. Dattas: J. A. Steyermark 13724 MO. DEKALB: E. J. Palmer 43749 MO. FRANKLIN: J. D. Sauer 1614 NO, WIS. Gentry: J. A. Steyermark 14968 MO. GREENE: P. C. Standley 8652 US, 9091 US. Jackson: B. F. Bush 12 US, 50 MO, 220 MO, US, 224 WIS, 227 KSC, MO, 313 GH, MO, 399 WIS, 444 MIN, MO, 446 MO, 516 MIN, MO, 1909 GH, MO, UC, US, 8175 ILL, MO, US, 8803 MO, 8845 WIS, 8878 ILL, UC, US, 13781 WIS; F. Bush 8 MO; R. Hoffman MO, WIS; K. K. Mackenzie ISC, KSC, MIN, RM. JaAs- PER: D. Demaree 4412 UARK; E. J. Palmer 425 MO, 1302 MO, WIS. 1560 MO. Jerrerson: J. A. Steyermark 8287 MO. LAFAYETTE: Kluhsmann & Trusik 36 COLO, OKL, RM, TEX; Trusik & Busch 11 TEX. Lincoun: J.D. Sauer 1615 NO, WIS; J.A. Steyermark 8938 M®. Livincston: S. Sparling 342 ISC, 539 ISC. McDonatp: B. F. Bush 15773 MO, NEB. Morcan: J. A. Steyermark 13154 MO. Perry: J. A. Steyer- mark 14062 MO. Puetps: J. H. Kellogg 192 MO, WIS. Pike: J. D. Sauer 1617 NO, WIS. Rattis: J. A. Steyermark 25808 MO. Saint Crater: J. A. Steyermark 7598 MO, 24320 MO. Saint Louis: G. Engelmann MO; N. M. Glatfelter US; J. H. Kellogg 885 MO; O. S. Ledman 9 MO; E. Mische MO; J. B.S. Norton MO, WIS; J. A. Steyer- mark 513A MO, 514 MO, 8869 MO, 8901 MO, 8902 MO, 8905 WIS, 8970 WIS, 8992 MO, US, 8996 WIS, 9020 MO. Satine: J. A. Steyermark 9342 MO, WIS, 9360 MO, 9361 WIS, 9375 MO, 9389 MO, US, 14811 MO. Warren: J. D. Sauer 1611 WIS. NEBRASKA (1860). Indefinite locality: W. C. Knight NEB, RM. Cass: T. A. Williams SDC, US. Crepar: T. A. Bruhin GH. Dopce: J. M. Bates 1569 (in part) NEB; G. C. Engberg NEB; W. Kiener 17734 GH; J. D. Sauer 1678 WIS. Dovuctas: W. Cleburne NEB; M. R. Gilmore MICH; Wiegand & Castle 825 CU. FRANKLIN: H. Hapeman DUKE, RM; J. D. Sauer 1683B WIS. Harr: J. D. Sauer 1679 WIS. HaAmitton: W. Kiener 15047 GH. Kearney: H. Hapeman OC. Knox: F. Clements 2755 (in part) NEB. Lancaster: J. M. Bates 4399 GH, NEB; U. G. Cornell NEB; W. Kiener 16989 GH; J. W. Morrow WIS; J. L. Sheldon WVA; J. G. Smith NEB; M.P. Somes ISC; O. E. Sperry NEB; C. A. Turrell ARIZ; H. J. Webber MO, NEB, 5325 NEB. Lincoin: F. Eastman 1464 RM. Nemana: J. M. Bates 3291 NEB; W. W. Hansen MICH; J. M. Winter 447 OC. Nucxotts: G. G. Hedgcock MO, WIS; W.L. Tolstead NEB. OToE: Anonymous NEB; G. G. Hedgcock MO. Pratte: J. D. Sauer 1692 WIS. Pork: E. A. Boostrom NEB. Sarpy: W. Cleburne NEB. SEwarp: W. Kie- ner 17087 MO. WEssTER: J. M. Bates GH, NEB, 3122 NEB, 3124 GH, NEB; W. L. Tolstead NEB. Yorx: W. Kiener 17840 GH. *NEW JERSEY (1879). Hupson: A. Brown US. *NEW MEXICO (1907). Roosevett: A. D. Stowell NMC. *NEW YORK (1924). Tompkins: S. H. Burnham 17854 CU, 19860 CU; Burn- ham & DeFrance 16950 CU, MO; W.C. Muenscher 15457 CU; C. L. Pratt 18521 CU. NORTH DAKOTA (1890). Cass: H. F. Bergman KSC; O. A. Stevens CAS, NDA, OC, RM, 753 GH, MO, NDA, US; L. R. Waldron 1306 NDA. Foster: W. P. MacDonald NDA. Suope: O. A. Stevens NDA. *OHIO (1871). LAKE: H.C. Beardslee OS. OKLAHOMA (1874). Indefinite locality: E. DeBarr 286 MSC; E. Palmer 277 MO. Cuoctaw: Hopkins & Cross 2270 OKL. CLEVELAND: W. H. Emig 385 US; M. Hopkins 778 OKL, 779 OKL, 29270 OKL; R. E. Jeffs OKL, WIS; Perkinson OKL. CoMANCHE: Mrs. J. Clemens 11560 GH, MO, OKL. Cratc: J. T. Monell MO. DEta- warE: R. Bebb 3698 OKL; B. F. Bush 15788 MO. GarFietp: H. B. Gephardt 564 US, 696 OKL, 1115 US. Kay: G. W. Stevens 1845 GH, ILL, MIN, OKL. KincFIsHER: R. Bollenbach 99 OKL, 106 OKL. McCurtain: U. T. Waterfall 9829 SMU. Murray: 1955] SAUER: AMARANTHUS 31 M. Hopkins 2023 OKL; Hopkins & Demaree 22 OKL; Hopkins & Van Valkenburgh 5424 OKL, 5474 OKL; Merrill & Hagan 821 TENN; G. T. Robbins 2728 OKL. MuskoceEe: R. Bebb 4994 GH, OKL; E. L. Little, Jr. 2064 OKL, 2067 OKL, 2717 OKL, 2804 OKL. OxLaHoma: D. Demaree 13272 GH, MO, 13273 OKL; B. Shimek IA; U. T. Waterfall 3560 OKL; S. S. White 1148 GH, 1181 GH. Payne: C. R. Atkins 16 SMU; J. W. Blankinship GH; K. Bradley 20 TEX; M. Cox 61 MO; J. M. Dyer 87 ARIZ, COLO; J. H. Kimmons GH, MO, US; R. E. Penn 64 WVA; N. R. Poteat 61 OKL; I. Sooter 90 TEX; F. A. Waugh 162 KSC, 167 MO. Tutsa: R. Luckhardt 160 OKL. Waconer: G. Ware WIS. *PENNSYLVANIA (1908). CHESTER: J. J. Carter MT. SOUTH DAKOTA (1891). Beanie: J. J. Thornber SDC. Brooxincs: Anony- mous SDC; M. Folds FLAS. Cray: W. H. Over 5137 US; S. S. Visher 4111 MO. Hucues: F. H. Sargent NDA. Kincssury: J.J. Thornber MO, SDC, UC, WIS, WVA. STANLEY: T. A. Williams 33 US. YAnxtTon: L. A. Bruce 65 US. TENNESSEE (1922). Knox: A. R. Bechtel CU, 10868 CU. LAKE: R. E. Shanks 13693 TENN. SHELBY: D. Demaree 19675 MO, 19723 MO. Tipton: Sharp, Fair- child & Clebsch 8067 TENN. TEXAS (1845). Indefinite locality: T. Drummond 240 (in part) GH. BELL: B. Mackensen 238 MO. Bexar: F. Lindheimer 142 GH, MIN, MO, US, WIS. Brazoria: B. F. Bush 1562 MO. Cameron: Bogusch & Molby 4190 ILL; R. Runyon 29 US, 4245 SMU, 4246 ARIZ, SMU, TEX. Cray: V. L. Cory 40741 GH; L. H. Shinners 12866 SMU, 12873 SMU. Cortin: L. H. Shinners 11712 SMU. Comat: F. Lind- heimer MO, SMU, 285B GH, MO, 456 GH, MO. Cooke: E. Whitehouse 19262 SMU, UC, US. Dattas: H. Eggert MO; H. L. Graham ISC; M. Hynes TEX; G. Letterman US, 93 WIS; C. L. Lundell 11653 GH, SMU; Lundell & Lundell 9656 GH, MICH, SMU; J. Reverchon GH, 829 MO, US; L. H. Shinners 10367 SMU; R. Van Vleet 334 SMU. Denton: Anonymous TEX; L. H. Shinners 9395 SMU, 11894 SMU; E.White- house 17426 SMU. Extis: L. H. Shinners 16727 SMU. Gatveston: G. L. Fisher UC, WIS, 613 US, 2066 US. GittEspiIE: F. Lindheimer 54 GH, MO. Grayson: E. White- house 17451 SMU, UC. Harris: G. L. Fisher WIS. Hipatco: V. L. Cory TEX. Hoop: L. H. Shinners 19089 SMU. Jackson: B.C. Tharp GH. Kaurman: V. L. Cory 52551 SMU. Lamar: G. Ware WIS. Maracorpa: V. L. Cory 11530 GH; B.C. Tharp 1593 TEX, US, 1594 TEX. McLennan: L. D. Smith 235 TEX, 268 TEX; C.L. York 46252 TEX; York & Smith 149 TEX. Navarro: J. F. Joor MO, US. SAN Patricio: F. B. Jones 352 SMU. Tarrant: G. W. Letterman 100 GH, MO; A. Ruth 233 CM, GH, IE ESMIN, SMU, US, WIS, 710 US, 1043 CU, GH, ISC, NDA, US; L..F. Ward US. Taytor: W. L. Tolstead SMU, 7734 GH. Travis: A. A. Armer 5387 TEX, US; J. E. Bodin 245 US; A. M. Ferguson 458 TEX; R.H. Painter 78 KSC; B.C. Tharp MICH, 892 TEX, US, 1276 TEX, US, 1388 TEX, US, 1501 TEX, US; Tharp & Barkley 15558 TEX; M.S. Young TEX. VALVERDE: C. Wright 582 (in part) GH, 1747 GH, US. Wasuincton: E. Brackett GH. Woop: E. Whitehouse 16482 SMU. ZAvALa: H.R. Reed GH. *WASHINGTON (1895). Kricxitat: W. N. Suksdorf 3676 GH. WISCONSIN (1914). MirwavukEE: H. P. Sartwell GH. SHEBOYGAN: C. Goessl WIS. 6. AMARANTHUS PALMERI S. Wats. Proc. Am. Acad. 12:274. 1877. Amaranthus Palmeri S. Wats. var. glomeratus Uline & Bray, Bot. Gaz. 19:272. 1894 (pro parte). Watson described this species from pistillate plants and gave no definite citations of staminate specimens (syntypes: J. L. Berlandier 2407, banks of Rio Grande, July, 1834, GH!, here designated as lectotype; also in MO!,US!; Dr. Edward Palmer 323, Larkin’s Station, San Diego County, California, 1875, GH!, MO! ). However, in both the original description 32 MADRONO [Vol. 13 and in a later publication Watson (1880, p. 42) suggested that certain staminate collections might represent the same species (Edward Palmer s.n., 80 miles east of San Diego on Fort Yuma road, August, 1875, GH!; L.J. Xantus 100, Cape San Lucas, Baja California, 1859-60, GH!, US!). It is fortunate that Watson did not cite the Xantus specimen as repre- sentative of his species, because it does not belong to the same taxon as the other specimens and has served as the type of various other taxa to be discussed below. The Berlandier specimen is an ideal representative of the species under consideration here; the Palmer specimens, although well within the limits of the species, are less typical and may show traces of mixing with A. Watsoni. The variety glomeratus is based on two speci- mens (syntypes: Edward Palmer 953, caespitose form producing great mats in dry places, river bottom, Colonia Lerdo, Sonora, April 26, 1889, 2 GH!, MICH!, US! ; 955, same place and date, ¢ GH!, US! ), the pistil- late plants being here assigned to A. Palmeri and the staminate to A. Wat- sont, although both are probably from a hybrid swarm of the two species. Plants stout and erect with ascending branches, usually about 1 m. but occasionally 2 or 3 m. tall; leaf-blade rhomboid-lanceolate, petiole strik- ingly long, about equalling blade; flowering and fruiting mainly in sum- mer and fall but occasional individuals flower in all months in the United States border states and in Mexico; thyrses flexible or moderately stiff, usually 20 to 30 cm. long, either all terminal on leafy branches or, if leaf- less branch thyrses present, these loosely arranged and each subtended by a leaf; bract usually 4 to 6 mm. long; midrib excurrent, moderately heavy in ¢, very heavy in 2; 6 flowers with 5 stamens, 5 tepals, the inner tepals 214 to 3 mm. long, obtuse or emarginate, the outer tepals 314 to 4 mm. long, acuminate, with conspicuous, long-excurrent midveins; 2 flow- ers with 5 recurved tepals, each with conspicuous, branched midvein, the inner tepals usually 2 to 24% mm. long, spatulate, emarginate, slightly denticulate, the outer usually 3 to 4 mm. long, acute, with midvein ex- current as rigid point; utricle 114 to 2 mm. long, circumscissile, thin, somewhat rugose; style branches usually 2, sometimes 3; seeds 1 to 1% mm. in diameter, obovate, lenticular, dark reddish brown. The species is at home along permanent or intermittent streams. About one-third of the collections bear such notations as: creek bank, river flood- plain, canyon bottom, arroyo floor, edge of marsh, by spring. In such places it commonly grows in silt as well as in sandy and gravelly soil. This is probably the weediest of all the dioecious amaranths and the only one in which collections from natural habitats are outnumbered by collections from artificial habitats: irrigation ditches, roadsides, railroads, dumps, fields, and gardens. Such sites are reported for almost all collections made on the margins or outside the coherent range of the species as well as for more than half of the collections from the heart of the range. The species shares various common names with the weedy monoecious species: pig- weed, careless weed, redroot in the United States, quelite and bledo in Mexico. Amaranthus Palmeri has been an important food plant, both as a SAUER: AMARANTHUS 210) 1955] ‘sasso1d Aq pojzeotpul sued pestdAzy “deul uorynqiiysip -2amppg snyJUvADULP °3 OTA S¥aLZROIDI aw. ) \ \ ' \ Nera t \ 1 4 aS - -- 34 MADRONO [Vol. 13 potherb and as a source of grain, for various Indian tribes: Mojave and Chemehuevi of Arizona (J. T. Kelly s.n., CAS! ), Papago of Arizona (Cas- tetter and Bell, 1942, pp. 61-62), Cocopa of Sonora (£. Palmer s.n., US!; Castetter and Bell, 1951, p. 189), Tarahumare of Chihuahua (C. V. Hart- mann 567, GH!), and others (Watson, 1889, p. 71). There is no evidence that the species has ever been systematically cultivated, although it may have been spread unintentionally by seed gathering. MEXICO. BAJA CALIFORNIA (1887). Comonpvu: Carter & Kellogg 3099 UC, WIS, 3100 UC, WIS; B. J. Hammerly 171 CAS, GH. Ensenapa: E. Palmer ARIZ, CAS, COLO, CU, FLAS, GH, ILL, KSC, MIN, MO, MSC, NEB, NMC, OC, POM, RM, SDC, TEX, UC, UTC, WIS. Mutect: H. Aschmann WIS. CHIHUAHUA (1885). AQuiLes SERDAN: E. H. Wilkinson ISC, OC, UC, US. Ascension: E. W. Nelson 6420 GH, US. Batopitas: C. V. Hartman 567 GH; E. Palmer V GH, US. Curnvuanua: E. Palmer 203 US, 350 US; C. G. Pringle 1112 MIN, US. Gateana: J. D. Sauer 1000 MO. Janos: E. A. Mearns 320 US; S. S. White 1102 GH, 1113 GH, MICH, 2519 ARIZ, MICH. Jimenez: S. S. White 2096 MICH. JUAREZ: B. Shimek ITA. MANUEL BeEnavipEs: I. M. Johnston 7972A GH. SAUCILLO: S. S. White 2288 GH, MICH. COAHUILA (1940). SrerRA Mosapa: Johnston & Muller 1050 GH, 1252 GH. COLIMA (1897). Cotmma: E. Palmer 152 US. DURANGO (1896). SANTIAGO PAPASQUIARO: FE. Palmer 428 GH, MO, UC, US. TEPEHUANES: G. L. Fisher 44234 GH. MO. TLAHUALILLO DE ZARAGOZA: H. Pittier 510 US: GUERRERO (1934). Coyvuca DE CaTaALaAn: G. B. Hinton 5532 GH, MO. JALISCO (1886). GuapaLayaRA: E. Palmer 624 GH, 627 US, 628 GH, MICH, US; C. G. Pringle 11314 GH, US. LAcos pE Moreno: C. O. Sauer 15 MO. MEXICO (1932). TEMASCALTEPEC: G. B. Hinton 2482 GH, US. NAYARIT (1892). Acaponeta: J. N. Rose 3127 US. Tepic: M. E. Jones 23291 MO; E. Palmer US. PUEBLA (1947). Acatitan: J. D. Sauer 1143 WIS. HuaQuecHuta: J. D. Sauer 1135 WIS. QUERETARO (1913). QUERETARO: Bros. Arsene & Agniel 10543 US. SAN LUIS POTOSI (1876). San Luts Portost: L. de la Rosa 1010 ILL; Parry & Palmer 786 (in part) GH, MO, US; J. G. Schaffner 885 GH, 886 (in part) GH; L.F. Ward US. SINALOA (1891). Indefinite locality: J. G. Ortega 4211 US. AHoME: E. Palmer 190 ARIZ, MICH, UC, US; Rose, Standley, & Russell 13347 US. Curtacan: T. S. Brandegee UC; H. S. Gentry 4939 ARIZ, MICH, MO, UC; E. Palmer 1721 GH, US; Rose, Standley, & Russell 14980 US. Det FUERTE: Rose, Standley, & Russell 13416 US, 13595 US. Mazatian: Rose, Standley, & Russell 14042A US. Rosario: Ferris & Mexia 5079 GH; J. N. Rose 1823 GH, US. SONORA (1869). Indefinite locality: C. Lumholitz 40 GH; E. Palmer US. Acua PriETA: S. S. White 4107 GH. AtAmos: H. S. Gentry 4851 ARIZ, GH, MICH, MO; Rose, Standley, & Russell 12665 US, 13031 US. Bacerac: S. S. White 2731 GH, MICH, 3807 GH. Bavispe: S. S. White 447 GH, 762 ARIZ, GH, MICH, 2889 ARIZ, GH, MICH. Casorca: E. Palmer 953 GH, MICH, US; G. Sykes US, 24 US. CucurPE: I. L. Wiggins 7164 GH, MICH, US. Guayrmas: E. Palmer US, 76 GH, 77 ARIZ, US, 78 MICH; US, 120 ARIZ, 127 GH, US, 128 US, 129 US; 130 US; 131-US,, 132 US, 133 US, 134 GH, US, 688 ARIZ, GH, MICH, MO, UC, US. HEermosIL1o: Wiggins & Rollins 148 GH, MICH, MO, UC, US. Macpatena: W. B. Kibbey 7052 US. NAcozari DE GArcIA: S. S. White 3962 GH. Nocates: D. Griffiths 6753 MO. 1955] SAUER: AMARANTHUS 35 UNITED STATES. ARIZONA (1865). Indefinite locality: W. W. Jones UC; O. Loew US; E. Palmer US; C. G. Pringle US; C. Smart GH, US. Cocuise: J. C. Blumer 1594 ARIZ, GH, ISC, KSC, MIN, MO, NEB, NMC, RM, US; C. B. Carter ARIZ; O. M. Clark 8613 ARIZ; D. Griffiths 1585 ARIZ; M. E. Jones 4224 MSC, OC, POM, US, UTC; J. G. Lemmon 465 GH, 2878 GH; E. A. Mearns 599 US, 625 US, 629 US, 664 US, 809 US, 810 US, 1096 US, 2006 US; J. J. Thornber ARIZ. Coconino: E. U. Clover 5075 SMU; Eastwood & Howell 6978 CAS; J. T. Howell 26562 ARIZ, CAS. Gira: J. W. Toumey 427 US. Grauam: R. H. Peebles 14508 US; J.T. Rothrock 379 GH, US. GREENLEE: A. Davidson UC. Maricopa: J. B. Feudge 2077 POM; J. J. Thornber ARIZ; S. E. Wolff 1847 US. Pima: J. J. Carlson CAS; R. Darrow GH; L. H. Dewey US; G. Engelmann MO; H. S. Gentry 5963 CAS; F.W.Gould 2901 (in part) ARIZ; H. W. Graham CM;; D. Griffiths 5897 US, 5920 US, 5927 US; J. A. Harris C16556 MIN; M. E. Jones POM; C. D. Marsh US; K. F. Parker 8053 ARIZ; A. R. Phillips 10 FLAG; C. G. Pringle CINC, CM, GH, MO; F. Shreve 4908 MICH; Sister Thomas-Marie 477 CM, RM; J. J. Thornber ARIZ, 201 MIN, MO, NMC, POM, UC, 2543 ARIZ; J. W. Toumey ARIZ, GH, UC, US; L. C. Whitehead ARIZ; W. F. Wickham IA; I. L. Wiggins 6992 MO, US. PINAL: R. H. Peebles 10011 CAS, UC, 10593 POM; J. J. Thornber ARIZ. SANTA CRUz: Darrow & Haskell 2006 ARIZ; D. Griffiths 6134 MO, US; M. E. Jones 22639 MO; Peebles, Harrison, & Kearney 5578 US; J. J. Thornber ARIZ. YAvapat: Coues & Palmer 169 MO, 568 MO; E. Jackson ARIZ; W. W. Jones UC, 249 GH; D. Mc- Dougal US; J. D. Sauer 1661 WIS, 1662 WIS. Yuma: A. Beard MO; I. T. Kelly CAS; Twist & Kelly 7 GH. CALIFORNIA (1875). Indefinite locality: E. E. Schellenger 84 UC. IMPERIAL: Macbride & Drouet 4510 KSC; R. McKee 13 US; E. Palmer GH; S. B. Parish 8269 GH; C. B. Wolf 2280 CAS, GH, POM, UC, 4336 COLO, UC. Inyo: M. F. Gilman 4310 POM. Los AncEtEs: L. C. Wheeler 1355 NO, POM, 1418 UC, 2102 CAS, POM. RIvERSIDE: H. M. Hall 8009 UC; L. S. Rose 36825 CAS; J. D. Sauer 1668 WIS. SAN BERNARDINO: J. M. Johnston GH, POM, RM;; S. B. Parish 10971 ARIZ, MIN, POM, UC; F. M. Reed 3683 RM, MO, UC; J. Roos 1895 POM. San Dreco: E. A. Mearns 2965 US; C. R. Orcutt 2066 US; E. Palmer 323 GH, MO. *StTaniIsLAus: R. F. Hoover 697 UC. *TLLINOIS (1896). Coox: A. Chase ILL. *KANSAS (1895). WyAnpoTTE: K. K. Mackenzie KSC, MO. LOUISIANA (1929). NAtcHITOCHES: G. Ware TEX, WIS. OrteAns: J.J. Mor- rison NO, WIS. *MISSOURI (1897). Jackson: B. F. Bush 200 GH, MIN, MO, UC, US, 8798 CAS, GH, ILL, MIN, MO, RM, US, 8800 GH, ILL, POM, US, 8877 CAS, GH, ILL, MIN, US, 9213 ILL, MIN, UC, 9475 ILL, MIN; O. C. Durham 73 FLAS; K. K. Mackenzie ISC, KSC, MIN, MO. NEW MEXICO (1881). Indefinite locality: E.O. Wooton 2727 US. BERNALILLO: Rose & Fitch 17798 US. Catron: E. A. Goldman 1579 US; E. O. Wooton US. Curry: H.C. Reynolds 276 NEB. Dona Ana: J. H. Bruce NMC; D. B. Dunn 8568 ARIZ, CAS, COLO, DUKE, FLAS, IA, IND, MT, NCU, NDA, NMC, NO, OC, SDC, TENN, UARK, WIS; P. C. Standley 444 MO, US; G. R. Vasey US; E. O. Wooton ARIZ, KSC, NMC, POM, RM, UC, US, 82 CAS, GH, ILL, KSC, MO, NMC, POM, RM, UC, US; Wooton & Standley US, 3223 MIN, RM, WIS. Grant: C. N. Barney NMC; O. B. Metcalfe US, 719 ARIZ, MIN, MO, NMC, RM, US, 721 ARIZ, GH, MIN, MO, US; J. G. Smith US. Harpinc: H. M. Hanson NMC. Hipatco: E. A. Mearns 2410 US; M.J. Murray CU, WIS. Luna: D. Griffiths 3333 US; H.T. Hen- son NMC; M. E. Jones POM; A.J. Mulford 1029 MO. OTERO: W.C. Alsdorf NMC; F. G. Plummer US. Quay: D. D. Suggs NMC. Sierra: Mrs. W. G. Beals 4 NMC; E. F. Holmes NMC; O. B. Metcalfe 1365 (in part) CAS, US, 1385 CAS, GH, MIN, MO, MT, POM. Socorro: C. L. Herrick 683 US;.G. R. Vasey GH, US. *NEW YORK (1936). QuEENS: J. Monachino 195 CU, TENN. 36 MADRONO [Vol. 13 OKLAHOMA (1926). CLeveLtaAnp: E. L. Little, Jr. 403 OKL. Jackson: M. Hopkins 874 OKL. McCtiain: M. Hopkins 33 OKL. OxiAHoma: U. T. Waterfall 1637 OKL, 1779 OKL, 2337 GH. *PENNSYLVANIA (1933). DELAWARE: L. C. Wheeler 5560 GH. PHILADEL- pHIA: T. O’Neill GH. TEXAS (1834). Indefinite locality: J. L. Berlandier 977 GH, 2407 GH, MO, US. ArcHer: L. H. Shinners 15868 SMU. Atascosa: E. J. Palmer 12920 GH, MO; H. B. Parks 20788 GH; L. H. Shinners 16917 SMU. Bastrop: H. H. Duval TEX; R. Mauermann 34 TEX. Bexar: C. R. Ball 914 US; R. D. Burr 317 TEX; Mrs. J. Clemens 190 MO, POM; dH. Eggert MO; G. Jermy MO; E. H. Wilkinson MO. BosguE: L. H. Shinners 15284 SMU. Brewster: V. L. Cory 26442 GH, 40344 GH; LRH, Shinners 8831 SMU; O. E. Sperry 7875 UC, US, T879 UG, US, T1272 UC,,US; B. H. Warnock 347 GH, 6667 SRSC, 7197 SRSC, 8024 SRSC. Brown: V. L. Cory 15331 GH, 15333 GH. CAMeEron: Mrs. A.M. Davis TEX; L. H. Shinners 17760 SMU. Cray: L. H. Shinners 12860 SMU. Coxe: V.L. Cory 37955 GH. Cororapo: B. Mack- ensen 208 MO. Dimmnit: V.L. Cory 29319 GH; S.S.Ivanoff 29291 GH; L.H. Shin- ners 17313 SMU, 17314 SMU. Epwarps: V.L. Cory 5245 GH, 38113 GH, 52493 RM, SMU, UC. Et Paso: B. Barlow UC; Mrs. J. Clemens POM; M. E. Jones POM; E. A. Mearns 1503 US; B. Shimek IA; G. R. Vasey US; B. H. Warnock 7277 SRSC, 8192 SMU, SRSC. FisHer: J. Brookes 41 TEX. Harris: G. L. Fisher 276 ILL, US. Hiparco: Ferris & Duncan 3067 CAS, MO; M. E. Jones MO; Lundell & Lundell 13337 SMU; L. H. Shinners 17153 SMU, 17169 SMU, 17897 SMU. Hupspetu: V. L. Cory 1488 GH; U. T. Waterfall 6276 MO. Jerr Davis: F. Barkley 14T798 TEX; V.L. Cory 9397 GH, 17639 GH; L. C. Hinckley TEX; B. H. Warnock 6940 SRSC, 7953 SRSC, 9228 SMU, SRSC; M.S. Young ILL, TEX, UC. Jim WE ts: R. Free- born 438 TEX. KAuFMAN: L. H. Shinners 15968 SMU. Kerr: A. A. Heller 1890 ARIZ, CAS, :CU,. GH, TA, TEL, ISC, MICH, MIN; MO: MSC2-RM..UG. cus: KieBERG: V. L. Cory 51319 SMU, TEX; L. H. Shinners 17021 SMU, 17068 SMU; J.F. Sinclair GH, TEX. La Sarre: L. H. Shinners 17280 SMU. Lavaca: G. L. Fisher 117 US. LivE Oak: L. H. Shinners 17000 SMU. Liriano: E. Whitehouse TEX. Lovinc: B.H. Warnock 10676 SMU. Lussock: E. L. Reed 1658 US. McLennan: L. D. Smith 315 TEX, 825 TEX. Mepina: V. L. Cory 11676 GH. Mipianp: V. L. Cory 40596 TEX, 40597 GH. Miram: L. H. Shinners 14729 SMU. Mitts: V. L. Cory 58207 SMU, 58208 SMU; L. H. Shinners 16837 SMU. MontacueE: L. H. Shinners 12832 SMU. Morris: V. L. Cory 56911 SMU. Nueces: V. L. Cory 20486 GH. PRresip10: W. W. Eggleston 17295 US; L. C. Hinckley ARIZ, GH, 2047 GH, TEX; T. H. Rogers SRSC; B. H. Warnock 94 TEX, US, 47830 SRSC; Watts & Finger SRSC. ReEEvEs: V. L. Cory 52239 SMU; Rose & Fitch 17896 US. Rerucio: E. R. Bogusch OC. San Patricio: F. B. Jones 377 SMU; Tharp & Brown 48199 TEX. TARRANT: A. Ruth 158 MICH, 967 (in part) CU, GH, ISC, MSC, SMU, US, WIS; E. White- house 16254 SMU. Tayvtor: W. L. Tolstead 7702 GH, SMU. TERRELL: W. B. Crockatt 50 NMC. Travis: B.C. Tharp CM, GH, MO, UC, 707 US, 1530 TEX, US, 1535 TEX, US, 1962 US; B. H. Warnock 66 TEX; M. S. Young TEX. VALVERDE: H. Eggert GH, MO. Warp: B. C. Tharp 3389 (in part) TEX. Wess: V. L. Cory 16872 GH; C.R. Orcutt 5720 MO; L. H. Shinners 17225 SMU. Wicuita: L. H. Shinners 15855 SMU. Wirparcer: E. Whitehouse 10945 (in part) SMU. Witracy: V.L. Cory 51467 GH, SMU; R. Runyon 2846 UC, US; L. H. Shinners 17113 SMU. 7. AMARANTHUS WATSONI Standl. Bull. Torrey Club 41:505. 1914. Amblogyne Torreyi A. Gray, Proc. Am. Acad. 5:167. 1861. (pro parte). Amaranthus Torreyi (A. Gray) Benth. ex. S. Wats. Bot. Calif. 2:42. 1880. (pro parte). Amaranthus Torreyi (A. Gray) Benth. ex S. Wats. var. suf- fruticosus Uline & Bray, Bot. Gaz. 19:272. 1894. Amaranthus Palmeri S. Wats. var. glomeratus Uline & Bray, loc. cit. 272. 1894. (pro parte). 1955] SAUER: AMARANTHUS 37 A. Watsoni KILOMETERS Fic. 9. Amaranthus Watsoni: distribution map. Atypical plants indicated by crosses. Amblogyne Torreyi was described from a very discordant group of specimens (syntypes: J. WM. Bigelow, Mex. Bound. Surv. 1190, mountains of the Cibola, a tributary of the Rio Grande, W. Texas, August, 1852, US!; C. C. Parry s.n., “Camp Green, New Mexico,” place unlocated; L. J. Xantus 100, Cape San Lucas, Baja California, 1859-60, ¢ 2 GH! 6 US! ). Gray cited another collection (H. Engelmann s.n., Lt. Bryan’s Expedition, South Fork of Platte, September, 1856, 2 GH! MO!) as rep- resenting a variety of the species; one sheet was inscribed A. Torrey angustifolia, but this name has not been published. Although Gray de- scribed the species as dioecious, the Bigelow and Parry specimens are not; Uline and Bray quite properly split them off as the types of a new mono- ecious species, Amaranthus Bigelovui, and a variety of the same, neither of which taxa concern us here. Only the Xantus collection belongs to the species now under consideration; the Engelmann collection belongs to the next species below. Uline and Bray effected a taxonomic separation of these two collections by splitting off a new variety under the misnomer suffruticosus, with the Xantus specimen as its type, thus leaving only the Engelmann collection in A. Torreyi proper. Unfortunately this procedure violated the rules, as Johnston has pointed out (1944, p. 155; 1948, p. 38 MADRONO [Vol. 13 193), since the element best fitting the original description was removed from the species. Johnston has proposed that the name A. Torreyi be re- instated for the monoecious species which has generally been called A. Bigelovit; one alternative would be to reinstate the name A. Torreyi for the species now being considered. Neither of these alternatives provides a happy solution, since for fifty years the name A. Torrey has been univer- sally applied to the next species treated, which includes the Engelmann specimen. I believe the best escape from this miasma is by discarding the name A. Torreyi completely as a nomen ambiguum, potentially a source of continued confusion and error. The earliest available name for the pres- ent species then becomes A. Watsoni (syntypes: Edward Palmer 312, on a sand spit near Guaymas, Sonora, October, 1887, ? US!, here designated as lectotype; also 6 2 in GH!; Edward Palmer 675, same area and year, 6 US!; paratypes: Edward Palmer 676, same area and year, 2 US!; Nelson & Goldman 7282, Santo Domingo to Matancito, Baja California, altitude 50 to 100 ft., November 14-15, 1905, 6 US!; Nelson & Goldman 7532, La Paz, Baja California, altitude 0 to 20 ft., February 17, 1906, 6 US!; and once again the well-worn Xantus 100. Plants moderately stout, low and bushy, with many ascending branches, or erect, to 1 m. tall; leaf-blades ovate to oblong, often deeply retuse, sometimes coriaceous; two flowering and fruiting seasons: plants which grow with winter rainy season mature in spring, those which grow with summer rains mature in fall; thyrses often extremely thick, flexible or fairly rigid, usually 10 to 20 cm. long, either all terminal on leafy branches or a few loosely arranged leafless branch thyrses present, the uppermost of these not subtended by leaves; bract usually 2% to 3% mm. long, with distinctive glandular pubescence on outer side, the midrib excurrent, moderately heavy in ¢, heavy in 2; ¢ flowers with 5 stamens, 5 tepals, the inner tepals 2'4 to 3 mm. long, obtuse or emarginate, the outer tepals 3 to 3%4 mm. long, acute, all tepals apiculate but dark midveins not ex- current; 2 flowers with 5 strongly recurved approximately equal tepals, all tepals usually about 214 mm. long, broadly spatulate and emarginate, with heavy, conspicuously branched, scarcely excurrent midveins; utricle about 114 mm. long, circumscissile, thin, somewhat rugose; style branches usually 2, sometimes 3; seeds 34 to 114 mm. in diameter, round or slightly obovate, lenticular, dark reddish brown. The species is at home in coastal dunes, on beaches and sandspits, and other places near the sea. It is the most abundant annual on some of the low desert islands of the Gulf of California, growing on guano-impreg- nated and weakly saline flats (Johnston, 1924, p. 1018; also his notes on specimens in CAS, cited below). Inland, the species stays close to water- courses and is common on the sandy floors of ravines and arroyos. The Colorado River and the associated irrigation works have apparently pro- vided the pathway by which this species, often mixed with A. Palmeri, has moved into southern California as a weed of irrigated fields and citrus groves. 1955] SAUER: AMARANTHUS 39 Like A. Palmeri, with which it is probably confused in some of the ethnobotanical literature, this species has served as an Indian food plant. Eighteenth century accounts of the Indians of central Baja California report that bushels of seed from wild b/edos were harvested in both spring and fall (Aschmann, 1953). The name, the region, and the kinds of sites mentioned all suggest that these plants were involved. MEXICO. BAJA CALIFORNIA (1860). Comonpu: T. S. Brandegee UC, US; T. Crocker CAS; Nelson & Goldman 7282 US; C. R. Orcutt 32 US; P.J. Rempel 135 ARIZ. EnsENADA: Carter, Alexander, & Kellogg 1930 UC, 1931 UC; I. M. Johnston 3527 CAS. La Paz: T. S. Brandegee UC; A. Carter 2717 UC, 2727 UC; I. M. John- ston 3032 CAS, 3225 CAS, GH, MO, UC, US; M. E. Jones 24452 MO; Nelson & Goldman 7532 US. Mutect: H. Aschmann WIS; H.S. Gentry 4065 MO, UC, 7831 UC, 7831A UC; I. M. Johnston 3652 CAS, 3743 CAS; E. Palmer 147 CAS, GH, US; I. L. Wiggins 7865 GH, MICH, UC, US. San Jose Det Caso: T. S. Brandegee 492 UC, US; L.J. Xantus 100 GH, US. Tonos Santos: B.J. Hammerly 292 CAS, US. SONORA (1887). CaBorca: E. Palmer 958 GH, US. Empatme: T. D. Mallery ARIZ. Guaymas: H. S. Gentry 4686 ARIZ, CAS, MICH, MO, UC, US; E. Palmer 311% ARIZ, GH, US, 312 GH, US, 675 US, 676 US; I. L. Wiggins 6356 MICH, POM, WS; 6357, US: UNITED STATES. ARIZONA (1912). Yuma: J.J. Thornber ARIZ. CALIFORNIA (1901). ImpertAr: L. Abrams 3993 GH, MO, POM; T. S. Bran- degee UC; M. E. Jones POM. Los AnceEteEs: L. C. Wheeler 963 POM, UC. RIvER- swe: T. S. Brandegee UC; J. D. Sauer 1667 WIS. 8. AMARANTHUS ARENICOLA I. M. Johnston, Jour. Arnold Arb. 29:193. 1948. Amblogyne Torreyi A. Gray, Proc. Am. Acad. 5:167. 1861 (pro parte). Amaranthus Torreyi (A. Gray) Benth. ex S. Wats. Bot. Calif. 2:42. 1880 (pro parte). By a comedy of errors, presented briefly in the discussion of the pre- ceding species, the name A. Torrey has been incorrectly but almost uni- versally applied to the present species. As a result this common and well- known species had no scientific name of its own until 1948, when Johnston named it A. arenicola (holotype: A. S. Hitchcock 428A, sandhills, Ham- ilton County, Kansas, 1895, 2 GH!; isotypes: ? KSC! NMC! RM! US! 6 MO!). Plants stout and erect, commonly 2 m. tall; leaf-blade usually oblong; flowering and fruiting entirely in summer and fall, mostly July through September; thyrses often thick, flexible, usually 10 to 20 cm. long, either all terminal on leafy branches or, if a few leafless branch thyrses present, these loosely arranged and each subtended by a leaf; bract usually 1% to 2% mm. long, the midrib scarcely excurrent, heavy in ¢, extremely heavy in 2; ¢ flowers with 5 stamens, 5 approximately equal tepals, the tepals usually about 3 mm. long, inner emarginate or obtuse, outer obtuse or acute, all apiculate but dark midveins not excurrent; ¢? flowers with 5 recurved spatulate tepals, each with conspicuous, usually branched mid- vein, the inner tepals 114 to 2 mm. long, emarginate or obtuse, the outer tepals 2 to 214 mm. long, obtuse, apiculate; utricle about 144 mm. long, 40 MADRONO [Vol. 13 circumscissile, thin, fairly smooth; style branches usually 3; seed 1 to 1% mm. in diameter, round, lenticular, dark reddish brown. Johnston’s name is well chosen; the plants seem to be very much at home in the sandhills. There and elsewhere the species is most abundant in such places as swales, dried ponds, lakeshores, river sandbars, edges of saline marshes, margins of hot springs. Most of the records are from sandy soil, but a few are from heavy silt or clay. The species is moderately suc- cessful as a weed. Almost half of the collections, including almost all those from isolated places outside the coherent range, were made in artificially disturbed habitats; fields, roadsides, railroad rights-of-way. UNITED STATES. COLORADO (1861). Indefinite locality: G. R. Kleeberger CAS; G. R. Vasey 509 CINC, GH, MO. Apams: J. A. Ewan 13968 COLO, NO; E. L. Johnston 275A MO; G. T. Robbins 853 UC. Baca: Weber & Anderson 5190 COLO, UC. BoutpEer: J. A. Ewan 12254 CAS, NO, UC; W. A. Weber COLO. CHAFFEE: E. Harper 167 WIS. CLEAR CREEK: C. C. Parry 323 GH, MO. DENVER: A. Eastwood COLO; W. Heustis COLO; M. E. Jones 579 POM, UTC; J. H. Smith 2038 US. Et Paso: R. Bacigalupi 904 GH, UC. Fremont: T. S. Brandegee 477 MO, UC, US, 846 MO, UC; G. Engelmann MO; G. Osterhout 3336 RM. Las ANIMAs: C. M. Rogers 5049 COLO, TEX, 5050 COLO, TEX. Lincotn: E. O. Wooton US. Locan: G. E. Osterhout 1141 RM. Morcan: G. Osterhout 7729 POM, RM; F. Rama- ley 16388 ARIZ, CAS, COLO, OKL, TEX, UC, UTC. Wasuincton: Ramaley & Ewan 16320 COLO. WELD: J. Ewan 12291 NO, WIS, 13988 COLO, NO. 14098 COLO, NO; E. L. Johnston 275 GH, US, 275B GH, MO, US, 277A US; E. J. Kraus WIS; F. Ramaley 12368 COLO, 12386 COLO, 12560 (in part) COLO; Ramaley & Ewan 16319A NO; L. F. Ward US. YumMA: Anonymous MIN, MO, WIS; G. E. Osterhout 4047 RM, WIS. *DISTRICT OF COLUMBIA (1896). E. S. Steele MO. *ILLINOIS (1898). Cook: L. M. Umbach WIS. CrawrForp: H. E. Ahles 4963 ID EI Gp *INDIANA (1896). LAKE: W. S. Moffatt 493 ILL; L. M. Umbach 388 TENN, 1419 WIS. *IOWA (1895). LEE: B. Shimek TEX. MuscaTIneE: F. Reppert IA. KANSAS (1876). Indefinite locality: A. S. Hitchcock 609 GH; L. G. Hoysradt US. Barton: A. S. Hitchcock KSC; P. P. Lorimer KSC. CHEYENNE: A. S. Hitchcock KSC. ComancHe: A.S. Hitchcock KSC. Dickinson: A. S. Hitchcock KSC. Epwarps: A. Finch KSC; A. S. Hitchcock KSC. EttswortuH: C. Weber 5 MO. FINNEY: R. Fritz 45 KSC; R. P. Murphy KSC; Rydberg & Imler 995 KSC; E. O. Wooton US. Forp: F.C. Gates 15974 KSC. Grant: A. S. Hitchcock KSC. Gray: A. S. Hitchcock KSC. Hamitton: A. S. Hitchcock 428A GH, KSC, MO, NMC, RM, US. HAskeEtt: A. S. Hitchcock KSC. HopcemMaAn: A. S. Hitchcock KSC. Kearny: A. S. Hitchcock KSC; Rydberg & Imler 928 MO. Kiowa: A. S. Hitchcock KSC; F. Rinkel GH, DUKE, TEX, US. Meape: A. S. Hitchcock KSC. Morton: Bloodhart & Poorman 9 KSC; A. S. Hitchcock KSC. Osporne: C. L. Shear 180 MO, 211 (in part) GH, NEB, US, 213 GH, NEB, US. PAwneer: A. S. Hitchcock KSC. Pratt: A. S. Hitchcock KSC. Reno: A. S. Hitchcock KSC. Rooxs: E. Bartholomew KSC, US. SEwarp: W. A. Kellerman KSC; Rydberg & Imler 846 COLO, KSC, UC. SHeErtpan: C. Weber 17 KSC. Starrorp: A. S. Hitchcock KSC; E. Maupin KSC. Stanton: A. S. Hitchcock KSC. Stevens: A. S. Hitchcock KSC. *MICHIGAN (1920). Lenawee: C. Billington US. Wayne: O. A. Farwell 5149 US. MISSOURI (1895). Jackson: B. F. Bush 518 CAS, MIN, MO; O. C. Durham 11232 ILL, MO. 1955] SAUER: AMARANTHUS ‘ ' e J ' t MILES i) e ~ 100 200 . ry 100 200 800 we) ae) a Sew ae meme wn Stet os ee es ae ee = ) METERS ; aos é eee Zee KILO \ e® x XY ae ‘ e e! e ue I ' x e° : ! e e r I e \ Safeco sere gece oA ee e > ! ) ! = area e o ee ° ! ye ax %2° r a % Oe) a SS = = — NA (| paneer campers brates a ape? ted ial - e e eo; @ q 1 = * x = \ .o ® r ! @ \ e e i} ‘ 1 e ' % : aa ' : 2 ! e e .*, ;o e®& © - i 1%e @ ° exe | ® -_-— e e oe --- ra i en ee e- PNG ig he) Oe ee ea 1 ! 4 ar an joao ---- x \ ! : | ' \ ! eens \ I Ns : o © ée ' ' ' ! ' ~ a ' ! oF i eo We oe e@ ~~ ' ] j \ ' at ! | I ' | ' ! “N J 7 cf ‘ ( ‘ Aq a ; b) & PA A Fic. 10. Amaranthus arenicola: distribution map. Atypical plants indicated by crosses. 41 42 MADRONO iVol. 13 NEBRASKA (1856). Indefinite locality: H. Engelmann GH, MO; F. V. Hay- den GH; G. R. Vasey US. Artuur: W. Kiener 15417 GH. Cuerry: E. Anderson SDC, WIS; J. M. Bates GH, MIN, NEB, 5223 GH, MIN, NEB, 5315 NEB, 5323 MIN, NEB, 5334 NEB; M. Dworak NEB; W. L. Tolstead 609 GH. FRANKLIN: J. Ewan 14797 COLO, NO; H. Hapeman CM, UC; J. D. Sauer 1683A WIS. GARDEN: C. H. Churchill NEB; H. Engelmann MO. Grant: Rydberg & Tulen 1664 CU, 1674 (in part) SDC. Hott: Engelman, Hodges, & Nielsen 3486 MIN. Kearney: H. Hape- man NO, POM, RM. Knox: F. Clements 2755 (in part) CU, GH, MIN, US. Lan- CASTER: J. L. Sheldon WVA. Lincotn: W. Kiener 17310 GH; J. D. Sauer 1691 WIS. PiaTTE: K. M. Wiegand 826 CU. SHERIDAN: F. Sandoz 111 NEB, 112 NEB, 113 NEB, 114 NEB. Tuomas: J.C. Blumer ISC; Pool & Folsom NEB; Pool & Williams NEB; Rydberg & Tulen 1370 NEB, SDC, 1513 NEB, US. Wesster: J. M. Bates NEB. *NEW JERSEY (1930). Burtincton: W. H. Witte RM. Campen: W. H. Witte ARIZ. GioucesTER: S. F. Blake 11260 GH, US. NEW MEXICO (1903). CuHAveEs: D. Griffiths 5702 US. OKLAHOMA (1853). Indefinite locality: G. W. Stevens 804 OKL. BECKHAM: B. Osborn 1348R OKL. BiatneE: G. W. Stevens 857% GH, OKL. Cappo: J. M. Bige- low US; M. Hopkins 2132 OKL. Cimarron: D. Demaree 13306 GH, MIN, OKL. ComMaAncHE: F. B. McMurry 529 OKL. Custer: L. Mericle 408 OKL, 720 OKL, 949 OKL, 1032 OKL; G. W. Stevens 916 GH, OKL. GArrietp: H. B. Gephardt 563 US. Greer: R. Bull 221 OKL; G. W. Stevens 1039 GH, ILL, MIN, MO, OKL, US. Harmon: M. Hopkins 1071 OKL, 1072 OKL. Kiowa: M. Hopkins 854 OKL; G.W. Stevens 1190 GH, MIN, OKL. Wasuita: C. T. Eskew 1498 OKL. Woops: G. W. Stevens 1702 GH, ILL, MIN, MO, OKL, US. *PENNSYLVANIA (1941). DELAWARE: L. C. Wheeler 5563 GH. SOUTH DAKOTA (1889). BENNETT: J. W. Moore 843 MIN. Lyman: T. A. Williams 143 CAS. PENNINGTON: W. H. Over 2384 US. WasHincTon: W. H. Over 2410 US. TEXAS (1859). Indefinite locality: V. Havard US; G. C. Neally GH, US; J. Reverchon MO. Atascosa: V. L. Cory 11677 GH, 19152 GH. Bexar: B. Mack- ensen 219 MO; H. B. Parks 27084 GH. Cameron: I. Shiller 915 TEX. CHILDRESS: Anonymous TEX. COLLINGSwortH: V.L.Cory 16127 GH. Cotte: V.L. Cory 35272 GH. Datiam: V. L. Cory 35418 GH. Dickens: V. L. Cory 16004 GH. FisHer: E. Whitehouse 16737 MIN, SMU. Hatt: J. Reverchon MO; E. Whitehouse 10743 SMU, UC. HarpemMan: C. R. Ball 965 US. Jerr Davis: V. L. Cory 40520 GH. KENEDY: V. L. Cory 17036 GH; R. Runyon 2847 US, 3582 US; L. H. Shinners 17803 SMU. 17090 SMU. Lovinc: B. H. Warnock 10693 SMU. Martin: V. L. Cory 40947 TEX. McLennan: L.D. Smith 1045 TEX. Mepina: E. Palmer 1131 GH, MO, US. Mitcu- ELL: R. W. Pohl 4328 ISC, SMU. Morris: V. L. Cory 56912 SMU. VAL VERDE: Parry & Bigelow 1195 US; L. H. Shinners 17347 SMU. Warp: B. C. Tharp 3389 (in part) TEX, US; Turner & Warnock 60 SMU, SRSC; B. H. Warnock 9029 SRSC, 9036 SMU, SRSC; Warnock & Parks 8823 SRSC. WHEELER: V. L. Cory 35311 GH. Wiparcer: EL. Eggert MO, WIS. Wink ter: L. C. Hinckley 4518 SRSC. *VIRGINIA (1939). DinwippiE: Fernald & Long 10633 CU, DUKE, GH, MO, US: WYOMING (1889). GosHEN: Brenckle & Stevens 39022 NDA. PiatTte: A. Nel- son 8581 CU, MO. *SwrEetwatTer: W. Cleburne NEB. 9. AMARANTHUS GREGGII S. Wats. Proc. Am. Acad. 12:274. 1877. Amaranthus Greggu S. Wats. var. Muelleri Uline & Bray, Bot. Gaz. 19:272. 1894. Amaranthus myrianthus Standl. Bull. Torrey Club 41:506. 1914. Amaranthus annectens Blake, Jour. Bot. 53:103. 1915. Watson described this species from a single collection, entirely of pis- tillate plants (J. Gregg s.n., frequent near the mouth of the Rio Grande, 1955] SAUER: AMARANTHUS 43 MILES 100 100 200 KILOMETERS Fic. 11. Amaranthus Greggii: distribution map. Atypical plants indicated by crosses. Tamaulipas, December 17, 1847, GH!, here designated as lectotype; also MO! ). Uline and Bray knew but one additional collection, also pistillate (F. Mueller s.n., near Vera Cruz, 1853), from which they described a new variety. They mentioned no concrete differences between the species proper and the variety and added the curious statement that the two are “inseparable.” Standley described A. myrianthus to include this variety but based it on a different type (holotype: Edward Palmer 266, sea level, vicinity of La Barra, 8 km. east of Tampico, Tamaulipas, February 1-8, 1910, 2 US!; isotypes: 2 GH! MO! US! WIS! ). The proper taxonomic disposition of this collection and one similar to it (Edward Palmer 511, same area and year) is not easy to determine. The living plants were badly injured by insects and had become half monstrous, with larvae instead of seeds in most of the utricles. These miserable plants appear to be mor- phologically intermediate between A. Greggiu and A. arenicola, and are conceivably hybrids between those species. Blake described his species from a single collection (holotype: A. Schott 360, seashore, Celestun, Yucatan, May 12, 1868, 2 BM, fragment US!; isotype: ¢ 2 GH!), which shows some quantitative differences from the type of A. Greggit. 44 MADRONO [Vol. 13 However, when all available specimens are assembled, there is no appar- ent discontinuity between the two forms. Plants stout, sprawling or erect, to 1 m. tall; leaf-blade strikingly thick and coriaceous with prominent nerves beneath, usually oblong but quite variable in shape; flowering and fruiting during summer and fall in nor- thern colonies, during all seasons in tropics; thyrses thick, flexible, 5 to 15 cm. long, either all terminal on leafy branches or, if a few leafless branch thyrses present, each subtended by a leaf; bract 11% to 2% mm. long, the midrib scarcely excurrent, heavy in ¢, very heavy in 2; ¢ flowers with 5 stamens, 5 approximately equal tepals, the tepals usually 2% to 3 mm. long, the inner emarginate or obtuse, outer obtuse or acute, all apiculate but dark midveins not excurrent; ? flowers with 5 recurved, approximately equal tepals which are usually 244 to 3 mm. long, the inner spatulate and emarginate, the outer obovate-oblong and obtuse, all with heavy, conspicuously branched, scarcely excurrent midveins; utricle about 3 mm. long, indehiscent, somewhat fleshy, fairly smooth; style branches usually 3; seed 1144 to 134 mm. long, obovoid, lenticular, dark brown. The dozen collections with habitat data are all from coastal sand dunes or from sea beaches. MEXICO. CAMPECHE (1933). Carmen: C. D. Mell 2015 US. TAMAULIPAS (1847). Matamoros: J. Gregg GH, MO; R. Runyon 475 US. Tampico: E. Palmer 266 GH, MO, US, WIS, 511 GH, MO, US. YUCATAN (1868). CeLEstun: A. Schott 360 GH, US. Dz1LAm GoNnzALEz: G. F. Gaumer 1243 GH, MO, US. Procreso: R. L. Crockett 69 US; Lundell & Lundell 8061 US; W. C. Steere 3112 MICH. UNITED STATES. TEXAS (1902). Gatveston: G. L. Fisher 102 MO, MT, US, 303 US, 605 GH, MT, RM, US, 612 GH, US; F. C. Gates 19217 KSC; L. H. Pammel ISC; J. Reverchon 2940 MIN, MO. WIS. Jerrerson: B.C. Tharp 3131 US. KieBerc: B. C. Tharp TEX. 10. Amaranthus Acanthochiton (Torr.) stat. nov. Acanthochiton Wrightu Torr. in Sitgr. Rep. Exp. 170. 1853. Non Amaranthus Wrightu 5: Wats. Proc.Am. Acad, 127275. 1877. The monotypic genus Acanthochiton was described from two early collections (syntypes: S. W. Woodhouse s.n., Sitgreaves expedition, Zuni Pueblo, September, 1851, 6 2 GH!, the 2 here designated as lectotype; C. Wright 1167, Rio Grande Valley, about 50 miles below El Paso, Sep- tember 6, 1849, 2 GH! US!). Since the epithet Wrighta cannot be re- tained when this taxon is assigned to Amaranthus, another name must be chosen. To maintain a continuity in names, Acanthochiton is herein given specific rank. Plants stout and erect, usually about 4 m. tall, with many ascending branches; leaf-blade narrowly lanceolate to linear, the nerves prominent beneath, the margin crispate; flowering and fruiting entirely in summer and fall, mainly August through November; thyrses rather thick and stiff, 1955] SAUER: AMARANTHUS 45 MILES 100 100 200 KILOMETERS Fic. 12. Amaranthus Acanthochiton: distribution map. usually less than 10 cm. long, entirely terminal on leafy branches; bracts dimorphic, those on ¢ plants with the ordinary lanceolate form common to most of the genus, 2 to 3 mm. long, with moderately heavy, excurrent midribs, those on 2 plants 5 mm. long or more, recurved, the excurrent midribs extremely heavy, the laminae accrescent, finally enfolding the flower, indurated, with prominent reticulate venation and crenate mar- gins; ¢ flowers with 5 stamens, 5 tepals, the inner tepals 24 to 314 mm. long, emarginate, the outer 3 to 4 mm. long, acute, all apiculate but dark midveins not excurrent; ? flowers usually with 5 tepals, the inner rudi- mentary, less than 1 mm. long, linear, the outer well-developed, sometimes 4 or 5 mm. long, broadly spatulate, with crenate margins and conspicuous branching venation; utricle about 2 mm. long, circumscissile, thin, some- what rugose; style branches usually 3; seed 1 to 114 mm. in diameter, obovate, lenticular, dark reddish brown. The species is at home on sand dunes, sandy riverbanks, and sandy places in general. The distribution may have been modified by Indian gathering of the plants for food. Cooked as greens, the plants have been an important food of such people as the Hopi since ancient times (Hough, 1898, p. 142). MEXICO. CHIHUAHUA (1852). Juarez: H. LeSueur 177 TEX, 278 GH, MO, SMU, TEX, 280 GH, TEX; C. G. Pringle 796 CM, GH, MIN, MO, MSC, UC, US, WIS. Ojtnaca: F. Shreve 9033 ARIZ, GH, MICH, UC. PRAxEpIs GUERRERO: G. Thurber 806 GH, 809 GH. UNITED STATES. ARIZONA (1896). GREENLEE: A. Davidson 1081 GH. Navajo: W. Hough 60 US; A. F. Whiting FLAG; M. Zuck UC, US. 46 MADRONO [Vol. 13 NEW MEXICO (1846). Dona Ana: F. R. Fosberg $3410 CAS, GH, POM, UC; G. R. Vasey US; E. O. Wooton ARIZ, COLO, NMC, OC, POM, RM, UC, US, 24 GH, MIN, MO, NMC, POM, RM, UC, US; Wooton & Standley 3156 IND, MIN, NMC, RM, US, WIS. Hipatco: E. L. Greene 272 GH, MO, POM. Luna: D. Grif- fiths 3330 US, 3333A US; M. E. Jones CAS, GH, MO, POM, RM, UC, US, UTC; A. I. Mulford 1030 ILL. Otero: F. S. Earle 416 US; B. Shimek IA. McKintey: S. W. Woodhouse GH. Rio Arripa: E. O. Wooton 2722 US. SAN Juan: P. C. Stand- ley 7887 US. Socorro: C. L. Herrick 840 US; A. Wislizenus 56 MO; E. O. Wooton NMC. Vatencia: E. F. Castetter 1312 RM; H. H. Rusby 365% CU. TEXAS (1849). Indefinite locality: C.C. Parry MO. Et Paso: V.L. Cory 30990 GH; D. B. Dunn 8791 CAS, DUKE, MT. NMC, TENN, WIS; V. Havard 49 GH; E. Stearns 436 US; B. C. Tharp MICH; B. H. Warnock 8244 SRSC, 8245 SMU, SRSC; E. Whitehouse TEX. Hupspetu: U. T. Waterfall 6575 GH, MO, SMU; C. Wright 1167 GH, US. Presipio: J. M. Bigelow 1186 US. Department of Botany, University of Wisconsin, Madison LITERATURE CITED ASCHMANN, H. 1953. The Indian population of central Baja California; Manuscript Thesis, Geography Library, Univ. Calif., Berkeley. CASTETTER, E. F., and W. H. BELL. 1942. Pima and Papago Indian agriculture; Inter- Americana 1:1-245. Univ. New Mexico Press. CasTETTER, E. F., and W. H. Bev. 1951. Yuman Indian agriculture. 274 pp. Univ. New Mexico Press. Gray, A. 1876. Notes on Acnida; Am. Nat. 10:487-489. Houcu, W. 1898. Environmental relations in Arizona; Am. Anthropologist 11:133- 155. Jounston, I. M. 1924. Botany of the expedition of the California Academy of Sciences to the Gulf of California in 1921; Proc. Calif. Acad. Sci. Ser. 4, 12:951-1218. 1944. Plants of Coahuila, eastern Chihuahua, and adjoining Zacatecas and Durango V.; Jour. Arnold Arb. 25:133-182. 1948. Noteworthy species from Mexico and adjacent United States II.; Jour. Arnold Arb. 29:193-197. Lanjouw, J., et al., editors. 1952. International code of botanical nomenclature; Reg- num Vegetabile 3. Lanjouw, J., and F. W. STaFrrev. 1954. Index Herbariorum, Part 1. The herbaria of the world; Regnum Vegetabile 2. Murray, M. J. 1940. The genetics of sex determination in the family Amaranthaceae; Genetics 25:409-431. Rose, J. N. 1895. Report on a collection of plants made in the states of Sonora and Colima, Mexico, by Dr. Edward Palmer; Contr. U.S. Nat. Herb. 1:293-366. STANDLEY, P. C. 1917. Amaranthaceae; N. Am. Flora 21(2) :95-169. TorrEY, J. 1853. Botany; Report on an expedition down the Zuni and Colorado Rivers by Captain L. Sitgreaves; 32nd Congress, 2d session, Senate, Exec. No. 59. Utine, E. B. and W. L. Bray. 1894. A preliminary synopsis of the North American species of Amaranthus; Bot. Gaz. 19:267—272, 313-320. 1895. Synopsis of North American Amaranthaceae; Bot. Gaz. 20:155—161. Watson, S. 1880. Botany; Geol. Surv. Calif. 2:1-552. 1889. Contributions to American Botany I. Upon a collection of plants made by Dr. Edward Palmer, in 1887, about Guaymas, Mexico; Proc. Am. Acad. 24:36-82. 1955] BIRDSEY: PSEUDOHOMALOMENA 47 THE STATUS OF PSEUDOHOMALOMENA PASTOENSIS M. R. BrrpsEY Pseudohomalomena pastoensis was described by A. D. Hawkes as a new genus and new species of the Araceae (Madrono 11:146-149. 1951). The author placed it in the subfamily Philodendroideae, tribe Philoden- dreae, subtribe Homalomeninae, with the statement that because of “‘its widely-spreading, almost flattened, large spathe” it ‘“‘is virtually unique in the subtribe.”’ A comparison of the type specimen (Espinosa 2866, Pasto, ““Ecuador,” UC 905798) with a photograph of the type specimen of Zantedeschia (Calla) aethiopica (No. 1081-1 in the Herbarium of the Linnean Society of London) indicates that the two are conspecific. Engler (1915) placed the genus Zantedeschia in the tribe Zantedeschieae of the subfamily Philodendroideae. In the description of Pseudohomalomena pastoensis certain statements are made concerning the habit of the plant that are not supported either by an examination of the specimen or by the collector’s notes. It is de- scribed as ‘‘terrestrial” and “apparently stemless’’; however, with the meagre material available one could as well assume that it was epiphytic and scandent. Likewise, the lamina is described as hastate with a caudate apex; actually the lamina of the type specimen is sagittate with a cus- pidate apex. The specimen appears to have the petiole and peduncle cut short; yet the measurements of the remaining parts were taken without qualification as to the probable missing parts. It is indicated in the Latin generic diag- nosis that the petiole is lightly vaginate, but this statement could not be derived from the cited specimen, as only a small part of the upper por- tion remains on the specimen and close observation reveals no vagination of any degree on this part. The presence of vagination is of importance inasmuch as the subtribe Homalomeninae, in which the author placed Pseudohomalomena, is defined by Engler and Krause (1912, p. 24) as having a vaginate petiole. In translating from the collector’s notes, an error has been made con- cerning the color of the spathe. “Color amarillo muy ckaro [claro], casi blanco”’ is translated as ‘“‘bright yellow, almost white.’’ More properly the phrase should be translated as color yellow, very light, almost white. This is in accordance with the spathe coloration that occurs in Zantedeschia aethiopica for the spathe often becomes cream color in age. The type locality, Pasto, was given in the article as being in Ecuador; although it is close to the border of that country, it is within Colombia. The collector’s notes clearly indicate that the plant was not native to Pasto for they read, ‘“‘cultivado 0 semiespontanea,” but this fact is not men- tioned in the article. There are numerous instances of localities where this 48 ; MADRONO [Vol. 13 plant has escaped from cultivation, for example: in Marin County, Cali- fornia (Howell, 1949, p. 97) and in Costa Rica (Standley, 1937, p. 146). It is, therefore, concluded from this study that Pseudohomalomena pastoensis A. D. Hawkes is a synonym of Zantedeschia aethiopica (L.) Spreng., the common White Calla Lily, a native of South Africa. Department of Botany, University of California, Berkeley LITERATURE CITED ENGLER, A. 1915. Araceae-Philodendroideae-Anubiadeae, Aglaonemateae, Dieffen- bachieae, Zantedeschieae, Typhonodoreae, Peltandreae. Das Pflanzenreich 64: (IV. 23. Dc) 1-78. Leipzig. ENGLER, A. and K. Krause. 1912. Araceae-Philodendroideae-Philodendreae. Homalo- meninae und Schismatoglottidinae. Das Pflanzenreich 55: (IV. 23, Da) 1-134. Leipzig. HowE Lt, J. T. 1949. Marin Flora. Berkeley. STANDLEY, P. C. 1937. Flora of Costa Rica. Publ. Field Mus. Nat. Hist. Bot. 391, Bot. Ser. 18:1-398. REVIEW How to Know the Grasses. By RicHarp W. Pout. 192 pp., 1954. Wm. C. Brown Company, Dubuque, Iowa. Spiral binding, $2.00; cloth binding, $2.75. This most recent addition to the ‘“Pictured-Key Nature Series” treats 293 of “... the commonest and most important species of American grasses — those that the beginner is most apt to meet, and those of importance in farming, gardening, weed control, range and pasture management. In addition to those keyed and illustrated, 91 others are mentioned in connection with closely related species, and their distin- guishing features are pointed out.” The book has a most helpful introductory section which points out (in a compact, illustrated key) the differences between the Juncaceae, Gramineae, and Cyperaceae and which also functions as an illustrated glossary for the more common terms the student must know to begin a study of the grasses. This introductory section is made more complete by the inclusion of a brief bibliography of useful books on grasses, several pages of directions for the collection and study of these plants, and a well illustrated key to the tribes of this family. The pictured keys to the species seem to be workable and well constructed. How- ever, the somewhat arbitrary selection of species necessitated by the geographic scope of the book may sometimes limit accurate field use of the book for identification to the species level. For example: Andropogon elliotti, found in southern Illinois, would key to A. vir- ginicus; and a student, not knowing the frequency of occurrence of A. elliottii and possibly not heeding the author’s note that “about fourteen other similar species or varieties occur in the southeastern states,’ might assume his work to be completed correctly. For the beginning student, however, such a mistake would probably be of no great consequence. The many grasses which are “common locally” throughout the United States could obviously not be included in this beginner’s book, and for this reason its greatest value, other than the text value of the introduction, will be in supervised field work and in class use to key out selected grasses. Within this scope, its content, size, and price should make it a popular member of the series and useful in a wide geographical area.—C. RitcHrE BELL, Department of Botany, University of Illinois, Urbana. INFORMATION FOR CONTRIBUTORS Manuscripts submitted for publication should not exceed 20 pages when printed, or contain more than 20 per cent illustrative material including tabular matter, unless the author agree to bear the additional cost. Subject to the approval of the Editorial Board, manuscripts may be published ahead of schedule as additional pages to the issue, provided the author assume the complete cost of publication. Shorter items, such as range extensions and other biological notes, will be published in condensed form with a suitable title under the general heading, ‘““Notes and News.” Institutional abbreviations in specimen citations should follow Lanjouw and Stafleu’s list (Index Herbariorum. Part 1. The Herbaria of the World. Utrecht. Second Edition, 1954). Articles may be submitted to any member of the Editorial Board. MADRONO A WEST AMERICAN JOURNAL OF BOTANY A quarterly journal devoted to the publication of botanical re- search, observation, and history. Back volumes may be obtained from the Secretary at the following rates: Volume I, 1916-1929 . . . . . $5.00 Volume IJ, 1930-1934. . . . . 5.00 Volume ITI, 1935-1936 . . . . 5.00 Volume IV, 1937-1938 . . . . 5.00 Volume V, 1939-1940. . . . . 5.00 Volume VI, 1941-1942 . . .. 5.00 Volume VII, 1943-1944 . . .. . 5.00 Volume VIII, 1945-1946 . . ... 5.00 Volume IX, 1947-1948 si! di Ae lies 6.00 Volume X,1949-1950. . . . . 7.00 Volume XI,1951-1952 . . .. 8.00 Volume XII, 1953-1954 . . .. . 8.00 Single numbers, Volumes land2 . ... . .50 Succeeding numbers, (except Volume XI, No.2). .. 1.00 Volume XI Nos2) 2) 92 1.50 The subscription price of MaproNo is $4.00 per year. If your institution does not now subscribe to MApRONo, we would be grateful if you would make the necessary request. Since there is no extra charge for institutional subscriptions, an individual may hold membership in the California Botanical Society on the basis of his institution’s subscription. Address all orders to: G. THomaAs Rossins, Corresponding Secretary Department of Botany . _ University of California, Berkeley 4, California , a, [sete ih | ————$——— ADRONO VOLUME 13, NUMBER 2 APRIL, 1955 Contents STUDIES IN NorTH AMERICAN VOLVOCALES. I. THE Genus Gonium, M. A. Pocock 49 FLORA OF THE CRESTED BUTTE QUADRANGLE, COLORADO, Jean H. Langenheim 64 Reviews: H. D. Harrington, Manual of the Plants of Colorado (G. Thomas Robbins) ; Lee Oras Overholts, The Polyporaceae of the United States, Alaska, and Canada (Lee Bonar) 79 NoTES AND NEWS 80 A WEST AMERICAN JOURNAL OF BOTANY f fae SHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY ey MADRONO A WEST AMERICAN JOURNAL OF BOTANY Entered as second-class matter at the post office at Berkeley, California, January 29, 1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price $4.00 per year. Published quarterly and issued from the office of Madrofio, Herbarium, Life Sciences Building, University of California, Berkeley 4, California. BOARD OF EDITORS HerserT L. Mason, University of California, Berkeley, Chairman EpGAR ANDERSON, Missouri Botanical Garden, St. Louis. LyMAN Benson, Pomona College, Claremont, California. HERBERT F. COPELAND, Sacramento College, Sacramento, California. Joun F. Davipson, University of Nebraska, Lincoln. Ivan M. Jounston, Arnold Arboretum, Jamaica Plain, Massachusetts. Mivprep E. Maruras, University of California, Los Angeles 24. MaArIon OwNBEY, State College of Washington, Pullman. Ira L. Wiccrns, Stanford University, Stanford, California. Secretary, Editorial Board — ANNETTA CARTER Department of Botany, University of California, Berkeley. Business Manager and Treasurer — RicHARD W. HoLm Natural History Museum, Stanford University, Stanford, California. CALIFORNIA BOTANICAL SOCIETY, INC. President: Lincoln Constance, Department of Botany, University of California, Berkeley, California. First Vice-president: John Thomas Howell, California Academy of Sciences, San Francisco, California. Second Vice-president: Mildred E. Mathias, Department of Botany, University of California, Los Angeles, California. Recording Secretary: Mary L. Bowerman, Department of Botany, University of California, Berkeley, California. Corresponding Secretary: G. Thomas Robbins, Department of Botany, University of California, Berkeley, California. Treasurer: Richard W. Holm, Natural History Museum, Stanford University, Stanford, California. Annual membership dues of the California Botanical Society are $4.00 for one person or $5.00 for two members of the same family. Either type of membership carries with it one subscription to Madrofio and all other privileges of the Society. Dues should be remitted to the Treasurer. General correspondence, changes of address, and applications for membership should be addressed to the Secretary. STUDIES IN NORTH AMERICAN VOLVOCALES. I. THE GENUS GONIUM! M. A. Pocock In the course of a field study of Volvox in North America, other mem- bers of the Volvocales were repeatedly encountered, one of the commonest among them being Gonium, which in spring and early summer is often found in rain-water pools similar to those in which Volvox may occur. No particular attention was paid to this genus during the journey across America, but later at the University of California time was diverted from the study of Volvox to this much simpler member of the Volvocaceae which was appearing in nearly all the soil cultures made in the laboratory as well as in many algal collections made in the neighborhood of Berkeley, California. Since most of the soil-samples used in these cultures were col- lected in Nebraska and California, where, too, the greater part of the field work was done, the present account is based on material from those two States. The work has continued in South Africa, using soil collected during the American trip some years previously, as well as some from Australia. Obviously the results obtained are by no means exhaustive, and further study would probably add considerably to our knowledge of the Gonium flora of North America. Of the five forms considered here, two have been found only in soil cultures and have not as yet been collected in the field, two others have been so collected and have also appeared in cultures; the fifth, G. soczale, was collected once in Berkeley. When possible the algae were isolated and raised in uni-algal cultures either by the soil-and-water method of Pring- sheim (1946, p. 13) or in culture solution made according to the recipe given by Juller (1937, p. 61); both single-colony and many-colony in- oculations were made, and all forms except G. sociale usually proved tolerant of culture conditions so that very rich growths of the various forms could be obtained. DESCRIPTION OF THE SPECIES Gonium octonarium sp. nov. (Lat., octonarium, composed of eight) (figs. 1-15). Species parva, saepe paene sine colore, cellulis octo, sex mar- ginalibus duabus minoribus centralibus, valde regulariter ordinatis: linea transversa duabus centralibus duabus marginalibus formata est, supra et infra duas centrales reliquis quattuor marginalibus positis; centralibus lateraliter paullum compressis, marginalibus omnibus a latere centralibus adiacenti plerumque paullum complanatis; in culturis recentibus cellulae chlorophyllo et pyrenoidibus carnet; in veteribus coenobia pigmentum 1 The work described here was done during the tenure of Senior Bursaries granted by the Council for Scientific and Industrial Research of South Africa. MapbroNno, Vol. 13, No. 2, pp. 49-80, April 29, 1955. BAY6 1955 50 MADRONO [Vol. 13 viride aliquando habent sed semper pallidissimum, pyrenoidibus tum prae- sentibus; in reproductione asexuali sex cellulae marginales sese dividunt plerumque prius quam centrales. Coenobium 46—54 pw X 38-45 pw; cellulis centralibus 9-12 » X 11-15 p, cellulis marginalibus 11-14 » X 13-16 p. A small species, often nearly colourless, composed of eight cells, six marginal and two smaller central cells, very regularly arranged; the two central cells and two marginal cells forming a transverse row, with the remaining four marginal cells above and below the two central cells; central cells slightly compressed laterally, all the marginal cells usually somewhat flattened on the side adjacent to the central cells; cells in young cultures lacking chlorophyll or pyrenoids, in older cultures the coenobia sometimes containing green pigment but always very pale, pyrenoids then present; in asexual reproduction the six marginal cells usually dividing before the central cells. O 9 O 10 Fics. 1-10. Gonium octonarium. Fic. 1. Nearly mature colony, colourless and with- out pyrenoids, showing attachment between cells, central nucleus with large nucle- olus, flagella, eyespots. Fic. 2. Colony from older culture showing pyrenoids sur- rounding central nucleated region. Fics. 3-5. Daughter colony formation: 3, ana- phase; 4, cleavage of third division; 5, cell division complete, embryo ready to invert. Fics. 6-10. Sexual reproduction: 6, sexual colony showing gamete formation; 7, con- jugation; 8, planozygote; 9, 10, zygospore before and after wall formation. Figs. 1, 2, 6 X 500; all other figures x 1000. 1955] POCOCK: GONIUM 51 HABITAT AND DISTRIBUTION. In rain-water pools, often those used as drinking pools by cattle grazing on pasture land. Hitherto known only from Nebraska and California; probably widespread but overlooked. Obtained only from soil cultures; not yet collected in the field. NEBRASKA: rain-water basin, Utica, Seward County; small pond in pasture land, Dinneen’s farm, near Exeter, Fillmore County. CALIFORNIA: pond in pas- ture land 4 miles south of Lemon Cove, Tulare County (type locality) ; vernal pools adjacent to the Santa Fe Grade near Los Banos, Merced County; inundated meadow south of Thornton (19 miles north of Stock- ton), San Joaquin County; shallow soil in holes in the rock in which rain or snow accumulates, El Moro Rock, Sequoia National Park, Tulare County. OBSERVATIONS. A very pretty and distinctive little species; when first observed in a soil culture from Utica, Nebraska, it was entirely destitute of chlorophyll and either colourless or faintly flesh-pink, with no sign of pyrenoids (fig. 1); later in the life of the culture it became very pale green. In cultures in which there is not much organic matter it soon de- velops chlorophyll and pyrenoids but it is typically much paler than other species of Gonium which are usually associated with it. When pyrenoids are present several develop in each cell in the cytoplasm around the central nucleus (figs. 2, 12, 13). The wall of the cell is closely adpressed to the protoplast, and even when treated with methylene blue can only be distinguished as a faint blue line with slight projections at the points of attachment between cells. The mode of attachment is interesting—normally the four cells of the median row are attached to one another at two points, whereas there is usually only one point of attachment between adjacent marginal cells and between the upper and lower two marginal cells and the central cells (figs. 1, 2, 12, 13). Exceptions to the rule are, however, not uncommon. The cells of the median row are closely approximated, the spaces between them and the upper and lower rows being comparatively large. The position of the eye- spot in the respective cells and the relation of eyespot, contractile vacuoles, and flagella (more than double body length) are shown in figure 1. The origin of the median row and its relation to the remaining four cells can easily be traced. After the second division two of the resultant cells widen tangentially, the two alternate cells radially. The third nuclear division, in which the axes of all four spindles are parallel, is immediately followed by cleavage which is radial in the two cells which had widened tangentially thus resulting in the two cells of the upper and lower marginal rows respectively, whereas in the two cells which had widened radially the cleavage is tangential and inclined to the plane of the marginal cells. As a result, the two inner cells, which are rather smaller than the two outer, lie in a plane below that in which the marginal cells all lie (figs. 3-5); the embryo is now a shallow saucer with the central cells at the bottom of the saucer (fig. 15). Inversion follows, which as usual in Gonium (cf., Pascher, 52 MADRONO [Vol. 13 1927, p. 414), consists of a reversal of the curvature of the young saucer- shaped embryo, thus bringing the two central cells slightly forward of the marginal cells to form the anterior pole of the colony as it progresses through the water. It is of interest to find that in daughter-colony for- mation these two central cells lag behind the others and may still be un- divided when all the marginal cells have either completed or nearly com- pleted embryo formation (fig. 14). Thus there is more than a suggestion of somatic differentiation. Throughout the process of cell-division and reorientation the embryo is attached to the parental flagella and the colony remains motile. Pascher states that the developing embryo becomes free from the flagella prior to inversion, but this is not normally the case in the forms studied by the writer; when it does occur it is probably an abnormality. Sexual reproduction was observed once in the early hours of the morn- ing (1 a.m. onwards) in a culture of soil from Utica. The culture was very vigorous with many young colonies in process of development, but it pre- sented a strikingly different appearance from those usually seen, as many small colonies showed groups of four cells in each constituent cell. These were arranged as if to form daughter colonies, but instead of escaping as such, the cells of each group proceeded to separate, escape from the paren- tal cell-membrane, and become gametes. In all the cases of conjugation observed, the gametes were unequal in size. The difference in size may be attributable to variation in size of the parent cell or perhaps to a difference in the number of gametes formed in each, although only four-celled groups were seen. In hanging drops in deep depression slides kept in a moist chamber, liberation and copulation of gametes continued for some hours, while colonies remained active for several days. The planozygotes were actively motile for some time, usually but not always settling down in less than an hour after completion of conjugation, then withdrawing their flagella and forming small rounded resting spores (4-5 » when first formed, enlarging to 6-8 » on completion of wall formation). At first the zygospore was almost colourless, but later became pale golden brown with a thin smooth wall. In soil cultures under favorable conditions, young colonies may appear within 24 hours after addition of water, or their EXPLANATION OF FIGURES Fics. 11-15. Gonium octonarium. Fic. 11. Nearly mature colony, colourless, without pyrenoids (Santa Fe Grade culture), stained with aceto-carmine to show nucleolus. X 1000. Fic. 12. Nearly mature colony from older culture (Utica culture), showing many small pyrenoids. * 1000. Fic. 13. Large green colony with many pyre- noids (Lemon Cove culture), showing mode of attachment between cells. x 960. Fic. 14. Mature colony (Santa Fe Grade culture), showing central cells undivided, marginal cells forming daughter colonies: three 4—celled, two preparing for last division and one with third cleavage nearly complete. x 1000. Fic. 15. Part of mature colony (Utica culture), showing two central and one marginal cell of central row undivided and two cells of upper row with daughter colonies ready to invert. x 1500. Fic. 16. Nearly mature colony of G. octonarium and young colony of G. multicoccum, stained with aceto-carmine (Lemon Cove culture). x 440. POCOCK: GONIUM Fics. 11-15. Gonium octonarium. 53 54 MADRONO [Vol. 13 appearance may be delayed to a later stage in the life of the culture; tem- perature appears to be the chief determining factor. The resting spores may retain their viability for years, in some cases the soil used in the cultures having been collected over four years previously. This species responds well to cultural conditions. Rich cultures were obtained from both single- and many-colony inoculations by using the soil-and-water technique, usually in Pyrex test tubes. Such cultures have been maintained for weeks, and in them development of pigment accom- panied by formation of pyrenoids is usually much more marked than in the original soil cultures. GONIUM PECTORALE Miller (1773). This cosmopolitan, typically 16- celled species is much the best known and is probably the commonest of all the colonial Volvocales. It is so frequently met that one is inclined to pass it by without further examination, but probably it includes many forms, some of which are certainly worthy of varietal rank; one such is described here. The typical form (var. pectorale) was also observed and appears to be widespread in North America. GONIUM PECTORALE Miiller var. pectorale (fig. 17). Colonies of this variety are characteristically bright green. In addition to the 16-celled colonies, 8- and even 4-celled individuals may occur, or according to Pascher (1927, p. 412, footnote), occasionally even colonies of between 16 and 32 cells. The coenobium takes the form of a nearly square plate, not quite flat but usually bent back slightly along the two diagonals. Mature cells are nearly spherical with a single large basal pyrenoid. Pascher gives 90 u as the limit of size, with the cells 10 u wide by 14 long, but the North American material may be considerably larger, well over 100 » in diameter with cells 18 » X 20 yu. In other respects it agrees reasonably well with the descriptions of European material. As the cell matures, the single pyrenoid increases in size almost filling the base of the massive bowl-shaped chloroplast so that the floor of the colourless nucleated region is pushed up, thus compressing the nucleus which consequently becomes spheroidal. Hence, though in polar view the colourless nucleated region is circular in outline, viewed from the side it appears elliptical (fig. 17). HABITAT AND DISTRIBUTION. Temporary fresh water pools, ponds in pasture land, lakes, etc. Probably widespread in North America. Collected in small lakes in Wisconsin; rain-water pools in Nebraska; in California collected near Walnut Grove, in Jackson Slough near Isleton (Sacramento County), and in the neighborhood of Stockton (San Joaquin County). Cultures of the latter provided most of the material used in the cytological investigation of this species (Cave and Pocock, 1951). Also in cultures of soil from Nebraska (Utica, Seward County) and California (Lemon Cove, Tulare County; Santa Fe Grade, Merced County). OBSERVATIONS. Daughter colony formation was never observed to occur until the colonies had reached a large size; there was no indication of cell 1955] POCOCK: GONIUM 55 differentiation within the colony as regards commencement of division— sometimes the cells of a colony began to divide more or less simultaneously, sometimes one or more of the central cells dividing before the marginal, but there was no definite order, and there appears to be a marked absence of synchronization between the cells in respect to division. Behaviour of different cultures is considered later. GONIUM PECTORALE var. praecox var. nov. (figs. 18, 19, 21). Coeno- bium adultum parvum, cellulis plerumque 16, ovatis, singulis unaquaque pyrenoidibus. Coenobium 50-64 » X 54-68 »;cellulis 10-12 » X 13-14 p. Mature colony comparatively small, usually rather pale green and slightly oblong in shape; cells somewhat elongated, ovoid in side view Fics. 17-20. Mature cells. 17, Gonium pectorale var. pectorale; (N, nucleus; NI, nucleolus; CV, contractile vacuole; P, pyrenoid) ; 18, 19, G. pectorale var. praecox; 19, showing reorientation of parts prior to division; 20, G. multicoccum. Fics. 21-25. Nearly mature colonies: 21, G. pectorale var. praecox; figs. 22-25, G. multicoccum; 22, showing slightly rhomboidal 16-celled colony; 23-25, 8-celled colonies showing various types of cell arrangement. All figs. ca. & 500. 56 MADRONO [Vol. 13 with apex usually wider than the base in which the single large pyrenoid lies, colourless apical region containing the spherical nucleus larger and more conspicuous than in var. pectorale (figs. 18, 21); in young colonies the central space often large so that the central cells appear to be pushed into the corners; bases of the two corner cells of the margin consequently slightly angular. HABITAT AND DISTRIBUTION. Fresh water ponds and pools. Obtained in a culture of soil from Lemon Cove, Tulare County, California (type lo- cality) associated with G. octonarium and the following species. Probably future study will show that it is not uncommon. OBSERVATIONS. Daughter colony formation may begin in quite small colonies whence the varietal name since precocious development is one of the chief distinguishing features of this variety. For example, in a colony measuring 50 » X 54 » all the marginal cells had already divided once or twice while the still undivided central cells measured 10 » X 13 », and even smaller colonies may have begun to divide. Division may start in either marginal or central cells or more or less simultaneously in all. No indication of somatic differentiation was observed. Nuclear division is preceded by a shift in the parts of the protoplast, the colourless apex moving to one side while the nucleus itself migrates toward the surface of the cell (fig. 19). This change in cell polarity takes place in the same direction in all the marginal cells, while a similar se- quence, not necessarily in the same direction, is seen in the four central cells. The regularity of this change in position of the constituents of the protolast is most striking. At the same time the pyrenoid begins to widen and as nuclear division takes place the pyrenoid also divides. Successive nuclear divisions are accompanied by division of the pyrenoids, and eventually each daughter cell receives a single small pyrenoid. Gonium multicoccum sp. nov. (figs. 20, 22-25, 26-30). Coenobium adultum magnum, cellulis 8, 16 aut 32, subglobosis, multis unaquaque pyrenoidibus. Coenobium (cellulis 16) 60-76 m X 62—78u; cellulis 13-18 » X 16=19 p. Adult coenobium large, of 8, 16, or 32 cells; cells more or less globose, with a number of pyrenoids of varying sizes almost surrounding the central spherical nucleus. Cells of the colony evenly spaced, separated by comparatively small spaces; no large central aperture. HABITAT AND DISTRIBUTION. Ponds, inundated meadow land, fresh- water basins, etc. UNITED STATES. Neprasxka: fresh-water basin, Utica, Seward County. CALIFORNIA: inundated meadow south of Thorn- ton (19 miles north of Stockton), San Joaquin County; pond 4 miles south of Lemon Cove, Tulare County (type locality). Also in cultures of soil from all three localities. AUSTRALIA. NEw SoutH Wa tes: Flooded meadow, Woodlawn Road near Lismore (soil culture). OBSERVATIONS. The cells of the colony are more closely apposed than in G. pectorale ; young colonies may show hardly any spaces between the 1955] POCOCK: GONIUM 57 cells. As development proceeds the cells separate slightly but evenly, and there is a marked absence of the large central aperture which char- acterizes the varieties of G. pectorale (figs. 26-29). Two of the central cells are normally joined across the comparatively small central space (figs. 22, 28). When first formed the cells of the young colony each contain a single basal pyrenoid, but while still quite small the pyrenoids begin to increase in number, first one and then another appearing until there may be 6 or 8, or in well grown colonies as many as 10 to 12 in each cell. For ex- ample, in acolony 29 » X 35 w with cells only 6 » in diameter, most of the cells already contained two pyrenoids; another still immature colony 65 » wide with cells 15 w in diameter had 5 or 6 pyrenoids in each cell. Apparently the secondary pyrenoids are formed de novo and not by di- vision of pre-existing pyrenoids; they are of varying sizes, the primary pyrenoid usually larger than those formed subsequently. In contrast to G. pectorale var. praecox, the colourless apical area is reduced to a mini- mum and is barely discernible in surface view of the colony (figs. 20, 22-24). In addition to the 16-celled colonies, 32-celled colonies may be fairly numerous in vigorous cultures, while on the other hand there is always a large proportion of 8-celled colonies. In colonies of the latter type the arrangement of the cells is strikingly different from that in G. octonarium. The commonest arrangement is three rows—two of three cells each with a central row of two cells—basically resulting from the arrangement of the cells at the 8-celled stage of division when there are four central cells in the form of a cross with four marginal cells in the angles of the arms of the cross (figs. 23, 29, 30). In other instances the rows are in the order of 3-3-2 cells (fig. 24), whereas in yet others as development proceeds the cells become further displaced and more or less rounded colonies result, with a single central cell surrounded by seven marginal cells (fig. 25). A close examination of the mode of attachment between cells may indicate how these various arrangements have arisen from the initial 8-celled pattern. It must be clearly understood that all these forms are due, not to fragmentation of an originally 16-celled colony but to the fact that the parent cell has divided three times only instead of four. The 32-celled colonies are, however, the most striking feature of this species. They are beautifully symmetrical, octagonal in outline but not quite isodiametric—two sides parallel to the longer diameter consist of four cells, whereas the remaining six sides are all 3-celled, the corner cell in each instance being common to two adjacent sides. The arrangement of the cells is 18 peripheral, 4 central and 10 intermediate (figs. 26, 27) When first observed in early summer of 1949 in a rich collection of algae from an inundated meadow south of Thornton, it was thought that the 32-celled plates might represent a species distinct from the 8- and 16-celled colonies with which they were associated. Single colony cultures were 58 MADRONO [Vol. 13 established using complete or nearly complete 32-celled individuals for inoculation, but in every case the resultant culture was composed of 8- and 16-celled colonies; later cultures raised from 16-celled inoculations sometimes produced 32-celled offspring. Obviously therefore the 32-, 16- and 8-celled colonies all belong to the same species, which, unlike G. pectorale, is characterized by the presence of many pyrenoids in each cell. When fully mature the constituent cells are as large as in G. pectorale var. pectorale, but since the intervening spaces are smaller the colonies are, as a rule, proportionately smaller. The dimensions given above are all for 16-celled coenobia; 32-celled individuals naturally reach a much larger size. The one shown in Figure 27, though far from mature, with cells still only 12 » X 13 » and containing only 3 or 4 pyrenoids, already measured 88 p X 96 p. In 1953 numerous cultures of soil from Lemon Cove, California, were made in South Africa at Grahamstown; several of these produced beauti- ful growths rich in 32-celled colonies, and it was possible to study the development and behaviour of the various types of colonies. Taking the 16-celled coenobium as the norm, it was interesting to find that on the whole the 8-celled form was the most abundant in young cultures, was the most stable and tended to reach maturity earliest, whereas 32-celled individuals made their appearance later in the life of the culture, developed more slowly than either the 8- or 16-celled colonies with which they were associated, full maturity culminating in the formation of daughter colonies being reached later than in either of the other forms; further, 32-celled colonies tend to fragment more readily. Sexual reproduction has not as yet been observed in this species. GONIUM SOCIALE (Dujardin) Warming (1876). This 4-celled species was collected in early summer by Dr. Lee Bonar in the fishpond in his garden at Berkeley. Unfortunately, little time could be spared for it and the few attempts made to get it established in culture failed, although it continued active for some weeks in the water in which it had been col- lected. Addition of culture solution to the original water did not notice- ably stimulate growth. The colonies were small and pale in colour and were apparently growing in water containing far less organic matter than in most of the soil cultures studied. Two-celled colonies sometimes occurred, but never colonies of EXPLANATION OF FIGURES Fics. 26-30. Gonium multicoccum. Fics. 26, 27. 32-celled colonies fixed in osmic acid vapour (Thornton Meadow culture) : 26, showing flagella and pyrenoids, « 430; 27, showing at a slightly lower focus the marginal cells with clear apical area, eye- spots, etc., * 490. Fic. 28. Nearly mature 16-celled colony with the central cells in focus to show absence of central aperture and attachment between the cells (Lemon Cove culture). * 960. Fic. 29. 8-celled colony from same culture stained with aceto- carmine. < 960. Fic. 30. Part of 16-celled colony showing undivided cells and de- veloping embryos in 2-, 4- and 8-celled stages. & 960. POCOCK: GONIUM Fics. 26-30. Gonium multicoccum. 59 60 MADRONO [Vol. 13 more than four cells. Colonies of this species were quite easily distinguish- able from the 4-celled colonies which may occur in any other species of Gonium. The species is apparently widespread but by no means common, and little is known about its life-cycle and behaviour. It has not appeared in any cultures of soil from North America nor has it yet been reported from South Africa. OBSERVATIONS ON CULTURES As the first rich cultures of Gonium pectorale made in the laboratory at Berkeley approached maturity, they were closely watched for dividing cells, but no matter what the hour of the day none could be found although many young colonies were always present. Observation was therefore started at night, and at once divisions were found; as night ad- vanced, the number of colonies showing stages in division increased. The first evening division apparently started between 7 and 8 o’clock and increased rapidly until about 9 p.m. when an optimum was reached and nearly all the larger colonies showed stages in division. This period of intense activity lasted for a time; then the number of dividing colonies gradually decreased until at about 10:30 p.m. no more divisions could be found. Observations were repeated on several succeeding nights with com- parable results except that there was a gradual shift in the time of suc- cessive stages to a slightly later hour. These observations led to the conclusion that in Gontum nuclear divi- sion, followed immediately by cell cleavage, was a nocturnal phenomenon. This conclusion, however, proved to be unwarranted for a few weeks later another culture was found to be dividing in the early afternoon, while in yet a third instance division took place in the morning. Quite recently, in a soil culture in which several forms of Gonium were flourishing, G. octo- narium was dividing actively at night while G. multicoccum showed hardly any stages in division but was found to be dividing freely in the early forenoon. There is therefore no stereotyped rule for the time at which division occurs in Gonium. All that can be said with certainty is that in any given population there appears to be an optimum time for nuclear division, but that this optimum may vary considerably in different populations or for different forms growing side by side in the same culture. It is difficult to estimate what factors may be operative; undoubtedly external factors play an important role in determining the behaviour of a culture, but obviously they are not alone responsible for variations in behaviour in this respect. Response to external conditions appears to be modified by some inherent factor or factors in any given strain and in any culture. Further, there appears to be a rhythm which is not strictly diurnal since there may be a progressive shift in the time of optimum division. Extended study of be- haviour in numerous cultures under varying conditions might do much to solve this as well as many other problems in the life cycle of Gonium and 1955] POCOCK: GONIUM 61 other green algae. Gonium would appear to be a particularly favorable subject for such a study. DISCUSSION In a cosmopolitan and moreover highly variable alga such as Gonium one hesitates to create new varieties or species; only when characteristic features, whether of structure or behaviour, prove constant and readily recognizable is such a course justifiable. The question then arises as to whether such variants constitute new species, or whether they should be regarded as varieties or merely forms of existing species. In the case of G. octonarium described here as new, from its first recognition there could be no doubt that it was a distinct, undescribed species. No other species of Gonium shows such constancy of number and arrangement of cells, combined with such individuality of cell-structure. But the other two taxa, both usually 16-celled with cell arrangement essentially similar to the cosmopolitan G. pectorale var. pectorale, could easily be passed over as that species. This is particularly true of the small form here named G. pectorale var. praecox in which there is a single basal pyrenoid. At first it was regarded as merely a form of G. pectorale var. pectorale, but there are distinct differences, not only in the shape of the cells with the con- spicuous apical region but also in their arrangement in the colony, in the precocious maturation and in the regularity of the shift in the polarity of the cell prior to division. These features are regarded as justifying varietal rank. In G. multicoccum, the differences are even more marked; here, although the shape of the cell is similar, its structure (with numerous pyrenoids scattered through the cytoplasm around the spherical nucleus and the small colourless apical region) and the compact arrangement of the cells in the colony without a large central space distinguish it from G. pectorale even when only the 16-celled form is considered. The numerous 8-celled, and still more the regular 32-celled colonies, further distinguish it; taken together these characters are regarded as sufficiently distinct to justify the view that it constitutes a separate species. Until comparatively recently the species of Gonium recognized as dis- tinct have all been either 4- or 16-celled, but in 1942 Prescott described a 32-celled species, G. discoideum |“‘constantesque 32 (raro 16) singulis cellulis” | from Louisiana. This species differs from G. multicoccum from California in the shape of the colony which is rounded in outline instead of octagonal, and in the form of the cells which are irregularly pyriform with two basal pyrenoids. The arrangement of the cells—4 central surrounded by two series of 10 and 18 respectively—is the same in both of these species. But in the California material (Thornton Meadow and Lemon Cove collections) 32-celled colonies were comparatively few and 16- and 8-celled colonies far more numerous. Furthermore, single colony cultures clearly demonstrated that the 32-celled colonies did not represent a species distinct from the 8- and 16-celled colonies with which they were associated. 62 MADRONO [Vol. 13 Crow (1927), in his account of abnormal forms of Gonium in Great Britain reports the occurrence of 8-, 4- and 2-celled colonies in G. pec- torale, and points out that such colonies may originate either by a reduc- tion in the number of cells formed by division of the parent cell or by fragmentation of normal 16-celled colonies. He adds: “the eight-celled form does not appear to be represented by a distinct species,” and dis- cusses possible reasons for this absence, concluding that a “‘central aper- ture” in the colony is necessary due to the way the organism swims. Hence, the discovery of an 8-celled species has special interest since it fills a gap in the series of types of colony structure within the genus. Crow’s ex- planation for the absence of such a species obviously becomes unnecessary. He also mentions the occasional occurrence of ring-shaped colonies formed by the dropping out of the central cells in normal 16-celled colonies. Simi- lar forms have been observed in the course of this work, but apparently the circular 8-celled colonies described above in G. multicoccum have not been noted in typical G. pectorale. Possibly they are peculiar to G. multi- coccum wherein other types of 8-celled colonies are also formed in un- usually large numbers. The number of pyrenoids in a cell may sometimes be of diagnostic value as a specific character, as in the case of G. multicoccum, but since in many algae the number varies during the life cycle, this feature must always be used with caution. In some algae the pyrenoids increase in number prior to cell division, in others they may disappear entirely. In Gonium there seems to be considerable variation in pyrenoid behaviour; in G. pectorale var. pectorale, pyrenoid behaviour has not been studied, but in the variety praecox division of the pyrenoid accompanying nuclear division seems to be the rule, whereas in G. multicoccum numerous pyrenoids are formed de novo during the maturation of the cell. The disappearance of the pyre- noids during cell division, a characteristic of many algae, does not seem to occur in Gonium. These statements are, however, made with some reserve and more work is needed to elucidate fully pyrenoid behaviour in the genus—possibly both methods of pyrenoid formation will be found to occur in any one form, one or the other predominating. A change in the polarity of the cell prior to division is a phenomenon which may be observed in various members of the Volvocaceae including Volvox (Pocock 1933, p. 587) ; it is fairly general in Gonium but nowhere has such regularity been observed as in G. pectorale var. praecox. Ex- ceptional cases may occur in which the regular “follow my leader”’ fashion fails and the nuclei of two successive cells come to lie on adjacent sides, but this is rare and usually the marginal cells and consequently the daughter colonies formed by them all face the same way, the central cells showing a similar mutual sequence. Later in development the embryo colonies tend to swing around until they lie in the plane of the parent colony. Possibly the change in polarity of the cell is directly connected with the retention of motility throughout daughter-colony formation. 1955] POCOCK: GONIUM 63 Since conflicting accounts have been given as to the way in which Gonium moves, it seems advisable to give a brief description of movement as observed during this work. As the colony is normally slightly convex, the central cells constitute the ‘anterior pole” of the coenobium; move- ment is two-fold: rotation on an axis perpendicular to the colony through the center and a forward progression along the line of this axis. The nor- mal position of the plate during movement is thus on edge, while at rest the tendency is for it to present a surface view. There is never any indication of “progression by a series of somersaults” as described by some workers, but the rotary movement may be slightly jerky. The beginning of somatic differentiation seen in Gontum octonarium is of particular interest since no such differentiation has been noted in any other species. Much still remains to be elucidated as to sexual reproduc- tion, in particular, whether the presence of plus and minus strains recorded by Schreiber (1925) in G. pectorale are general for all taxa and whether or not all are dioecious. Apparently sexual reproduction is seldom ob- served, but possibly this may be because the process is a nocturnal phe- nomenon rather than a rare one. The longevity of the resting spores is probably even more pronounced than has been shown in the course of this work. SUMMARY Gonium collected in the field has been supplemented by material ob- tained from cultures of soil collected mainly in Nebraska and California. Five distinct forms of Gonium have been studied of which two are de- scribed as new species—G. octonarium (8-celled) and G. multicoccum (8-, 16- and 32-celled)—-and one as a new variety—G. pectorale var. praecox,; the two remaining forms are G. pectorale var. pectorale (nor- mally 16-celled, as in var. praecox) and the 4-celled G. soczale. The new species and variety are described and observations on the occurrence, life cycle, behaviour and results of cultures of the various forms are recorded. The discussion deals with taxonomic relationships within the genus; the diagnostic value of pyrenoids and their number and behaviour; change in polarity within the cell prior to division; movement of the colony, and incipient somatic differentiation as seen in G. octonarium. ACKNOWLEDGMENTS My grateful thanks are due to the Faculty of the Department of Botany, University of California, Berkeley, where the work was begun, and especially to the Chairman of the Department, Dr. Lee Bonar, who also supplied me with the material of Gonium sociale ; to Professor Twy- man of the Botany Department of Rhodes University where it was con- tinued; to Dr. Johannes Proskauer, Dr. Isabella Abbott, and Dr. Mary L. Bowerman, all of the University of California; and to Dr. Walter Kiener, Biologist of the Nebraska Game, Forestation and Parks Commission, for material and soil samples; to my brother, Professor L. G. Pocock of Can- 64 MADRONO [Vol. 13 terbury College, University of New Zealand, and to Mr. S. Whiteley of Rhodes University for the Latin descriptions; and finally to Dr. G. F. Papenfuss and Dr. Marion S. Cave for so kindly reading the manuscript, seeing it through the press, and for much helpful advice. Grahamstown, South Africa LITERATURE CITED Cave, Marion S. and Mary A. Pocock. 1951. Karyological studies in the Volvo- caceae. Am. Jour. Bot. 38:800-811. Crow, W. B. 1927. Abnormal forms of Gonium. Ann. Mag. Nat. Hist., Ser. 9, 19: 593-601. FritscuH, F. E. 1935. The structure and reproduction of the algae. Vol. 1. Cambridge. 791 pp. JuLter, E. 1937. Der Generations- und Phasenwechsel bei Stigeoclonium subspino- sum. Archiv Protistenk. 89:55—93. MUtter, O. F. 1773. Vermium terrestrium et fluviatilium, seu animalium infusiorum, helminthicorum et testaceorum, ... Vol. 1, pt. 1 [32]-+ 135 pp. Copenhagen and Leipzig: Heineck & Faber. Pascuer, A. 1927. Die Siisswasserflora Deutschlands, Osterreichs und der Schweiz, Vol. 4, Jena. 506 pp. Pocock, Mary A. 1933. Volvox in South Africa. Ann. So. Afr. Mus. 16:523-646. Prescott, G. W. 1942. The algae of Louisiana, with descriptions of some new forms and notes on distribution. Trans. Am. Micr. Soc. 61:109-119. PRINGSHEIM, E. G. 1946. Pure cultures of algae, their preparation and maintenance. Cambridge. 119 pp. SCHREIBER, E. 1925. Zur Kenntnis der Physiologie u. Sexualitat hoherer Volvocales. Zeitschr. Bot. 17:336-376. SmiTH, G. M. 1950. The fresh-water algae of the United States. 2d ed. New York. 719 pp. WarMInG, E. 1876-1877. Om en fircellet Gonium (Dujardin’s Tetramonas socialis ?). Bot. Tidsskr. Raekke 3, 1:69-83, pl. 1. FLORA OF THE CRESTED BUTTE QUADRANGLE, COLORADO JEAN H. LANGENHEIM Although the flora of the eastern slope of the Colorado Rockies has been studied in considerable detail, the western slope has received little attention. Several general surveys which include incomplete floral lists have been published, the most complete being Brandegee’s (1876), which was intended to be a supplement for southwestern Colorado to Porter and Coulter’s “Synopsis of the Flora of Colorado” (1874). Charles F. Baker also published a very incomplete list for the La Plata Mountains in 1898. Names of plants are mentioned in several vegetational surveys of the state (Robbins, 1910; Cary, 1911), and Schmoll (1935) discussed the vegeta- tion of the Chimney Rock Area, Archuleta County. However, there are no truly definitive lists for local areas other than that by Graham (1937) for the Colorado portion of the Uinta Basin. 1955] LANGENHEIM: CRESTED BUTTE 65 The following floristic list was prepared in conjunction with a detailed ecological study in the Crested Butte Quadrangle, and represents collec- tions made to document that study. It is prepared for publication previous to the ecological study at the request of several workers who will find it useful to their research in this area. Voucher specimens are in the herbaria at the University of Colorado and the University of Minnesota. Since the rarer species were deemed to be more useful to workers at the University of Colorado, in many instances the only collections of these are at that institution. Large but incomplete collections are in the herbaria at the Colorado A. & M. College, State University of Iowa, and the University of California at Berkeley. I wish to acknowledge the assistance of W. A. Weber, University of Colorado, in verifying or determining the collection in general. The majority of the Compositae, Gramineae, and Cyperaceae were submitted to H. D. Harrington, Colorado A. & M. College. I am also grateful to the following people for assistance in their special groups: C. R. Ball, Salix; Lyman Benson, Ranunculus; Bernard Boivin, Thalic- trum, Lincoln Constance, Umbelliferae and Hydrophyllaceae; G. J. Goodman, Eriogonum; F. J. Hermann, Juncus; C. L. Hitchcock, Draba and Lathyrus; D. D. Keck, Penstemon; G. B. Ownbey, Corydalis; C. L. Porter, Astragalus and Oxytropis ; C. O. Rosendahl, Saxifragaceae ; E. T. Wherry, Polemoniaceae ; Robert E. Woodson, Asclepias. The Crested Butte Quadrangle is located in west-central Colorado on the southwest flank of the Elk Range. It is about 35 miles north of Gun- nison and 25 miles southwest of Aspen in Gunnison County (fig. 1). The altitude ranges from 8,000 to 13,500 feet. Many types of bedrock occur here although ninety percent is sedimentary rock; the remainder consists primarily of igneous intrusives. The topography is rugged, being in an early mature stage of the erosion cycle modified by glaciation. Climatic conditions are diverse. The only weather station in the area is located at Crested Butte, 8,867 feet, where an average annual rainfall of 28 inches is reported (U. S. Dept. of Commerce, 1952), but weather bureau esti- mates are as high as 50 inches in portions of the area. There are two maxima, of precipitation: from July to September, and during January. The heavy accumulation of winter snow supplies a persistent source of moisture throughout the summer at high elevations. At Crested Butte the mean temperature for July is 56.8° F. and for January, 13.6° F.; the absolute maximum temperature on record is 91° F. and the absolute minimum, — 42° F. (U.S. Dept. of Agric. Yearbook, 1941). Although the temperature extremes at higher elevations are unknown, it is significant that freezing temperatures occur at irregular intervals throughout the growing season. The vegetation of this area has been disturbed little by the influence of man since the establishment of the Gunnison National Forest in 1909. However, despite a controlled program of grazing which has prevented widespread erosion, the composition of the grassland and meadow com- 66 MADRONO [Vol. 13 Fic. 1. Index map of Colorado showing location of Crested Butte Quadrangle. munities has been altered somewhat. A fire history is evident as having occurred between 1880 and 1900, thus coinciding with the peak of mining activity. From this brief discussion, it is evident that the area is one of highly diverse environmental conditions which could be expected to select a rich flora. The percentages of species representing different floristic elements, as recognized by W. A. Weber (personal communication), are shown in Figure 2. Although the area of the Crested Butte Quadrangle is located on the border of the Colorado Plateau, it is still a part of the Rocky Mountain system. Thus it is not surprising to find that a majority of the species at the elevation of the study area have Rocky Mountain affinities. Further analysis of this Rocky Mountain element shows that the highest percentage of species is either widespread throughout the Rockies or oc- curs primarily within the Central Rockies; few have Northern Rocky affinities. Species with circumboreal affinities or with distributions wide- spread over western North America are also common. There is a relatively small representation of species with Great Basin-Colorado Plateau affini- ties, but observations at lower elevations in the adjacent Gunnison River Valley indicate, as one would expect, a higher representation of species with these relationships. The flora is organizable into five vegetational zones: sagebrush, aspen, spruce-fir, upland herb, and alpine. The sagebrush zone characterizes the 1955 | LANGENHEIM: CRESTED BUTTE 67 50 ROCKY MTS. 40 30 20 WIDESPREAD WESTERN NO. AMERICA WIDESPREAD ROCKY MTS. CENTRAL ROCKY MTS. CIRCUMBOREAL PERCENTAGE BOREAL AMERICA 10 GREAT BASIN-COLORADO PLATEAUS CENTRAL & SOUTHERN ROCKY MTS. WIDESPREAD NO. AMERICA SOUTHERN ROCKY MTS. CENTRAL & NORTHERN ROCKY MTS. NORTHERN ROCKY MTS. ADVENTIVE SOUTHWESTERN U.S, A B Fic. 2. Floristic relationships of species in the Crested Butte area: A. Percentages of species representing floristic elements recognized by W. A. Weber. B. Analysis of the species in the Rocky Mountain element (see A) into component elements. 8,500-9,500 altitudinal range in the quadrangle and continues downward to approximately 7,500 feet in the adjacent Gunnison River Valley. It seldom occurs above 10,000 feet, but has been observed in patches near timberline. Artemisia tridentata is the dominant; other common shrubs are Chrysothamnus viscidiflorus and C. parryi. Festuca thurberi is the most prominent grass, and characteristic forbs are Achillea lanulosa, Erio- gonum neglectum, Vicia americana, Arenaria congesta, Erigeron specio- sus, Lathyrus leucanthus, and Potentilla pulcherrima. The aspen zone ranges from 8,500 to approximately 11,200 feet. Be- tween 9,500 and 10,500 feet it forms a distinct altitudinal belt, occurring as a continuous open forest below the spruce-fir zone or as groves scattered through the Festuca thurberi grassland. Populus tremuloides is the domi- nant. The most common forbs in the mature undergrowth are Thalictrum fendlert, Ligusticum porteri, Vicia americana, and Erigeron elatior. Bro- mus ciliatus is the most frequent grass. The Engelmann spruce-subalpine fir zone is the most extensive one in the Crested Butte area, occurring from 8,500 feet along streamsites to 12,500 feet as patches of KrummAolz. It is altitudinally well defined, how- ever, only from approximately 10,500 to 11,500 feet. Picea engelmannu and Abies lasiocarpa are the only tree dominants. A tall shrub layer is composed primarily of Rzbes cereum and R. wolf. Vaccinium myrtillus and V. caespitosum make up a low shrub layer which dominates the forest 68 MADRONO [Vol. 13 floor. Pedicularis racemosa, Arnica cordifolia, Fragaria ovalis, and Pole- monium delicatum are the most frequent herbs. Lodgepole pine, aspen, ~ and Festuca thurberi with characteristic associates replace burned spruce- fir forest. Replacement has been most extensive by the grassland in this area. The upland herb zone is most prevalent from approximately 11,500 to 12,500 feet, but also is characteristic of non-forested areas as low as 10,500 feet. This zone consists of a luxuriant assemblage of grasses, sedges, and showy forbs. The most common forbs are Senecio crassulus, Ligusti- cum porteri, Lupinus parviflorus, Delphinium barbeyi, Polygonum bistor- toides, and Helianthella quinquenervis. Carex ebenea and C. chalceole pis are the most common sedges, with Phleum alpinum, Poa alpina, P. arctica, and Trisetum spicatum the most important grasses. The alpine zone is best developed between 12,500 and 13,500 feet, although alpine assemblages interfinger with the upland herb zone as low as 12,000 feet. A well-defined population pattern could not be discerned in this altitudinal range. Characteristic groups of species occur on fell fields, boulder fields, on talus, and along streamsides and snowbanks, but dominance varies widely from stand to stand. Arétes on which fell fields predominate provide most of the area available to alpine species here. This habitat is relatively barren with only scattered herbaceous growth. Hy- menoxys grandiflora, Artemisia scopulorum, Oxytropis deflexa, Erigeron pinnatisectus, Polemonium viscosum, and Silene acaulis are the most fre- quent forbs. Patches of sedges with some included grasses occur locally. Kobresia bellardu, Carex elynoides, and C. hepburnu comprise the promi- nent sedges. The common grasses are Poa alpina, Trisetum spicatum, Fes- tuca ovina var. brachyphylla, and Agropyron trachycaulum. An extensive bunch grassland dominated by Festuca thurbert occurs within the matrix of all zones from 8,500 to 12,500 feet, being most preva- lent on deep soils on south-facing slopes. Bromus ciliatus and Agropyron trachycaulum are other characteristic grasses. The most frequent forbs are Achillea lanulosa, Potentilla pulcherrima, Linum lewisu, Erigeron speciosus, Thalictrum fendleri, Lathyrus leucanthus, and Vicia americana. In considering the floristic relationships of the communities in the area, it was arbitrarily decided to include only those species occurring with a frequency of 50 per cent or more since only these contribute significantly to the definition of homogeneity of the vegetational pattern. The floristic relationships of these species within the vegetational patterns are shown in Figure 3. Species with less frequency contribute only to minor diversi- ties within the patterns. Boreal, and Central and Northern Rocky Mountain elements are rep- resented in all of the communities, although the latter plays only a minor role in each community. The boreal element occurs most prominently in the aspen community, where it plays the dominant role. Also species with distributions widespread in western North America contribute to all of the vegetational patterns, but they play the dominant role in the sage- 1955] LANGENHEIM: CRESTED BUTTE 69 brush, fescue, and spruce-fir communities, being especially noteworthy in the sagebrush community where 63 per cent display this relationship. The Central Rocky Mountain element is represented in all but the fescue and sagebrush communities; it is the most conspicuous element in the upland herb community and takes second place in the alpine zone. The Central and Southern Rocky Mountain element is present in all communi- ties except the alpine, and it places second in the upland herb, fescue, and aspen communities. Species widespread throughout the Rocky Mountains are represented in all but the upland herb community, but assume no out- standing role in any community. Four other elements are represented only in a limited number of com- munities. The circumboreal element is restricted to the alpine, upland herb, and aspen communities, but dominates the alpine zone. Species with Great Basin-Colorado Plateau affinities are represented only in the sage- brush and alpine communities which are at elevational extremes. A similar situation in regard to the presence of desert elements in the alpine zone of the Sierra Nevada was noted by Went (1953). Species with Northern Rocky Mountain affinities and those widespread over North America play only insignificant roles in a few communities. It should be pointed out that, despite the evident relationships of the flora (fig. 2) and the floristic relationships of those species which are at least 50 per cent frequent in the communities (fig. 3), the physiognomic expression of the communities is more similar to that in the adjacent mountainous areas in the Great Basin and Colorado Plateau than to that on the eastern slope of the Rockies. Also the zonal sequence appears to be in response to a gradient toward more extreme xeric conditions than in many other localities on the western slope of the Rockies and in adjacent areas in eastern Utah. These conclusions are borne out by the following features. A distinguishable assemblage of grasses and forbs, which occurs in the Crested Butte area between the alpine and spruce-fir zones, has not been recorded on the eastern slope of the Rockies. This upland herb zone does occur, however, in the Uinta Mountains (Graham, 1937) and the Wasatch Mountains (Ellison, 1954). The spruce-fir zone on both the eastern slope and in mountainous areas in the Great Basin usually occu- pies about a 2,000-foot altitudinal range, but here, except on north ex- posures, it is restricted to about 1,000 feet, and after fire is replaced more commonly by a grassland than by lodgepole pine or aspen forests. A forest dominated by Pinus ponderosa var. scopulorum and Pseudotsuga taxifolia, which is general on the eastern slope below the spruce-fir zone, is represented in this area only by scattered or isolated trees along canyon walls. This pine-Douglas fir zone is also absent over much of the western slope and adjacent areas in eastern Utah. A zone dominated by Pinus ponderosa var. scopulorum, but usually without Pseudotsuga taxifolia as an associate, does occur, however, in many places in the Great Basin. In place of the pine-Douglas fir zone, below the spruce-fir zone, there usually occurs an altitudinally-defined zone of aspen. Although aspens are abun- 70 MADRONO [Vol. 13 e e e e z 5 ” ” wn ” oc %) - - = Ee ” e = = = = = LS < =m < 7) > > > > = uw = re St ge lei eyes) Recents af _ oO = oc a= ea rE) Ao ro] Sie) ro) RT| Ww Ww re) a Eee °o a a ay =a r=) a ro) =° = _ li ro) ° o roo) ro) oO ao z => So ro) ” ” z z Oo a | | 4 aay Pe 5) ae | & Hal Ls =.) z Z F a 4 a eran a uJ o 2 o£ = Fe qi ASPEN é i — | foo Fo] mz lJ eee F g Ec FESCUE al 50 40 30 SAGEBRUSH 20) ae 0 lu = Fic. 3. Floristic relationships of species which are at least 50 per cent frequent within the principal communities in the Crested Butte area. dant on the eastern slope, they usually do not form a well-defined belt such as has been reported for western Colorado and central Utah (Baker, 1925), the LaSal Mountains (Tanner and Hayward, 1934), the Uinta Mountains (Graham, 1937), the Wasatch Mountains (Lull and Ellison, 1950), and many other areas in the Great Basin. The interzonal grassland community dominated by Festuca thurberi also has been reported else- where only from the Abajo Mountains (Ellison, unpublished ms.). Gener- ally, over many areas on the western slope of the Rockies and in the Great Basin, the next zone below the aspen belt is one dominated by oaks, fol- lowed by one dominated by pinyon and junipers. In the Crested Butte area these zones are absent and in their place is one dominated by Artemz- sta tridentata. Discussions by numerous authors concerning the zonal sequences and the presence or absence of ponderosa pine, oak, pinyon- juniper, and high-altitude sagebrush zones are correlated by Langenheim 1955] LANGENHEIM: CRESTED BUTTE 71 (1953). The relatively unique zonal sequence in the area of the Crested Butte Quadrangle may result partially from a steep precipitation gradient between 8,000 and 10,000 feet and partially from the fact that 66 per cent of the area is southwest-facing. DOCUMENTARY COLLECTIONS Although specimens were collected throughout the area of the quad- rangle, intensive collecting was done over the 100 square miles where the vegetation was mapped. Because field investigations of this type are de- pendent upon sampling procedures, it is improbable that the list is com- plete. Therefore species (marked with an asterisk) from nearby similar habitats outside the area are included. Eventually it is expected that they will be found within the area of the quadrangle. The list includes 579 species, 245 genera, and 64 families. Numbers following specific epithets are the collection numbers of the author. ACERACEAE Acer glabrum Torrey, 68, 270 *A. negundo L., 1430 ANACARDIACEAE *Rhus trilobata Nutt., 1328 APOCYNACEAE Apocynum androsaemifolium L., 500 ASCLEPIADACEAE Asclepias hallii Gray, 476, 788 A. speciosa Torrey, 1376 *A. subverticillata (Gray) Vail, 1441 BERBERIDACEAE Berberis repens Lindl., 184-48, 19 BETULACEAE Alnus tenuifolia Nutt., 135-48, 1270 Betula glandulosa Michx., 317-48, 27 *B. occidentalis Hook., 135-48, 1379 BORAGINACEAE *Cynoglossum officinale L., 1377 Cryptantha bakeri (Greene) Payson, 1209 Eritrichium elongatum Wight, 444, 1014, W274 2109. Hackelia floribunda (Lehm.) Johnst., 107, 434, 1420, 3944 Lappula redowskii (Hornem.) Greene, 114, 1263 Mertensia ciliata (James) G. Don, 105-48 M. franciscana Heller, 3957 M. fusiformis Greene, 10, 11, 1222 M. viridis Nels. var. cana (Rydb.) Wil- liams, 102-48, 4, 276, 341, 362, 396, 808, 1063, 1113, 1300 Plagiobothrys scopulorum (Greene) Johnst., 1141 CALLITRICHACEAE Callitriche palustris L., 444-48 CAMPANULACEAE Campanula parryi Gray, 328, 376, 593, 3866 C. rotundifolia L., 366, 593, 2145 C. uniflora L., 439 CAPPARIDACEAE Cleome serrulata Pursh, 2016 CAPRIFOLIACEAE Linnaea borealis L., 142, 481, 3914 Lonicera involucrata Banks ex Sprengel, 182-48, 38 Sambucus racemosa L., 180-48, 31, 3904 Symphoricarpos occidentalis Hook., 76 S. tetonensis A. Nels., 121, 1325 S. vaccinioides Rydb., 780, 1205, 1314, 1535, 3938 CARYOPHYLLACEAE Arenaria congesta Nutt., 98-48, 358, 795 A. macrantha (Rydb.) Nels., 3968 A. lateriflora L., 126 A. obtusiloba (Rydb.) Fernald, 427, 2105 A. sajanensis Willd., 70, 905, 1361, 2106 A. verna L., 455 Cerastium arvense L., 1225, 1521 C. beeringianum Cham. & Schlecht., 560-48, 52, 814 Lychnis apetala L., 901, 1461 L. kingii Wats., 433 Paronychia pulvinata Gray, 1288 Silene acaulis L., 538-48 S. menziesii Hook., 174, 822, 882 S. scouleri Hook., 356, 416, 425 Sagina saginoides (L.) Karst., 3844 Stellaria longipes Goldie, 98, 739, 1278 S. jamesiana Torrey, 1392 S. umbellata Turcz., 526 CELASTRACEAE Pachystima myrsinites Raf., 424-48, 544- 48, 24 "9 MADRONO CHENOPODIACEAE Chenopodium album L., 1455 C. fremontii S. Wats., 323 a Monolepis muttalliana (Schult.) Greene, SB2360 COMPOSITAE Achillea lanulosa Nutt., 515-48, 1112, 3879 Agoseris aurantiaca (Hook.) Greene, 22, POL ore A. glauca (Pursh) D. Dietr., 130, 494, 920, 972, 892, 1097, 3936 Anaphalis margaritacea (L.) Benth. & Hook., 528 Antennaria anaphaloides Rydb., 325 A. concinna A. Nels., 350 A. microphylla Rydb., 829, 1136 A. parviflora Nutt., 87 A. rosea Greene, 89-48, 59, 129 Arnica cordifolia Hook., 46, 64, 90, 371, 1367, 1399, 3908 A. fulgens Pursh, 493 A. latifolia Bong., 400 a, 487 A. mollis Hook., 3963 A. parryi Gray, 584, 970, 1395, 3907 A. rydbergii Greene, 1001, 2111 Artemisia dracunculus L., 531-48, 214, 346, 3929, 3945 A. frigida Willd., 90, 1531, 3863 A. ludoviciana Nutt., 371, 825, 841, 3969 A. scopulorum Gray, 215-48, 820, 1047 A. spithamaea Pursh, 390 A. tridentata Nutt., 572, 3864, 3928 Aster bigelovii Gray, 3868, 3930 A. coloradensis Gray, 961 A. engelmannii Gray, 1521, 3903 A. foliaceus Lindl., 511, 1100, 3869, 3943 A. foliaceus Lindl. var. frondosus Gray, 1083 A. glaucodes Blake, 450 A. leucanthemifolius Greene, 507 A. porteri Gray, 1261 Bahia dissecta (Gray) Britton, 466 Brickellia grandiflora (Hook.) Nutt., 1433 Chaenactis alpina (Gray) Jones, 331,585, 1504, 3889, 3897, 3954, 3972 C. douglasii (Hook.) H. & A., 113, 386, 1333 Chrysopsis hispida (Hook.) DC., 464, 3920 C. villosa (Pursh) Nutt., 278, 220, 400 a, 570, 1011, 3947 Chrysothamnus parryi (Gray) Greene, 555, 594, 1121, 1128, 3856 C. viscidiflorus (Hook.) Nutt., 374, 1122, 1127, 3857 Cirsium drummondii T. & G. var. ac- aulescens (Gray) Macbride, 1130 C. eatonii (Gray) Robins., 3887 C. parryi (Gray) Petrak, 3885 C. scopulorum (Greene) Cockl., 921, 1129, 3896, 3962 [Vol. 13 Crepis intermedia Gray, 153, 375, 1337 C. nana Richards, 102, 1427, 1509, 1519, 3922 Erigeron acris L. var. debilis Gray, 3918 E. compositus Pursh var. glabratus Ma- coun, 53 E. coulteri Porter, 488, 923, 3910, 3917 E. elatior (Gray) Greene, 378, 1015, 1041 E. flagellaris Gray, 243 E. lonchophyllus Hook., 361, 595 E. melanocephalus A. Nels., 15, 910, 2148 E. peregrinus (Pursh) Greene, 99, 549, 936, 1023, 2148, 3911 E. pinnatisectus (Gray) Nels., 26, 104, 359 E. simplex Greene, 3956 E. speciosus (Lindl.) DC. var. macran- thus (Nutt.) Cronquist, 367-48, 3880 . subtrinervis Rydb., 478 . superbus Greene, 3919 . unalaschkensis (DC.) Vierh., 3967 . ursinus D. C. Eaton, 3913 . vagus Payson, 283, 432, 988, 1462, 1472, 2055 Gnaphalium wrightii Gray, 1111 Haplopappus clementis (Rydb.) Blake, So H. croceus Gray, 1044, 2095 H. macronema Gray, 903 H. parryi Gray, 495, 564, 587, 1032, 3895 H. pygmaeus (T. & G.) Gray, 1503 H. uniflorus (Hook.) T. & G., 1461, 3960 Helianthella quinquenervis (Hook.) Gray, 1042, 1415 Helenium hoopesii Gray, 381, 1419 Hieracium albiflorum Hook., 2083 H. gracile Hook., 969 Hymenoxys acaulis (Pursh) Parker, 1007 H. grandiflora (T. & G.) Parker, 541-48, 3951 Lactuca pulchella (Pursh) DC., 3986 Matricaria matricarioides (Less.) Porter, 3925 Senecio acutidens Rydb., 109, 151, 388 ambrosioides Rydb., 347, 365, 3862 amplectens Gray, 401, 525, 935, 2153 atratus Gray, 1065, 3865 bigelovii Gray, 924, 1016, 1417 carthamoides Greene, 554, 3921 crassulus Gray, 448, 491, 951 . crocatus Rydb., 953 dimorphophyllus Greene, 1095 fendleri Gray, 1384 harbourii Rydb., 963 integerrimus Nutt., 759 . porteri Greene, 537, 1520, 3899 pudicus Greene, 1066, 3886 . purshianus Nutt., 288 . saxosus Klatt., 78, 1258, 1291, 1350, 2089 . serra Hook., 503, 504, 1082, 1505, 3948 . soldanella Gray, 17, 277, 935, 1458 S. triangularis Hook., 154 et es el NN NNNNNNNNNNNNNHH 1955] S. werneriaefolius Gray, 106, 1234, 1248, 2100 Solidago altissima L., 1431 S. ciliosa Greene, 1096 S. decumbens Greene, 1002 S. glutinosa Nutt., 21, 596 S. sparsiflora Gray, 321, 449, 1435 Taraxacum eriophorum Rydb., 420 T. erythrospermum Andrz., 173 T. officinale Wiggars., 37, 3941 Tetradymia canescens DC., 462, 1355, 1412 Townsendia exscapa (Rich.) Porter, 725, L227 T. incana Nutt., 2013, 2020 T. leptotes (Gray) Osterh., 725, 1292 T. rothrockii Gray, 52, 422, 772, 1256, 1310 ; Tragopogon dubius Scop., 199, 346 Viguiera multiflora (Nutt.) T. & G., 1456, 3881 Wyethia arizonica Gray 224-48, 35 CONVOLVULACEAE Convolvulus arvensis L., 384 CORNACEAE Cornus stolonifera Michx., 423-48, 1315, 3974 CRASSULACEAE Sedum integrifolium (Raf.) A. Nels., 45 S. rhodanthum Gray, 295-48, 556 S. stenopetalum Pursh, 304-48, 460 CRUCIFERAE *Arabis crandallii Robins., 1264 A. drummondii Gray, 265, 728, 957, 3935 A. fendleri (Wats.) Greene, 1336 A. lemmonii Wats., 1116 Brassica kaber (DC.) L. C. Wheeler, 465 Capsella bursa-pastoris (L.) Medic., 309, 3933 Cardamine cordifolia Gray, 29 Descurainia pinnata (Walt.) Britt., 317 D. richardsonii (Sweet) O. E. Schulz, 704, 3942 Draba aurea Vahl, 50, 343, 1272, 1277 D. crassa Rydb., 443 D. crassifolia R. Graham, 428, 819, 844 D. incerta Payson, 3965 D. lanceolata Royle, 60 D. nivalis Lilj. var. exigua (O. E. Schulz) C. L. Hitch., 849, 1078 D. oligosperma Hook., 85, 92, 721, 941, ES D. spectabilis Greene var. oxyloba Gilg & O. E. Schulz, 556-48, 18, 63, 428, 760, 975, 2066 D. ventosa Gray, 1476, 3964 Erysimum alpestre Rydb., 188-48, 722 E. asperum (Nutt.) DC., 133, 755 E. nivale (Greene) Rydb., 61 E. wheeleri Wats., 1282, 2130 Lepidium ramosissimum Nels., 3831 Lesquerella alpina (Nutt.) Wats., 198-48 LANGENHEIM: CRESTED BUTTE (IS Physaria acutifolia Rydb., 295, 720, 1200 Rorippa nasturtium-aquaticum (L.) S. & T., 576-48, 108 R. obtusa (Nutt.) Greene, 418, 1140 Sisymbrium altissimum L., 115 Smelowskia calycina (Desv.) C. A. Mey., 392, 810, 1266, 2107 Thelypodium wrightii Gray, 700-48, 1432 Thlaspi alpestre L., 39, 387 T. arvense L., 330-48, 176, 513 CYPERACEAE Carex aquatilis Wahl., 192, 1089 C. chalciolepis Holm, 918, 1474 C. drummondiana Dewey, 986, 1477, 1478 . ebenea Rydb., 950 . elynoides Holm, 561, 960 . engelmannii L. H. Bailey, 1073, 1075 . festivella Boott, 202, 230 . geyerl Boott, 736, 836, 893, 955 . haydeniana Olney, 1507 . hepburnii Boott, 82, 324, 958, 987, 1043 C. media Vahl var. stevenii (Holm) Fern., 55 C. nigricans C. A. Mey., 1424 C. nova L. H. Bailey, 3-48 C. pyrenaica Wahl., 1110 C. rostrata Stokes, 257-48, 443-48, 192 C. siccata Dewey, 758 Eleocharis pauciflora (Lightf.) Link, 826, 1028, 1528 Kobresia bellardii (All.) Degland, 1052, 1080, 1082 Scirpus americanus Pers. var. polyphyllus (Boeckel) Beetle, 509 S. validus Vahl, 442-48 OGieAe!e.e ELAEAGNACEAE Shepherdia canadensis (L.) Nutt., 150 EQUISETACEAE Equisetum arvense L., 299-48, 1245 E. variegatum Schleich., 548, 1523 ERICACEAE Arctostaphylos uva-ursi (L.) Spreng., 537-48, 147 Chimaphila umbellata (L.) Nutt., 1070 Pyrola asarifolia Michx., 277-48, 513-48, 258, 479, 1387 P. chlorantha Sw., 888 P. secunda L., 145, 482, 532 P. uniflora L., 273-48, 309-48, 519 Vaccinium caespitosum Michx., 1, 1106 V. myrtillus L., 1106 a EUPHORBIACEAE Euphorbia robusta (Engelm.) Small, 74, (its GENTIANACEAE Gentiana amarella L., 467-48, 72, 492, 989, 1105 | G. barbellata Engelm., 1003, 3915, 3953 74 MADRONO G. calycosa Griseb., 73, 568, 1068, 3916, 3952 G. prostrata Haenke, 442, 909 G. romanzovii Ledebour, 1108, 1084, 1466, 3961 G. thermalis Kuntze, 267-48, 269-48, 40, 452 Swertia perennis L., 236-48, 237-48, 1109, 3894 S. speciosa Griseb., 66 GERANIACEAE Geranium fremontii Torrey, 368-48, 783 G. nervosum Rydb., 188 G. richardsonii F. & T., 757 GRAMINEAE Agropyron saundersii (Vasey) Hitchc., 3877 A. scribneri Vasey, 278, 1055 A. trachycaulum (Link) Malte, 247-48, 334, 397, 917, 933, 977, 2136, 2165, 3833 Agrostis alba L., 1137 A. palustris Huds., 508 A. scabra Willd., 1110 b . Arrhenatherum elatius (L.) Presl., 1131 Blepharoneuron tricholepis (Torrey) Nash, 236 Bouteloua gracilis (HBK.) Lag., 1501 Bromus ciliatus L., 238, 1020 B. frondosus (Shear) Woot. & Standl., 352, 787, 978 B. inermis Leyss., 334, 337, 999 B. polyanthus Scribn., 254-48, 329-48, 223, 383, 979, 3854 B. porteri (Coult.) Nash, 3985 B. tectorum L., 1326 Calamagrostis canadensis (Michx.) Beauv., 897 C. inexpansa Gray, 546 C. purpurascens R. Br., 993, 1009, 1037, 1057, 1104 Catabrosa aquatica (L.) Beauv., 349 Deschampsia caespitosa (L.) Beauv., 515, 557, 946, 947, 1509 Elymus glaucus Buckl., 3855 Festuca idahoensis Elmer, 802, 1126 F. ovina L., 931, 1106, 1061 F. ovina L. var. brachyphylla (Schultes) Piper, 13, 412, 550, 793, 798, 817, 1479 F. thurberi Vasey, 527-48, 75, 157, 995, 3878 Glyceria striata (Lam.) Hitchc., 336, 1030 Hordeum brachyantherum Nevski, 3838, 3887 H. jubatum L., 3832 Koeleria cristata (L.) Pers., 895, 896, 2164, 3867 Melica spectabilis Scribn., 328-48 Muhlenbergia montana (Nutt.) Hitchc., 1130, 3849 Oryzopsis hymenoides (R. & S.) Ricker, 370, 902, 2094 [Vol. 13 Phleum alpinum L., 94-48, 100-48, 224- 48, 54. 229) 937 P. pratense L., 243-48, 1125 Poa alpina L., 97-48, 244-48, 11, 229, 948 . arctica R. Br., 246-48, 256-48, 1051 . glaucifolia Scribn. & Will., 1052 . interior Rydb., 824, 3873 . leptocoma Trin., 527, 966 macroclada Rydb., 980, 981 . occidentalis Vasey, 222 palustris L., 66-48, 2139 reflexa Vasey & Scribn., 1403 . rupicola Nash, 43, 415, 821, 929, 983, 1480 Sitanion hystrix (Nutt.) Smith, 335, 1453, 2087, 3847, 3858 Stipa comata Trin. & Ruprecht, 1267 S. columbiana Macoun, 3855 S. lettermani Vasey, 971, 998, 2163, 3872 S. viridula Trin., 333 Trisetum spicatum (L.) K. Richt., 264, 820% 920571037, 1058 a-Racla-Ba-la-ha-la-la-ha-) GROSSULARIACEAE Ribes cereum Dougl., 429-48, 21, 80, 1067, 1235 R. montigenum McClatchie, 389-48, 428- 48, 1318 R. Wolfii Rothr., 391-48, 7, 22, 78, 1067 HYDROPHYLLACEAE Hydrophyllum capitatum Dougl., 16, 1220 H. fendleri (Gray) Heller, 34, 832, 1294 Phacelia leucophylla Torrey, 227, 2138, ZI 7 P. sericea (Graham) Gray, 189, 273, 2023 HYPERICACEAE Hypericum formosum HBK., 454 IRIDACEAE Iris missouriensis Nutt., 30 Sisyrinchium montanum Greene, 514-48, 1524 J UNCACEAE Juncus alpinus Vill., 1529 J. balticus Willd., 1025, 1510 J. drummondii E. Meyer, 887, 1077, 1081, T4715 odo) J. longistylis Torrey, 1527 J. mertensianus Bong., 553 J. parryi Engelm., 554 J. saximontanus A. Nels., 452-48, 191, B91 232. 8271 O8d Luzula parviflora (Ehrh.) Desv., 967, 12761472 L. spicata’ (L.)) D@.7475: LABIATAE Agastache urticifolia (Benth.) Rydb., 345, 1454, 2133, 3940 Dracocephalum parviflorum Nutt., 3842 *Mentha arvensis L., 1448 Monarda fistulosa L., 1437 1955] Monardella odoratissima Benth., 55 Prunella vulgaris L., 105, 190 LEGUMINOSAE Astragalus agrestis Dougl., 199 A. alpinus L., 255, 485 *A. anisus Jones, 1219 A. debilis (Nutt.) Gray, 1021 A. decumbens (Nutt.) Gray, 149 A. drummondii Dougl., 461, 1217 A. occidentalis (Wats.) Jones, 296, 833 Glycyrrhiza lepidota Pursh, 1443, 3950 Hedysarum occidentale Greene, 381-48, 406, 2146 Lathyrus leucanthus Rydb., 72, 751, ZHa3 L. leucanthus Rydb. var. laetivirens (Greene) C. L. Hitch., 143 a Lupinus argenteus Pursh subsp. argen- teus, 187-48, 188-48, 466-48, 957, 1375 L. caudatus Kell. subsp. caudatus, 131, 799 L. lepidus Dougl. ex. Lindl. subsp. caespi- tosus Detling, 1224 L. sericeus Pursh subsp. sericeus, 32, 81 Melilotus officinalis (L.) Lam., 2162 Oxytropis campestris L. var. gracilis (A. Nels.) C. L. Porter, 1056 O. deflexa (Pall.) DC. var. sericea T. & G., 571-48, 312, 583, 951, 2110 O. lambertii Pursh, 457, 2002 O. podocarpa Gray, 402, 408, 724, 2097 Thermopsis montana Nutt., 1208, 2031 Trifolium dasyphyllum T. & G., 559, 1114 T. hybridum L., 337-48 T. nanum Torrey, 1271, 2103 T. parryi Gray, 65, 558, 1087 T. pratense L., 118 Vicia americana Muhl., 752, 2134 LILIACEAE Allium acuminatum Hook., 2004 A. brandegei Wats., 36 A. cernuum Roth., 490-48 A. geyeri Wats., 43 Calochortus gunnisonii Wats., 327 Erythronium grandiflorum Pursh, 211- 48, 62 Lloydia serotina (L.) Sw., 302 Smilacina racemosa (L.) Desf., 182 ». stellata (L.) Desf., 179-48, 57 Streptopus amplexifolius (L.) DC., 836, 1G97. Veratrum californicum Durand, 199-48, 526-48 Zygadenus elegans Pursh, 261-48, 38, 208, 1400 LINACEAE Linum lewisii Pursh, 71 LYCOPODIACEAE Lycopodium annotinum L., 518-48 MALVACEAE Sidalcea candida Gray, 1139, 1383, 1450 S. neo-mexicana Gray, 477, 2006 LANGENHEIM: CRESTED BUTTE 75 NYCTAGINACEAE Allionia linearis Pursh, 469 ONAGRACEAE Epilobium alpinum L., 536, 1518 E. angustifolium L., 483, 3902 E. latifolium L., 533, 3898 E. lactiflorum Hausskn., 211 E. paniculatum Nutt., 322, 1131, 2159, S875: Gayophytum nuttallii T. & G., 2160, 3840 Oenothera caespitosa Nutt., 1268 O. flava (Nels.) Garrett, 1413 O. hookeri T. & G., 377-48 OPHIOGLOSSACEAE Botrychium matricariaefolium A. Br., 3837 ORCHIDACEAE Calypso bulbosa (L.) Oakes, 740 Corallorhiza maculata Raf., 2080 C. trifida Chat., 1368, 1423 Goodyera oblongifolia Raf., 2085 Habenaria hyperborea (L.) R. Br., 254, 545, 900 H. obtusata (Pursh) Richards, 319-48 H. unalascensis (Spreng.) Wats., 2081 Listera cordata (L.) R. Br., 1005 Spiranthes romanzoffiana Cham., 889 OROBANCHACEAE Orobanche fasciculata Nutt., 123 PAPAVERACEAE Corydalis aurea Willd., 101, 1257 C. caseana Gray subsp. brandegei (Wats.) G. B. Ownbey, 588, 1600, 2117 Papaver radicatum Rottb., 434, 806, 1458 PINACEAE Abies lasiocarpa (Hook.) Nutt., 97, 3970 Juniperus communis L., 144 J. scopulorum Sargent, 1366, 1440, 2113 Picea engelmannii (Parry) Engelm., 620-48, 3927 P. pungens Engelm., 621-48, 3983, 3984 Pinus aristata Engelm., 1287 P. contorta Dougl. var. latifolia Engelm., 613-48, 3893 P. edulis Engelm., 1439 P. flexilis James, 1008, 1237, 3975 P. ponderosa Laws. var. scopulorum Engelm., 1422 Pseudotsuga taxifolia (Poir.) Britton, 93, 3926 PLANTAGINACEAE Plantago major L., 3830 P. tweedyi Gray, 3833 POLEMONIACEAE Collomia linearis Nutt., 525-48, 348 Gilia aggregata (Pursh) Spreng., 85, 753, 780, 3931 G. attenuata (Gray) A. Nels., 1006 G. calcarea Jones, 1202 76 MADRONO Leptodactylon pungens (Torrey) Rydb. subsp. eupungens (Brand) Wherry, 403, 703, 1262 Phlox caespitosa Nutt., 16, 1107, 1279 P. caespitosa Nutt. subsp. pulvinata Wherry, 370-48, 543-48 P. multiflora A. Nels. subsp. depressa (A. Nels.) Wherry, 116, 1000 Polemonium confertum Gray, 24, 3912 P. delicatum Rydb., 6, 99, 136 P. foliosissimum Gray subsp. archibaldae (A. Nels.) Wherry, 316-48, 193 P. viscosum Nutt., 47, 779 POLYGONACEAE Eriogonum coloradense Small, 171-48, 198, 285, 1062, 3874 E. neglectum Greene, 120, 234, 489, 543, 839 E. racemosum Nutt., 376, 1534 E. subalpinum Greene, 326-48, 233, 756 Oxyria digyna (L.) Hill, 462-48, 423, 1517 Polygonum aviculare (L.) Lam., 2161, 3883, 3987 P. bistortoides Pursh, 291 P. douglasii Greene, 501 P. viviparum L., 175-48, 50 Rumex acetosella L., 269 R. densiflorus Osterh., 212, 3835 R. triangulivalvis (Danser) Rech., 3833 POLYPODIACEAE Cryptogramma crispa (L.) R. Br. var. acrostichoides (R. Br.) C. B. Clark, 540, 884 Cystopteris fragilis (L.) Bernh., 61, 512 *Pellaea occidentalis (E. Nels.) Rydb., 2022 *Pteridium aquilinum (L.) Kuhn var. pubescens Underw., 426-48 PORTULACACEAE Claytonia lanceolata Pursh, 382-48, 460-48 C. megarhiza (Gray) Parry, 429, 1305 Lewisia pygmaea (Gray) Robins., 44, 498, 1369 PRIMULACEAE *Androsace carinata Torrey, 1428 A. septentrionalis L., 343-48, 8, 28 Dodecatheon radicatum Greene, 226, 310 *Primula angustifolia Torrey, 1289, 1301 P. parryi Gray, 437-48, 440-48 RANUNCULACEAE Aconitum columbianum Nutt., 127-48, 509-48 Actaea arguta Nutt., 778, 1255 Anemone globosa Nutt., 130-48, 534-48, 99 A. narcissiflora L., 546-48, 14, 1365 A. parviflora Michx., 554-48, 74, 496, 1306, 1365 [Vol. 13 A. patens L., 185, 726, 1243 Aquilegia coerulea James, 134-48, 215, 840, 962, 2119 A. elegantula Greene, 736, 763, 2096 A. coerulea James X A. elegantula Greene, 730, 885 *A. micrantha Eastw., 1382, 2021 Caltha leptosepala DC., 540-48, 29, 1251 *Clematis ligusticifolia Nutt., 1446, 1449 Delphinium barbeyi Huth., 14, 209, 1418 D. nelsonii Greene, 727 Ranunculus adoneus Gray, 112-48, 578- 48, 579-48, 419, 934, 982, 1390 R. alismaefolius Geyer var. montanus S. Wats., 111-48, 581-48, 590, 701, 851, 1226 R. cymbalaria Pursh, 472 R. cymbalaria Pursh var. saximontanus Fern., 738 R. eschscholtzii Schlecht., 582-48, 883-48, 171, 520, 718 84710860 1258 R. inamoenus Greene, 7, 257, 314, 702 R. inamoenus Greene var. alpeophilus (A. Nels.) L. Benson, 766, 1048 R. uncinatus Don var. earlei (Greene) L. Benson, 580-48, 852, 1500 Thalictrum alpinum L., 1296, 1373, 1525 T. fendleri Engelm., 181, 775, 837, 2084, 3949 Trollius laxus Salisb. var. albiflorus Gray, 166, 170, 1250 RHAMNACEAE Ceanothus fendleri Gray, 458, 776, 1378 ROSACEAE Amelanchier pumila Nutt., 152-48, 475-48 Cercocarpus montanus Raf., 79, 1327 Dryas octopetala L., 824, 2108 Fragaria ovalis (Lehm.) Rydb., 153-48, 477-48, 4, 27, 520 Geum macrophyllum Willd., 256 G. rivale L., 160, 261 G. rossii Seringe, 563-48, 3955 G. triflorum Pursh, 13 Holodiscus discolor (Pursh) Maxim var. dumosa (Nutt.) Dippell, 480-48, 476, 898, 1414, 2125 Ivesia gordonii (Hook.) T. & G., 83, 224 Potentilla anserina L., 326 P. arguta Pursh, 51 P. diversifolia Lehm., 58, 292, 1049 P. fruticosa L., 111, 497, 3852 P. hippiana Lehm. x P. pulcherrima Lehm., 781, 3848, 3859 P. nivea L., 294, 304 P. norvegica L., 3890 P. pennsylvanica L., 158, 3884 P. pulcherrima Lehm., 88, 3861 P. rubricaulis Lehm., 391, 770, 773, 994, 1297 Prunus virginiana L., 67, 1317 Purshia tridentata (Pursh) DC., 568-48, 468, 1260 1955] Rosa woodsii Lindl., 368-48, 364, 3851 *Rubus parviflorus Nutt., 1323 R. strigosus Michx., 139 Sibbaldia procumbens L., 568-48, 197, 1468, 2155 Spiraea caespitosa (T. & G.) Nutt., 91-48, 1480, 1502, 3971 Sorbus scopulina Greene, 225 RUBIACEAE Galium bifolium Wats., 3836 G. boreale L., 350-48, 112, 135, 3882 G. trifidum L., 3846, 3923 G. triflorum Michx., 1386 SALICACEAE Populus angustifolia James, 1034, 1069 P. balsamifera L., 177, 1067, 3973 P. tremuloides Michx., 1537 Salix anglorum Chamisso var. antiplasta Schneider, 547-48, 101, 108, 110, 111, 410, 447 b, 818, 1094 S. barclayi Andersson, 341-48, 376-48, 41, 93, 102, 106, 107, 118, 133, 894 S. barclayi Andersson var. hebecarpa An- dersson, 120-48 S. brachycarpa Nutt., 373-48, 376-48, 105, 109, 245, 411, 486, 581, 716 S. cascadensis Cockl., 985 S. drummondiana Barratt var. subcoeru- lea (Piper) Ball, 28, 84, 715 . Irrorata Andersson, 700 . melanopsis Nutt., 732 . hivalis Hook., 447 a . nivalis Hook. var. saximontanus (Rydb.) Schneider, 103-48, 307, 597 . planifolia Pursh var. monica (Bebb) Schneider, 104-48 . pseudocordata (Andersson) Rydb. var. aequalis Andersson 707, 731, 733 S. pseudolapponum Von Seeman, 549-48, 48, 112, 117, 298, 308, 342, 807, 940, 1093 S. scouleriana Barratt, 253, 717 S. scouleriana Barratt var. coetanea Ball, EON 215) S. wolfii Bebb, 714, 115, 116, 318 WM WM MMMM SAXIFRAGACEAE Heuchera parvifolia Nutt. var. flavescens R. B. & L., 194, 1274 Lithophragma bulbifera Rydb., 1232 Mitella pentandra Hook., 835, 1004 M. stauropetala Piper var. stenopetala (Piper) Rosend., 164, 769 Parnassia fimbriata Banks, 598-48 Saxifraga arguta Don, 600-48, 259, 502 S. bronchialis L., 601-48, 360 'S. caespitosa L. subsp. exaratoides (Si- mon) Engl. & Irmsh., 430, 1464, 2101 S. cernua L., 804, 1470 S. columbiana Howell, 1091 S. chrysantha Engelm., 1460 S. debilis Engelm., 499, 881 LANGENHEIM: CRESTED BUTTE 77 S. flagellaris Willd. var. platysepala Traut., 435, 803, 1459 S. oregana Howell, 259, 311, 1091 S. rhomboidea Greene, 546-48, 599-48, 602-48, 289, 424 Sullivantia purpusii (Brand) Rosend., 1020 SCROPHULARIACEAE Besseya alpina (Gray) Rydb., 438, 2104 Castilleja chromosa A. Nels., 708, 1210, 2011 C. linariaefolia Benth., 219, 3932 C. occidentalis Torrey, 1079, 2057, 3958 C. rhexifolia Rydb., 394-48, 3959 C. septentrionalis Lindl., 495-48, 235 Chionophila jamesii Benth., 436, 809, 1469 Mimulus guttatus DC., 406-48, 582 Orthocarpus luteus Nutt., 357 Pedicularis crenulata Benth., 411-48, 68 P. groenlandica Retz., 398-48, 572-48, 557 P. parryi Gray, 565, 2147 P. paysoniana Pennell, 204, 249, 973 P. procera Gray, 355, 1416 P. racemosa Dougl., 244, 2150 P. scopulorum Gray, 69, 815 Penstemon barbatus (Cav.) Roth. subsp. torreyi (Benth.) Keck, 492-48 P. caespitosus Nutt., 83 P. crandallii A. Nels., 774 P. harbourii Gray, 412-48, 942, 2152, 3900 P. humilis (Nutt.) Gray, 575-48 P. rydbergii A. Nels., 498-48, 196, 1332 P. strictus Benth., 132, 339 P. teucrioides Greene, 1298 P. whippleanus Gray, 420-48, 248, 490, 1391 Veronica alpina L., 405-48, 40, 2149 V. americana Schwein., 80, 790, 2000 V. serpyllifolia L., 3845 SELAGINELLACEAE Selaginella densa Rydb., 56, 3967, 3988 SPARGANIACEAE Sparganium angustifolium Michx., 454- 48 UMBELLIFERAE Angelica ampla. Nels., 3841, 3892 A. grayi C. & R., 418-48, 919, 2158 Conioselinum scopulorum (Gray) C. & R., 1026, 3906 Heracleum lanatum Michx., 504-48, 505- 48, 377 Ligusticum porteri C. & R., 503-48, 200, 216, 994, 3905, 3934 Lomatium dissectum (Nutt. ex Torr. & Gray) Mathias & Constance, 2142 L. simplex (Nutt.) F. Macbr., 712, 771, 1274 Oreoxis alpina (Gray) C. & R., 252, 440, 710, 2036 Osmorhiza obtusa (C. & R.) Fern., 789, 954, 1398 78 MADRONO O. occidentalis (Nutt.) Torrey, 217, 3901 Oxypolis fendleri (Gray) Heller, 419-48, 94, 228, 530, 1092 Pseudocymopterus montanus (Gray) C. & R., 422-48, 519-48, 564-48, 6, 114, 385, 711, 723 [Vol. 13 VIOLACEAE Viola adunca Sm., 540-48, 585-48, 590-48, 592-48, 2, 53, 58, 286, 567, 1241 V. adunca Sm. subsp. ashtonae Baker, 212-48, 597-48 V. canadensis L., 595-48, 26 Urtica gracilis barons V. nuttallii Pursh, 586-48, 488-48, 591-48, VALERIANACEAE 593-48, 12, 1221 Valeriana capitata Pall. subsp. acutiloba V- Palustris L., 1425 (Rydb.) F. G. Meyer, 5, 300, 404, 1050 V.rugulosa Greene, 587-48, 589-48, 594- V. edulis Nutt. ex. T. & G., 1040, 1364, 48 1421 Department of Botany, University of California, Berkeley LITERATURE CITED BAKER, C. F. 1898. Botanizing in the La Plata Mountains. Plant World 2:29-32. BAKER, F. S. 1925. Aspen in the Central Rocky Mountain region. U.S. Dept. Agric. Bull. 1241:1-46. BRANDEGEE, T.S. 1876. The flora of southwestern Colorado. Bull. U.S. Geol. and Geog. Survey Terr. 2:227-248. Cary, M. 1911. A biological survey of Colorado. No. Am. Fauna 33. 256 pp. Exuison, L. 1954. Subalpine vegetation of the Wasatch Plateau, Utah. Ecol. Monog. 24:89-184. . Festuca thurberi grassland. Indian Creek, Albajo Mts., Utah. Unpublished data in files of Intermountain Forest and Range Experiment Station, U.S. Forest Service, Ogden, Utah. GraHAM, E. H. 1937. Botanical studies in the Uinta Basin of Utah and Colorado. Ann. Carnegie Mus. 26. 432 pp. LANGENHEIM, J. H. 1953. The plant communities and their environment in the Crested Butte area, Gunnison County, Colorado. Ph.D. thesis (unpublished), Univ. Minnesota. Lut, H. and L. Errison. 1950. Precipitation in relation to altitude in Central Utah. Ecology 30:479-484. Porter, T. C. and J. M. Coutter. 1874. Synopsis of the flora of Colorado. U.S. Geol. and Geog. Survey Terr. Misc. Publ. 4. 180 pp. Rossins, W. W. 1910. Climatology and vegetation in Colorado. Bot. Gaz. 49:256—280. SCHMOLL, H. M. 1935. Vegetation of the Chimney Rock area, Pagosa-Piedra region, Colorado. Private edition, distributed by Univ. of Chicago Libraries, Chicago, Il. TANNER, V. M. and C. L. HAywarp. 1934. A biological survey of the LaSal Mts., Utah. Utah Acad. Sci. 11:209-235. UnItTeEpD STATES DEPT. AGRICULTURE YEARBOOK. 1941. Climate and Man: 798-808. UnitTEeD STATES DEPT. COMMERCE. 1952. Climatological Data, Annual Summary, Col- orado 57:172-181. WENT, F. W. 1953. Annual plants at high altitudes. Madrono 12:109-114. REVIEWS Manual of the Plants of Colorado. By H. D. HaArrincton. x + 666 pp., Sage Books, Denver. $8.00. The identification of plants in much of the Rocky Mountain area has been greatly hampered for many years by the lack of up-to-date manuals. The two manuals which have been of most value for Colorado, namely, Coulter and Nelson’s “New Manual 1955] REVIEWS 79 of Rocky Mountain Botany” (1909) and Rydberg’s “Flora of the Rocky Mountains and adjacent plains” (1917) are not only out of print but are considerably outdated in their nomenclature and taxonomy. A reflection of the need for an up-to-date treat- ment may be gained from a statement in the introduction of the present work that “about one out of every 30 species listed here constitutes a new record for the state or at least is not credited to Colorado in the monographs and manuals.” Dr. Harrington, Associate Professor of Botany and Curator of the Herbarium at Colorado A. & M. College, Fort Collins, has prepared analytical keys and complete descriptions for the identification of 2,794 species and 351 infraspecific taxa of ferns and flowering plants in the state. He has incorporated about 350 entities in the text for which he has seen no actual specimens from the state, but which are plants actually reported for Colorado or listed from adjacent areas and to be expected within the state. However, the source of these reported occurrences is not indicated. Following the introduction there is a discussion of the vegetation zones in Colo- rado contributed by David F. Costello, Range Conservationist of the United States Forest Service. This provides a general topographic and floristic analysis of the veg- etation with the characteristic taxa of the various vegetation types enumerated. A key to the families precedes the main taxonomic treatment and a complete glossary and index to common names and genera concludes the volume. In the taxonomic treatment proper, which forms the bulk of the manual, estab- lished common names are employed (but none are coined) for families and genera. No common names, however, are given for species. Infraspecific taxa are cited as straight trinomials with their rank indicated in parentheses following the name. The place of publication for each species is given. The book has been printed by photo- offset from typescript and the small type (especially in the keys) is not easily read. Oftentimes, the impression is unevenly reproduced so that there are lighter or darker sections on the same page. Alignment of the right hand margins of the typescript would have improved the appearance of the finished text and would also have effected an economy with respect to the number of pages. Dr. Harrington has maintained a fairly conservative viewpoint in his delimitation of taxa and much of the work is in accord with current treatments. He has wisely drawn upon the assistance of specialists for the accounts of some genera, either in whole or in part: Scirpus by Alan Beetle, Juncus by F. J. Hermann, Salix by E. C. Smith, Oxytropis and Astragalus by C. L. Porter, Penstemon by C. Wm. T. Penland, Hymenoxys by K. F. Parker, and Balsamorhiza, Wyethia and Helianthella by W. A. Weber. Likewise, contemporary taxonomic papers have been closely followed in the preparation of many other groups. The manual appears to be a careful compilation of existing knowledge regarding the Colorado flora, and notes of critical value, while limited, are scattered throughout the text. The distributions given for the various species are based on specimens actually studied and these seem to have been prin- cipally specimens in the herbaria of the Colorado-Wyoming area. Valuable collec- tions of Colorado plants in some of the larger herbaria of the United States appear not to have been consulted. The reviewer has noted some Colorado collections cited in taxonomic papers from these larger herbaria which either extend the range of the species or establish an otherwise doubtful record for the state and which are not accounted for in the present manual. Although altitudinal ranges are given for most species, an arbitrary device is employed which divides the state into nine equiformal areas for describing the range of each species. The ponderous size (8%4” & 11” and 2” thick) scarcely makes the book a “manual” that can be used with ease in the field. The technical approach and the lack of illustrations, as well as the cost, may deter the lay student or interested amateur from using the book. Nevertheless, the manual represents a milestone in Colorado botany which should prove of much help to the professional taxonomist. It is to be hoped that it will provide a stimulus and a working foundation for much needed studies of a highly interesting and rich flora——G. THomas Rossins, Department of Botany, University of California, Berkeley. 80 MADRONO [Vol. 13 The Polyporaceae of the United States, Alaska and Canada. By LEE ORAS OVER- HOLTS. Prepared for publication by Josiah L. Lowe. University of Michigan Scientific Series, Volume XIX. xiv + 466 pp., 132 pls. 1953. University of Michigan Press, Ann Arbor. $7.50. Mycologists, plant pathologists and others have long awaited the appearance of this work as Dr. Overholts was the leading specialist in the Polyporaceae in America for many years. It is the first unified comprehensive treatment of this family for the area. The first draft of the manuscript was written in 1933, but publication facilities were not secured at that time. Dr. Overholts made corrections and additions for new material as his health would allow until the time of his death in 1946. It is especially fortunate that Dr. Josiah L. Lowe, a widely known and able specialist in this field was willing to complete the work necessary for publication of the manual. He has included new material in the keys, text, and illustrations and has brought the whole to a well-finished product. The introduction gives a summary of the history of classifications of the family, and of the variations in generic concepts held by different students of these fungi. An explanation is given of the morphological characters and terminology employed. Also the biology and the role of these fungi in nature is discussed. The taxonomic treatment is the conservative one that characterized Dr. Overholts’ earlier publications. The key to genera limits the family to ten old, long established genera. Two of these, the non-pileate genera Merulius and Poria, are not treated in this manual. This treatment is in striking contrast to those of several other workers who have segregated a large number of other genera within the family. The author defends his position in this matter in his introductory chapter. The more recent publications of Donk (1933), Pilat (1936-1942), Cunningham (1946-1948), and Pinto-Lopes (1952) employ quite different criteria as their basis of classification. The first two are cited, but none of them is reviewed, since they appeared after the manuscript was written. The keys to species are well constructed and simple. A conservative synonomy for each species is provided. Clear concise technical descriptions, using both macroscopic and microscopic characters, are given for each species as well as habitat, distribution and published illustrations. Two hundred thirty-five species and twelve varieties are so treated. An outstanding feature of this manual is the abundant and excellent illustrations consisting of 675 half-tone illustrations and more than 200 line drawings. The half- tones, although necessarily reduced in size, are clear and furnish information as to macroscopic characters, while the line drawings illustrate the character of the hyme- nium and spores. The citation of publication data for each species, the synonyms listed, and the bibliography of 238 titles provide excellent references to the literature of the field. A glossary furnishes an explanation of the terminology used. This manual will be a valuable standard work for many years to come, and the editor and publishers are to be congratulated on the quality of the publication — LEE Bonar, Department of Botany, University of California, Berkeley. NOTES AND NEWS Authors of scholarly works which may be of great importance to a particular field and yet, because of their probable limited sale, may be too expensive for publication in the United States, will be interested to learn of the organization of the International Scholars Forum. The Advisory Board of the Forum has arranged to receive and ap- praise for publication by a European press such manuscripts as may be submitted to them by American authors. Details of the plan may be obtained from the Librarian of the Honnold Library, Claremont, California. INFORMATION FOR CONTRIBUTORS Manuscripts submitted for publication should not exceed 20 pages when printed, or contain more than 20 per cent illustrative material including tabular matter, unless the author agree to bear the additional cost. Subject to the approval of the Editorial Board, manuscripts may be published ahead of schedule as additional pages to the issue, provided the author assume the complete cost of publication. Shorter items, such as range extensions and other biological notes, will be published in condensed form with a suitable title under the general heading, “Notes and News.” Institutional abbreviations in specimen citations should follow Lanjouw and Stafleu’s list (Index Herbariorum. Part 1. The Herbaria of the World. Utrecht. Second Edition, 1954). Articles may be submitted to any member of the Editorial Board. MADRONO A WEST AMERICAN JOURNAL OF BOTANY A quarterly journal devoted to the publication of botanical re- search, observation, and history. Back volumes may be obtained from the Secretary at the following rates: VolumeI,1916-1929 . . . . . $5.00 Volume IT, 1930-1934. . . . . 5.00 Volume III, 1935-1936 . . . . 5.00 Volume IV, 1937-1938 . . .. . 5.00 Volume V, 1939-1940. . . .. . 5.00 Volume VI, 1941-1942 . . .. 5.00 Volume VII, 1943-1944 . . . 5.00 Volume VIII, 1945-1946 . . .. 5.00 Volume IX, 1947-1948. me wee te. 6.00 Volume X, 1949-1950. . . .. . 7.00 Volume XI, 1951-1952... wt, 8.00 Volume XII, 1953-1954 . . ... 8.00 Single numbers, Volumes'l and 2> 5 (° 2) 2) & 50 Succeeding numbers, (except Volume XI, No.2). . . 1.00 Volume X1,,.No.27.2 "2 25 We 1.50 The subscription price of MapRoNo is $4.00 per year. If your institution does not now subscribe to MADRONO, we would be grateful if you would make the necessary request. Since there is no extra charge for institutional subscriptions, an individual may hold membership in the California Botanical Society on the basis of his institution’s subscription. Address all orders to: G. THomMAS Rossins, Corresponding Secretary Department of Botany University of California, Berkeley 4, California VOLUME 13, NUMBER 3 JULY, 1955 Contents ERNEST BRowN Bascock, G. L. Stebbins 81 SOME CONSIDERATIONS OF THE GENERA ECHINOCYSTIS AND ECHINOPEPON IN THE UNITED STATES AND NorTHERN Mexico, Kenneth M. Stocking 84 MOSSES OF CALIFORNIA V. PTERIGONEURUM OVATUM AND P. SUBSESSILE, Leo Francis Koch 100 CHROMOSOME NUMBERS IN MENTZELIA (LOASACEAE), Henry J. Thompson and Harlan Lewts 102 CHROMOSOME COUNTS IN THE SECTION SIMIOLUS OF THE GENUS MIMULUS (SCROPHULARIACEAE), Robert K. Vickery, Jr. 107 NoTEs ON NEvADA MimuLus, Gabriel Edwin 110 Notes AND News: MONOCHORIA VAGINALIS IN CALI- FORNIA, J. M. Tucker and B. J. McCaskill 112 A WEST AMERICAN JOURNAL OF BOTANY PUBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY ey MADRONO A WEST AMERICAN JOURNAL OF BOTANY Entered as second-class matter at the post office at Berkeley, California, January 29, 1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price $4.00 per year. Published quarterly and issued from the office of Madrofio, Herbarium, Life Sciences Building, University of California, Berkeley 4, California. BOARD OF EDITORS HERBERT L. Mason, University of California, Berkeley, Chairman Epcar ANDERSON, Missouri Botanical Garden, St. Louis. Lyman BENSON, Pomona College, Claremont, California. HERBERT F. COPELAND, Sacramento College, Sacramento, California. Joun F. Davinson, University of Nebraska, Lincoln. IvAN M. JounstTon, Arnold Arboretum, Jamaica Plain, Massachusetts. Miuprep E. Marutas, University of California, Los Angeles 24, Marion OWNBEY, State College of Washington, Pullman. Ira L. Wiccrns, Stanford University, Stanford, California. Secretary, Editorial Board — ANNETTA CARTER Department of Botany, University of California, Berkeley. Business Manager and Treasurer — R1icHARD W. HoLm Natural History Museum, Stanford University, Stanford, California. CALIFORNIA BOTANICAL SOCIETY, INC. President: Lincoln Constance, Department of Botany, University of California, Berkeley, California. First Vice-president: John Thomas Howell, California Academy of Sciences, San Francisco, California. Second Vice-president: Mildred E. Mathias, Department of Botany, University of California, Los Angeles, California. Recording Secretary: Mary L. Bowerman, Department of Botany, University of California, Berkeley, California. Corresponding Secretary: G. Thomas Robbins, Department of Botany, University of California, Berkeley, California. Treasurer: Richard W. Holm, Natural History Museum, Stanford University, Stanford, California. Annual membership dues of the California Botanical Society are $4.00 for one person or $5.00 for two members of the same family. Other classes of membership in the Society are: Life ($100), Sustaining ($25 annually), and Honorary (elective). All types of membership carry with them one subscription to Madrofio and all other privileges of the Society. Dues should be remitted to the Treasurer. General corre- spondence, changes of address, and applications for membership should be addressed to the Secretary. 1955] STEBBINS: BABCOCK | 81 ERNEST BROWN BABCOCK 1877-1954 With the passing of Ernest B. Babcock in Berkeley, California, on De- cember 8, 1954, the botanists of California lost one of their most dis- tinguished colleagues and a kindly, warm-hearted friend. Professor Babcock was best known to botanists as the author of one of the finest monographs in the field of plant science, his monumental work on Crepis. This work, which was the climax of thirty years of intensive research by himself and his associates, brought together a larger body of information about the systematics, cytogenetics, and evolution of a large genus than had ever been gathered before its publication in 1947. His remarkable collection of herbarium specimens of species of Crepis and related genera, part of them collected in the field, and part of them made from plants grown in the garden and greenhouse, was presented before his death to the Herbarium of the University of California. As the author of several new species of Crepis, he combined the meticulous attention to accuracy of detail which is necessary for a taxonomist with the vision and understanding which he used to excellent advantage in the ensuing long series of experiments and observations in the field of cytogenetics. Above all, he was an evolutionist. A deeply religious man, he looked upon the successive evolution of ever more varied forms of life as the strongest evidence which man can obtain regarding the master plan of the Creator. Mr. Babcock was born in Edgerton, Wisconsin, on July 10, 1877. He came to California in 1896 with his parents, Emilus Welcome and Mary Eliza (Brown) Babcock. His interest in plants had already begun at that time, since he more than once told his friends about the enthusiasm with which, as a child, he tended the plants in his mother’s conservatory. After graduating from the Los Angeles Normal School, he went to the University of California, where he remained for the rest of his life. He received the B.S. degree in 1905, and in 1907 was appointed to the University faculty, on which he served for forty years. In 1913, he became chairman of the newly founded Division of Genetics in the College of Agriculture, and retained that position until his retirement in 1947. Professor Babcock was introduced to the flora of California by Harvey Monroe Hall, who became his friend during undergraduate years in Ber- keley, and with whom he took several long botanizing trips in the High Sierra. Until Dr. Hall’s death, he and Babcock were inseparable friends. They published together a systematic and genetic study of Hemizonia, and Dr. Hall’s stimulus did much to inspire Professor Babcock to begin his study of Crepis. In 1908, he married Georgia Bowen, a childhood friend and neighbor from Edgerton, Wisconsin. She survives him. Professor Babcock’s career is noted for a number of accomplishments in the field of Genetics and Plant Breeding. The most noteworthy of these Maprono, Vol. 13, No. 3, pp. 81-112, July 8, 1955. ‘Jat 751955 82 MADRONO [Vol. 13 ERNEST Brown BABCOCK was the authorship, with his colleague R. E. Clausen, of the textbook, Genetics in Relation to Agriculture. This work appeared in two editions, the first one in 1918 and the second in 1927. For many years it was a standard textbook in the field of Genetics. In his earlier years as a plant breeder he made the hybridizations which led to the production of the Babcock peach, a variety which until recent years was one of the fore- most of the varieties suited to the warmer parts of California. When Professor Babcock began his studies of the genus Crepis in 1920, the idea of using cytogenetic information to clarify taxonomic relation- ships was a novel one. The enthusiasm with which he and his associates embarked upon this project, the obvious value of even their earlier pub- lications, and the striking clarity of the exhibits which he presented at various meetings of national scientific societies and international con- gresses did much to impress botanists with the importance of this type of work. To those botanists interested in evolution who attended the Sixth International Congress of Genetics at Ithaca in 1932, Professor Babcock’s “Evolutionary Tree” of Crepis, with living plants of representative species and illustrations of their chromosomes placed side by side in order of 1955] STEBBINS: BABCOCK 83 their phylogenetic position, became a major center of interest, and made a lasting impression. Not long after he had begun his studies of the Old World species of Crepis, he turned his attention to the much smaller number of species endemic to North America. He found there a pattern of relationships entirely different from that in the Old World, and based upon hybridiza- tion, polyploidy, and particularly apomixis. With characteristic thorough- ness, and in collaboration with the present writer, he worked out the entire pattern of relationships among these species, and published this study as the first monographic contribution on the genus itself. Mean- while, he had become interested in the genera related to Crepis, and either carried out himself or promoted the execution by his associates of a series of monographic studies which showed clearly the position of the genus within the family Compositae. These preliminary monographic studies paved the way for the publica- tion of the final great work, which coincided with Professor Babcock’s retirement from the active faculty of the University of California. Pro- fessor Babcock’s monograph of Crepis has been acclaimed by all who have used it as taxonomically sound, broad in scope, and meticulously accurate in execution. It will remain for many years as a classical model for monographic studies. Professor Babcock’s career did not end with his retirement. During the last seven years of his life, he devoted much energy toward advancing the cause of scientific research, both in connection with the California Academy of Sciences, of which he became President in 1953, and with the Institute of Forest Genetics at Placerville. As Vice-President of the Forest Genetics Foundation, he was the stimulus of a nation-wide move- ment to encourage research toward the utilization of the best germ plasm available in the restoring of our forest resources by replanting. This final activity emphasized one of Professor Babcock’s outstanding qualities. For his inspiration he always looked toward the future rather than back- ward toward the past. As might be expected of a scientist with such outstanding attainments, Professor Babcock belonged to a large number of scientific societies, and became an officer of many of them. He was President of the California Botanical Society in 1941, and of the Society for the Study of Evolution in 1952. He was a member of the American Society of Naturalists, the Washington Academy of Sciences, and the National Academy of Sciences. His Alma Mater recognized his preeminence in 1944 by awarding him the Faculty Research Lectureship and in 1947 by conferring upon him the honorary degree of LL.D. All those who knew him, whether intimately or casually, whether as students, colleagues, or friends, recognized in Ernest Babcock an excep- tional personality. His broad understanding of nature, combined with his kindliness, generosity, and faith in the goodness of all mankind created a feeling of good will among a wide circle of humanity.—G. L. STEBBINS, Division of Genetics, University of California, Davis. 84 MADRONO [Vol. 13 SOME CONSIDERATIONS OF THE GENERA ECHINOCYSTIS AND ECHINOPEPON IN THE UNITED STATES AND NORTHERN MEXICO KENNETH M. STOCKING The genera Sicyos and Sicyoides as originally proposed in 1737 by Linnaeus encompassed, in addition to the Old World species, what were later to be recognized as the New World genera Echinocystis, Echino pe pon, and Marah. The present writer is treating these three New World genera herein and in a subsequent paper. In addition to these three genera the following key includes Brandegea and Vaseyanthus, two genera whose ranges coincide in part with those of Echinopepon and Marah, plants of which, in sterile condition, have often been confused with them. KEY TO THE FIVE GENERA 1. Ovary 1-celled, ovule 1; fruit usually smooth or nearly so, less than 1 cm. long; Perennld! 2) esa POP Ho eae eh ete Boy ice eee Brandegea 1. Ovary more than 1-celled, ovules more than one; fruits usually conspicuously spiny, more than 1 cm. long; annuals except Marah. 2. Plants perennial, stems from large perennial tuber, seeds at least 7 mm. thick. SP ee, re tie eee, oe ae Marah 2. Plants annual, seeds less than 6 mm. thick. 3. Sepals and petals 6; seeds large, 14-18 mm. long; northern Arizona, northern New Mexico and northeastern United States . . . Echinocystis 3. Sepals and petals 5; seeds less than 10 mm. long; Mexico, southern Arizona, and New Mexico. 4. Fruit globose; ovules normally 2; Lower California . . Vaseyanthus 4. Fruit ovoid-cylindrical, ovules 4 or more, Argentina to Mexico, southern Arizona, and southern New Mexico . . . . . . Fic. 1. Early stages in germination of Marah seeds; a, M. macrocar pus var. macro- carpus; b, M. horridus; c, M. oreganus; d, M. watsonii; e, M. fabaceus var. agrestis. Roots AND TuBErs. Tubers have been observed to develop from the hypocotyl in Marah fabaceus, M. horridus, M. macrocarpus, M. oreganus, and M. watsonu. The writer has not germinated seeds of M. gilensis or M. guadalupensis. During the first five years tubers have been seen to have a shape which tapers above toward the stem base and less abruptly below toward the tap root. From a study of numerous more mature tubers of the above mentioned species which the author has dug out and re- planted, it can be conservatively estimated that the tubers remain fusi- form for several to many years. As tubers mature, they become more globose. Some of the heaviest specimens dug, especially those of M. mac- rocarpus, had a greater diameter than length. These were growing where rocks limited downward growth of large organs. The basal portion of one large tuber of /. horridus was split so that it somewhat resembled the legs of a man. The heaviest tuber dug by the author weighed 58 kg. (fig. 2). One has been reported to weigh about 90 kg. (Science News Letter, 1948). Beneath the bark, which becomes quite rough and thick in old tubers, 1955 | STOCKING: MARAH el) lies the large, bulky area of starchy parenchyma which contains concen- tric rings of xylem vessels. (In digging to locate the tubers, the author knows, when he smells an odor like that of fresh-cut potatoes, that he has cut into a tuber.) The core of old tubers is dark, hard, and without starch. The concentric rings in the starchy parenchyma are not well enough defined to allow a very accurate estimate of age; however, the author did estimate the age of the 58 kg. specimen (fig. 2) to be fourteen years. In one growing season and in adverse light conditions, one tuber of M. fabaceus var. agrestis reached a length of 13 cm. and a diameter of 1.7 cm. Fic. 2. Tuber of Marah fabaceus var. agrestis from Linden, California: weight, 58 kilograms; estimated age, fourteen years. Numerous lateral roots arise along the tuber. On large specimens, these are of sufficient size to have considerable storage function. The tap roots in plants only a few years old are approximately as long as the tuber and are still fusiform. As the plant ages, lateral roots approxi- mate or surpass the diameter and length of the tap root. Dr. Flora M. Scott (1943) has found that the transition to root structure occurs near the base of the storage region, the tuber presumably developing from the hypocotyl and stem base. The upper 20 cm. portion of a 65 cm. long tuber of M. fabaceus var. agrestis was cut off and the remainder left rooted . in the ground. Healthy shoots grew from an area on the upper edge of the rooted portion. GEOGRAPHICAL DISTRIBUTION AND INTER-RELATIONSHIPS OF THE SPECIES OF MARAH Marah oreganus has a range wider than that of any other species of the genus. It may be considered primitive in its possession of relatively large calyx lobes and of a comparatively spine-free fruit. Marah guadalupensis, a little-collected species, and one which may now be extinct because of the goats which have overrun the island of Guadalupe, has flowers and fruit very much like M. oreganus, and may be the closest relative of M. 116 MADRONO [Vol. 13 oreganus. The range of M. Watsonii is adjacent to that of M. oreganus, and its flowers, fruits, and seeds are so much like those of M. oreganus that one may well speculate that 17. watsonii is a second species which was derived from M. oreganus. Marah fabaceus is a third possible species which may have been de- rived from M. oreganus. In its geographical position and morphological characteristics it lies between each of the two pairs of species: M. ore- ganus-M. gilensis, and M. horridus-M. macrocarpus. It is interesting to note that M. fabaceus var. fabaceus has a range nearly adjacent to that of all the species except M. gilensis and M. guadalupensis. Marah macro- carpus var. major and var. micranthus represent groups which only re- cently have been geographically isolated. Marah macrocarpus var. major needs further collecting inasmuch as the Channel Islands contain many apparent intermediates between the species and its variety. SOME GENERAL ECOLOGICAL RELATIONSHIPS The author has spent many hours in the field locating, observing, col- lecting fruits and seeds of, and digging tubers of each of the species of Marah except M. guadalupensis. In addition to the results recorded here, he has experienced loss of weight, callouses, and a severe case of poison oak. From one to four tubers of each of the following species have been transplanted to a plot of sandy soil in Fresno, California: M. fabaceus var. agrestis, one tuber; M. horridus, four tubers; M. macrocar pus, three tubers; and M. watsonii, one tuber. The five largest tubers, weighing be- tween 6 and 31 kg. when dug, have grown rather well. When one tuber of M. fabaceus var. fabaceus, two tubers of M. fabaceus var. agrestis, one tuber of M. horridus, one tuber of M. macrocarpus var. macrocar pus, one tuber of M. oreganus, and three tubers of M. watsoni were planted in Stockton, California, even better results were obtained. All produced flow- ers except MW. fabaceus var. fabaceus. One plant of M. macrocarpus and two of M. watsoni produced easily distinguishable mature fruits and seeds. Marah horridus and M. oreganus also produced identifiable fruits. In no case was there any appreciable change in the floral characteristics of any species. Thus under garden conditions character differences of the reproductive organs were found to remain constant in these species. The leaves of M. horridus and M. oreganus were noticeably smaller in the sunny plots than in their original shady canyon homes. TEMPERATURE Factors. The family Cucurbitaceae is primarily one of the tropics and subtropics. The perennial genus Marah is exceptional in that it is able to prosper even at latitudes of more than 45° N. This is due to the development of deeply-buried large tubers. However, no species of Marah has been found growing higher than the highest elevations of the Upper Sonoran Life-Zone. Minimum temperatures here are lower than those of the habitat of M. oreganus in the coastal Transition Life-Zone. Marah horridus, at an elevation of 1500 m. near King’s Canyon, M. wat- sonu near Paradise, California, and M. macrocarpus var. macrocarpus 1955] STOCKING: MARAH ew near Mount Wilson all grow to within a few hundred meters of the edge of the coniferous forest, but none have been observed to grow in this forest. In each of these places, plants grow near or on a ridge where air drainage is excellent, where snows do not remain long, and where the ground does not freeze to tuber depth. MotstureE Factors. Marah oreganus is adapted to areas where the annual precipitation is more than 200 cm., while M. macrocarpus, M. gilensis, and M. fabaceus var. agrestis are adapted to many areas where there is less than 25 cm. of rain. The other species and varieties receive intermediate amounts of rain. Occasional snows fall in the higher ranges of all of these plants, especially those of M. oreganus, M. watsoni, and M. horridus. LicHuTt Factors. Individual plants of all the species of Marah except M. guadalupensis have been observed by the writer to grow well both in direct sunlight and in partial shade. The latter type of environment, es- pecially where there are shrubs or other low plants over which the Marahs twine, supports the greatest numbers of these cucurbits. The quite elon- gated stems and very large and relatively delicate leaves of many shade- dwelling plants of M. oreganus and some of M. macrocarpus var. major are striking. All species of Marah blossom by late winter or early spring. Marah wat- sonu and M. horridus have been observed to possess open blossoms as the plants push through the soil surface. Stems of MM. macrocarpus var. mac- rocarpus, which came through the surface of the soil in June, produced buds which did not blossom. These observations suggest that, photo-peri- odically speaking, all species of Marah are short-day plants. EpapuHic Factors. It was observed that all of the tubers dug were growing in well-drained, well-aerated soils: M. korridus in soils derived from decomposed granite; M. macrocarpus var. macrocarpus in the same, in gravelly soil, or in other soils on slopes; M. watsonz in friable soils; M. fabaceus in a variety of light soils; and M. oreganus in more or less light loam. No tuber of VW. gilensis was dug, but the plants observed were grow- ing in gravelly soil. After one has dug a number of tubers, he begins to associate them with friable soils, sands, gravel, and tangled masses of roots. When one observes the depth at which the tubers are buried, he can ap- preciate the place of good drainage in their survival. The tuber top of M. fabaceus var. agrestis was 17 cm. below the surface of the earth less than six months after the seed germinated. Others, measured as they were dug, were: M. horridus, elevation 1500 m. near King’s Canyon, Fresno County, California, buried 25 and 38 cm.; M. macrocarpus var. macrocarpus in the San Gabriel River wash near Arcadia, Los Angeles County, California, buried 15 and 30 cm.; and on the Angeles Crest Highway, just below the turn-off to Mount Wilson Observatory, Los Angeles County, Califor- nia, and a short distance below the Transition Life-Zone, buried 30 and 118 MADRONO [Vol. 13 38 cm. Tubers of M. watsonu, M. fabaceus, and M. oreganus were buried at comparable depths, but the exact measurements were not taken. Acid-base relationship were, as far as yet observed, not found to be critical factors. Plants of Marah oreganus and M. fabaceus var. fabaceus, natives of the somewhat acid soils of the coastal Sequoia sempervirens forest, when transplanted to a slightly alkaline soil of the San Joaquin Valley appeared to have made a satisfactory adaptation to the change. PLANT AssocraTESs. In the coastal mountains of central and northern California both M. fabaceus var. fabaceus and M. oreganus have been col- lected by the author as they were growing over Sequoia sempervirens in association with Quercus agrifolia, and Corylus rostrata var. californica. Marah fabaceus var. agrestis in the foothills around much of the Great Valley of central California is associated with Pinus sabiniana, Quercus douglasu, Quercus wislizeni, Aesculus californica, and several shrubs, especially Ceanothus spp. and Arctostaphylos spp.; M. watsonii and M. horridus share restricted parts of this range with M. fabaceus var. agrestis. M. macrocarpus var. macrocarpus is associated with Quercus agrifolia, Yucca whipplei, Rhus ovata, and many other species of woodland and shrub-land plants of southern California. At elevations of 4,500 feet or lower, M. gilensis, a species of Arizona and New Mexico, is common mostly in thickets along streams. The plant observed to be most frequently associated with most species of the genus Marah was Rhus diversiloba. OTHER PLANT RELATIONSHIPS. Individual Marah plants compete ef- effectively with grasses, other angiosperms, and gymnosperms. Since growth begins very early in the season they have a real advantage over almost all other deciduous plants. As far as has been observed, they have no natural plant parasites, except molds, and these seem to damage only plants injured in transplanting. ANIMAL RELATIONSHIPS. Very few vertebrates eat any of the vege- tative parts of Marah. The name Marah is derived from the Latin word amarus which means bitter. Most animals respond to this bitterness as man does. Damage done by western striped and twelve-spotted cucumber beetles and by squash bugs has been observed to be of little significance except to a few new shoots. A large portion of one growing tuber had been eaten by what the author, who has investigated many gopher underground networks, considered to be a very hungry gopher. Certainly some rodent had eaten it. There is no evidence that domesticated animals browse on Marah. Numerous plants of several species have been observed to prosper in various well-grazed pastures. Clearing of land and cultivation have reduced the numbers of Marah plants. However, very few agriculturists have been concerned enough with the “wild cucumbers” that grow along fence rows to try to eliminate them. A present threat of probably greater significance to their survival is that of their repeated poisoning by various herbicides. Fires in four burned-over areas studied appeared only to have eliminated competition for the spreading colonies of M. fabaceus var. agrestis and M. watsonit. 1955] STOCKING: MARAH 119 Rodents, especially the ground squirrel, Citellus beechyi, compete for the seeds of Marah. In Fresno County near Alcalde, ground squirrels were seen eating many green seeds. Undoubtedly rodents are among the most important means of seed dispersal. Gravity on steep hillsides has repeat- edly been seen to be another means of seed distribution. A number of species of small black ants have often been studied as they climbed about the flowers of M. fabaceus var. agrestis in the field. Similar ants have been observed on flowers of MM. macrocarpus var. macrocar pus and M. horridus in the field, and on all of the cultivated species and vari- eties. These ants apparently find the trichomes of the stems, peduncles, and pedicels excellent ant ladders. When observed under the microscope, the ants were found to have pollen on their bodies. A few honey bees and small beetles were observed on the flowers, but never in sufficient numbers to be of much significance as pollinating agents. ECONOMIC IMPORTANCE In the well-developed agricultural areas of California, the various spe- cies of Marah serve as breeding places and distribution points for various species of insect pests. On Marah fabaceus var. agrestis growing adjacent to several cultivated crops, the author has noted large numbers of colonies of the rapidly-reproducing western striped cucumber beetle, Diabrotica trivittata, of the twelve-spotted cucumber beetle, Diabrotica duodecimo- punctata, and the squash bug, Anasa tristis, and in his garden these same insects occurred on M. macrocarpus var. macrocarpus, M. oreganus, M. horridus, and M. fabaceus var. fabaceus. E. O. Essig (1926), says that the larvae of the striped cucumber beetle commonly attack the roots of cucumbers, melons, pumpkins, squash, and other cucurbits, eating linear ’ holes toward the bases of the plants, while the adults feed on the tops and also on beans, beets, corn, peas, sunflower, almond, apple, prune, and other plants. The larvae of the twelve-spotted cucumber beetle feed upon the roots and tubers of various plants and often do considerable damage, especially to grasses, corn, millet, oats, rye, wheat, and weeds. Adults are often serious pests feeding on many kinds of plants including fruit trees, flowers, field, forage, and truck crops. Adults and young of the squash bug do great damage to many cucurbits, and are apparently a car- rier of vine wilt disease. Because of its strong cathartic properties, a substance from the tuber of Marah fabaceus has been used in a laxative called Stroughton’s Bitters. Another substance found in Marah tubers has the property of dilating the eyes. Neither of these two substances is at present being used. — The author has speculated that since various species of Marah have stored hundreds of pounds of starch per acre while growing in such waste places as the San Gabriel wash, Los Angeles County, and the chaparral- covered slopes of many hills of California, it might sometime be a practical plan to harvest and use this starch. It should not be difficult to leach out any poisonous substances in much the same way that manioc is treated to obtain tapioca. } 120 MADRONO [Vol. 13 Seeds of Marah show promise of being a valuable source of oil. The seed’s contents will ignite when heated with a match. GENETIC CONSIDERATIONS Hybridization. Five species of Marah were cultivated in a small plot within a radius of six meters; the seeds of M. watsonu and M. macrocar pus var. macrocarpus taken from this plot showed no signs of hybridization, nor did plants grown from these seeds. Marah fabaceus var. agrestis and M. watsoni, growing within a few feet of each other near Placerville, Cali- fornia, the type locality of the latter, also show no signs of interchange of genes between species. Marah fabaceus var. fabaceus and M. oreganus grow close together in the Sequoia sempervirens forest near Pescadero, California; M. fabaceus var. agrestis and M. horridus are found in close proximity in the woodland near Bagby, California, yet each of these spe- cies has been observed to remain morphologically distinct. A study of the distribution maps of Marah (fig. 5) shows other areas shared by more than one species. The author has yet to find evidence of any interspecific hybridizing among specimens which have been collected in flower and fruit. Apparent intergrades between varieties are rather common, however, especially between M. fabaceus var. fabaceus and M. fabaceus var. agres- tis in central California and between M/. macrocarpus var. macrocar pus and M. macrocar pus var. major on the Channel Islands. Specimens of reddish-seeded races of M. macrocar pus var. macrocarpus, growing in the San Gabriel River wash near Arcadia, California, and of M. fabaceus var. agrestis at a point east of Linden, California, have been found by the author. A specimen of M. oreganus (Tracy 3532 UC) and one of M. watsonu (Heller 11812 CAS, DS, UC) likewise represent red races. CHROMOSOME CONSIDERATIONS. McKay (1931) found 32 to be the diploid chromosome number of M7. macrocarpus, M. fabaceus, and M. oreganus. Whitaker (1949) also obtained a chromosome count of 32 for M.macrocar pus. GENERIC RELATIONSHIPS The closest relatives of Marah are apparently Echino pepon and Echino- cystis. One may speculate that the genus Echinopepon is the oldest of these three genera. It ranges from central Argentina to New Mexico and Arizona, and may have given rise to the genus Echinocystis. Echinocystis, found chiefly in the northeastern parts of the United States and adjacent parts of Canada, in addition to having many floral characteristics in common with Echinope pon is, like it, an annual, with epigeous germina- tion. Marah may have been derived from Echinocystis in the eastern part of the Columbia Plateau in the Snake River region. Marah is more like Echinocystis in its irregular method of dehiscence and its seed size than it is like Echinopepon. It is unlike either Echinopepon or Echinocystis in being a perennial with large tubers and in having hypogeous germination. 1955] STOCKING: MARAH 124 In the vegetative condition, species of the three genera mentioned above, and, in addition, those of Brandegea and Vaseyanthus are often confused. They may be separated readily, however, on the basis of flowering and fruiting characters (Stocking, 1955, p. 84, key). The range of the genus Brandegea coincides with that of the northwestern part of the range of Echinope pon and the southern part of the range of Marah, that of Vasey- anthus, with the western part of the range of the genus Echinopepon, es- pecially in Lower California. TAXONOMIC CONSIDERATIONS The type specimens of some of the species of Marah were destroyed by the San Francisco fire of 1906 or in other ways. In these cases lectotypes have been suggested. All measurements recorded, except those of floral parts and stem length, were made of dry materials. Flowers of the five species of California Marah were preserved in dilute formalin; flowers of all other species were boiled in water. Where measurements of fresh and variously preserved materials were compared, no significant differences were observed. Stem length of the California species of Marah was taken in the field. Only mature seed measurements are given since the light-col- ored immature seeds of Marah are noticeably larger than the mature seeds. Specimens cited in this study were assembled in the herbarium of the University of Southern California, Los Angeles. Abbreviations used in citations are those proposed by Lanjouw (1954). In addition, KMS refers to the private herbarium of the author and USC refers to University of Southern California. Marah Kell. Proc. Calif. Acad. ser. 1 (1): 38. 1855. Type: M. muri- catus Kell. [now known as M. oreganus (Torr. & Gray) Howell, Dunn in Kew Bull. 4: 145. 1913]. Megarrhiza Torr. Pacif. R. Rep. 6, part 3, num- ber 2: 74. 1858, nomen nudum ; Pacif. R. Rep. 12, part 2, number 3: 61. 1861, based on Sicyos oreganus Torr. & Gray, Fl. N. Am. 1: 542. 1840. Echinocystis (section Marah) Cogn. in Monogr. Phan. 3: 816. 1881, based on Marah muricatus Kell. Proc. Calif. Acad. ser. 1 (1): 38. 1855. Plants sub-glabrouse to somewhat vestite, climbing or trailing, monoe- cious, with deeply striated annual stems arising from 1 to 5 “necks” on large, fusiform or globose, perennial tubers. Leaf blades suborbicular, cordate, more or less deeply 5—7-lobed or cleft; basal sulci narrow and closed to sometimes broad and opened; tendrils unifid to trifid on pe- duncles somewhat shorter than petiole length; petioles 0.5—1 times leaf diameter. Staminate flowers in racemes or panicles, axillary, tardily de- ciduous; pedicels persistent; calyx teeth small or obsolete, alternate with corolla lobes; corolla campanulate or rotate, inserted on the calyx, the surfaces, especially the upper, more or less glandular and with trichomes, usually 5— (4-8-) merous; 3 (—4) anthers fused into a cylindrical or somewhat flattened globose head; filaments a fused column. Single pis- tillate flower from same axil as staminate inflorescence and commonly 122 MADRONO [Vol. 13 larger than staminate flower; stigma discoid to sub-globose, style short or nearly obsolete; ovary (2—) 4 (—8) celled, ovules 1—4 per cell. Mature fruit a turgid capsule, globose-ellipsoidal or broadly fusiform, pendant, its spines large and numerous to small and almost lacking, irregularly de- hiscent at or near the apex, the ripe seeds falling or being ejected through a jagged opening left by splitting or dropping of the beak; dry fruit yellowish-brown, sometimes with mature seeds held by septae for several months. Seeds large; commonly brownish-grey, olive, or tan; cotyledons large; seed coat thick-walled, lignified, suberized. KEY TO THE SPECIES 1. Corollas of all mature flowers rotate or only slightly cup-shaped. 2.Corollas of all flowers slightly cup-shaped, white; fruit oblong-cylindrical, ovules more than 4; southern California and Toner California A ae te Bie ei agee 6. M. Pilar aera 2. Corollas of all fener ete ek mobos ovules 4 or fewer. 3. Staminodia present in pistillate flowers; flowers white or nearly so; Arizona, New Mexico ... 2 St 71 Miseedlensas 3.Staminodia absent in Distillate owes lowers velle wc white; California i 4. M. fabaceus 1. Corollas a all ean Hower: STRIATE, 2.Staminate flowers less than 8 mm. in diameter; ovary, fruit, and seeds globose; leaves glaucous beneath, oe dissected ; nOLEheneeenn California SMe atone 2 Staminnee ower more (hen 8 mm. in Picts ovary vem fruit elongated. 3. Calyx teeth more than 1 mm. long; linear-lanceolate; Guadalupe Island, Mexico... . 2.M. guadalupensis 3. Calyx teeth less than 1 mm. joneta delvords to eabat ce: lanceolate. 4. Ovaries and fruit ovate; spines inconspicuous; seeds discoid; humid Pacific Northwest: 4 2). circ als ape CV OLE CALS 4. Ovaries and fruit ane Baines eonenicnoue, seeds cylindrical; foothills and mountains around southeastern San Joaquin Valley . 5. M. horridus 1. MARAH OREGANUS (Torr. & Gray) Howell, Fl. NW. Am. 1: 239. 1898. Sicyos oreganus Torr. & Gray, Fl. N. Am. 1: 542. 1840. Type: ‘‘on the Oregon (Columbia River, from near its mouth to Kettle Falls: Dr. Scouler, Douglas, Mr. Tolmie,” Scouler (NY! photograph KMS! ). Megarrhiza oregona Torr. Pacif. R. Rep. 12, part 2, number 3: 61. 1861. Echinocystis oregona (Torr. & Gray) Cogn. Mem. Acad. Sci. Belg. 28: 87. 1878. Micrampelis oregona (Torr. & Gray) Greene, Pittonia 2: 129. 1890. Marah muricatus Kell. Proc. Calif. Acad. ser. 1, 1: 38. 1855. Type: “‘declivities of the hills back of the Mission Dolores, on Mr. Hutchinson’s ranch near San Francisco,” (specimen did not survive the San Francisco fire of 1906). Echinocystis (section Marah) marah Cogn. Monogr. Phan. 3: 817. 1881, based on Marah muricatus Kell. Micrampelis marah (Cogn.) Greene, Pittonia 2: 129. 1890. It might be noted that Torrey and Gray first used the specific name oreganus in 1840 and that it agrees with the masculine name Marah. 1955 | STOCKING: MARAH 123 Aerial stems from 1 the first year to 12 in older plants, 2-5 mm. in diam., 1-7 m. long, sparsely pubescent to sub-glabrous, internodes 5—15 cm. long; leaf blade sub-orbicular, 8-20 (—35) cm. in diam. with broad, rounded sinuses between the 5—7 more or less shallow lobes, the base of blade cordate, apex of terminal lobe acute, often acuminate, others obtuse or acute, the margin of blade entire, sub-undulate, sub-crenulate to den- ticulate, the surfaces sub-glabrous to sparsely pubescent; petiole (2—) 4-8 (—12) cm. long, pubescent; tendrils trifid or bifid, the peduncle heavy, 2—7 cm. long; staminate flowers in racemes, rarely in panicles, the racemes 10-30 cm. long with 5-20 flowers, seldom more than 10 mature flowers open at one time, on pedicels 5—12 (—20) mm. long, the calyx lobes broadly triangular to subulate, to 0.67 mm. long and 1 mm. wide, with conspicuous trichomes sometimes present at apices, the corolla cam- panulate, 10 mm. long, 12-15 mm. in diam., white, basal part somewhat greenish, its lobes oblong-ovate, 5-6 mm. long, 2—4 mm. wide at base, with apices obtuse, the stamens with anther head cylindrical, 2—2.5 mm. in diam., slightly longer than wide and with apex flattened, the 5 anthers inverted, U-shaped, with filament column 2 mm. long; pistillate flower with calyx lobes deltoid, subulate, or filiform, to 1 mm. long, the corolla campanulate, 15-17 mm. in diam., its lobes oblong, unequal, 8-10 mm. long, 3-4 mm. wide at base, and with apices acute, the staminodia more or less adherent to style, the stigma broadly obconic and 3 mm. in diam., the style to 3 mm. long and 2 mm. in diam.; immature fruit oblong-ovate, mature fruit ovate, attenuate at both ends, 4.5-6.5 (—8) cm. long, 3-4 (—4.5) cm. in diam., with alternating broad longitudinal, darker and lighter bands of color, the spines few, weak or almost lacking, to 6 mm. long, the peduncle 2—3 (—3.5) cm. long, the carpels 2-3, each with 1—2 (—4) ovules, these often inclined toward the apex from the horizontal; seeds discoid, surfaces slightly undulating, 16-20 mm. in diam., 8-12 mm. thick, with conspicuously ridged hilum 4-6 mm. long and 0.5—1 mm. high, dark reddish-brown, encircled with inconspicuous, slightly grooved line. Humid Pacific Coast areas from San Mateo and Santa Clara counties in California to southern Vancouver Island in British Columbia, inland especially along waterways, as far east as the Snake River valley in eastern Oregon (fig. 5). Representative specimens seen. CALirorniA. Alameda County: Oakland Hills, May, 1881, Rattan 49619 DS; Glenn County: summit of ridge north of Black Butte, July 13, 1944, Howell 19758 CAS. Humboldt County: south of point of Cape [Mendo- cino], May 20, 1933, Tracy 12227 DS, GH, UC, US. Marin County: Tamalpais, Brandegee, DS 7502; wooded hillside 3.3 miles east of Dillon Beach, March 29, 1947, Stocking 19 KMS, USC. San Mateo County: common about the hills, April 4, 1902, Baker 450 DS, GH, POM, UC, US; Sequoia sempervirens forest, 1.5 miles east of Pescadero, May 15, 1947, Stocking 40 KMS, USC. Santa Clara County: Black Moun- tain, May, 1903, Elmer 4733 ARIZ, DS, UC, US. Orecon. Benton County: northeast of Corvallis, May 22, 1917, Anderson USC; near bank of Snake River at landing, east Oregon, May 27, 1901, Cusick 2523 GH, UC, US. WasuincrTon. Klickitat County: Grand Dalles, March 26, 1934 Jones 4461 CAS. Snohomish County: Marysville, May, 124 MADRONO [Vol. 13 1928, Grant USC. BRITISH COLUMBIA. North Saanich, Vancouver Island, July 10, 1917, Newcombe UC. The race with reddish seeds collected at Eureka (Tracy 3532 UC) might well be further investigated. 2. MARAH GUADALUPENSIS (Wats.) Greene, Leafl. Bot. Obs. 2: 36. 1910. Megarrhiza guadalupensis Wats. Proc. Am. Acad. 2: 138. 1876. Type: Guadalupe Island, growing on high rocks, 1875. Palmer 33 (GH! photographs KMS! fig. 3, isotypes GH! US! ). Echinocystis guadalupensis (section Marah) guadalupensis (Wats.) Cogn. Monog. Phan. 3: 819. 1881. Micrampelis guadalupensis (Wats.) Greene, Pittonia 2: 129. 1890. Stem to 3.5 mm. in diam., glabrous to puberulent, internodes 5—20 cm. long; leaf blade broadly orbicular to ovate, (10—) 15-20 (—25) cm. in diam., broadly 5-lobed, the lobes triangular-ovate to oblong, acute or obtuse, the basal sulcus of blade 3-6 cm. deep, narrower to wider than deep, the margin of blade sinuate to denticulate or somewhat dentate, the upper surface slightly papillate-scabrous, minutely pubescent, especially on veins, the lower surface more nearly glabrous; petiole 5—8 (—10) cm. long, sub-glabrous to puberulent; tendrils bifid or trifid, the peduncle rather rigid, 3-5 (—8) cm. long; staminate flowers in raceme or panicle 10-20 (—30) cm. long with 6—12 flowers per raceme and up to 6 racemes per panicle, on pedicels 3-8 mm. long, the calyx lobes linear-subulate, 1.5-2 mm. long, 0.5—0.8 mm. broad at the base, the corolla broadly cam- panulate, 12-20 mm. in diam., its tube 4—5 mm. long, its lobes unequal, triangular-oblong, 6—7 mm. long, 3—5 mm. wide at base, sub-obtuse, with margins and inner surfaces glandular-punctate with stalked glands, the stamens with anther head sub-globose, 2.33 mm. in diam., thecae ir- regularly contorted, the filament column 1.5 mm. long, .06 mm. in diam. ; pistillate flower 18-24 mm. in diam., with calyx lobes linear, 6-7 mm. long, 0.33 mm. wide at base, the corolla open-campanulate, its tube 1.67-2 mm. long, its lobes oblong-lanceolate, 8-10 mm. long, 2-4 mm. wide at base, obtuse, the stigma thickened, discoid, 2.5 mm. in diam., 0.67—1 mm. thick, the staminodia occurring under stigma edges, subulate, 0.5 mm. long, 0.17 mm. in diam., the style 0.33—0.67 mm. long, 0.33 mm. in diam.; fruit ovoid, short beaked, 6 cm. long, 5 cm. in diam., striated from base to apex, short pubescent, its spines weak, 1—3 mm. long, 0.25—0.5 mm. in diam. at base, glabrous to puberulent; peduncle 6-8 cm. long, puberulent; seeds 2, ovoid-lenticular, 28 mm. long, 25 mm. wide, 14 mm. thick (when slightly immature), olivaceous, circumferential line incon- spicuous. Known only from Guadalupe Island off the west coast of Baja Califor- nia, Mexico, latitude 29° N., longitude 118° W. Representative specimens seen. March-June, 1897, Anthony 234 GH, POM, UC, US; March 25, 1897, Brandegee UC; winter, 1892 and 1893, Franceschi POM, UC, US; 1875, Palmer 33 GH, US. Hanna and Anthony (1923) state that goats have denuded the island. It is possible that this species has become extinct. 125 : MARAH STOCKING 1955] “Gysi) HYD Slsi ‘eg sampog ‘addy ‘sisuadnyppons "py $ (4 YI) ZOSL SC aadsapunag ‘snuns 2) am 40 YD ¢ Oly C 126 MADRONO [Vol. 13 3. MARAH WATSONII (Cogn.) Greene, Leafl. Bot. Obs. 2: 36. 1910. Echinocystis muricatus Kell. Proc. Calif. Acad. ser. 1, 1: 57. 1885, not Marah muricatus Kell., 1855. Type: vicinity of Placerville (did not sur- vive the San Francisco fire of 1906). Megarrhiza muricata ( Kell.) Wats. Proc. Am. Acad. 11: 139. 1876. Echinocystis watsonu Cogn. Monogr. Phan. 3: 819. 1881. Micrampelis watsonu Greene, Pittonia 2: 129. 1890. The following topotype represents the author’s concept of the species and may be taken as a lectotype until such time as some other specimen, historically with better claim to this status, may be located: Institute of Forest Genetics, 3 miles east of Placerville, El Dorado County, Robbins 1057 (UC! photograph KMS! fig. 4). Aerial stems from 1 the first year to 5 in older plants, 1-3 mm. in diam., 1-3 m. long, nearly glabrous with few scattered hairs, the internodes 3-12 cm. long; leaf blade orbicular, 3-8 cm. wide, almost as long, 5-cleft, the lobes heavily clavate, often further 2- or 3-lobed, apices of lobes obtuse (sometimes acute), some minutely mucronate, basal sulcus of blade 2 cm. deep, about half as wide, the margin entire or with occasional small teeth; petiole 2—5 (—7) cm. long, sub-glabrous; tendrils bifid or undivided, the peduncle 1-3 cm. long; staminate flowers 3-12 per raceme, the racemes 4-12 (-—20) cm. long, the pedicels filiform, 10-15 mm. long, the calyx lobes lanceolate, 0.5 mm. long, to 0.4 mm. broad at lower part of base, the corolla campanulate, more closed than in M. oreganus, 5-6 (—8) mm. in diam., its tube 3-4 (—6) mm. long, greenish, especially when young, its limbs triangular-ovate, to 2.5-3 mm. long, not quite as wide at the base, obtuse, glandular-papillate on inner surface, the anther head cylindrical, to 1.33 mm. in diam., to 2.5 mm. long, the filament column to 1 mm. long, 0.67 mm. in diam.; pistillate flower with calyx lobes linear-subulate, to 1 mm. long and 0.75 mm. broad at base, the corolla 8-12 mm. in diam., its tube to 4 mm. long, its lobes to 5 mm. long and 3 mm. wide at base, the apices obtuse, the staminodia oblong, to 0.67 mm. long and 0.33 mm. wide, the stigma rounded, obconic, surface rather smooth, 3 mm. in diam., somewhat less thick, the style 0.5 to 1 mm. long, slightly wider than long, the ovary globose, tapering to a tip above when young, with few broad- based prickles; fruit globose, somewhat flattened at the two poles, 2—3 cm. in diam., polar axis 0.67 to 0.75 as long as diam., the spines usually 1 or more, weak, glabrous, 1-2 mm. long, near peduncle, the dark green merid- ional lines conspicuous, the coats thin, non-rigid when dry; peduncle (2.5—) 3-4 (-4.5) mm. long; seeds globose, 11-14 mm. in diam., 2— (1-4) per mature fruit, in 2 locules, slightly flattened away from the poles, greyish brown, mottled with a reticulum of dark lines, mature seeds out- lined by an inconspicuous black band. This is the most slender and least branched species of the genus. Chaparral and woodland of the Upper Sonoran Life Zone of central and north central California (fig. 5). Representative specimens seen. Amador County: 2.1 miles northeast of Plymouth, March 21, 1947, Stocking 11 USC. Butte County: 8 miles north of Oroville, April 27, 127 : MARAH STOCKING "(9y311) OSN FT € 8U1YI0I ‘s2]Sa43D IP A snaapqvf *W * (YL) ON ZSOr suzqqoy ‘3d4}0}99] $ RS & Be SEES ER a nninennnnnnnnre - ee acer Siac SE GEE é ¢ “UOSIDRL YD ADI ‘b “OTA ‘ AOe? eM 128 MADRONO [Vol. 13 1914, Heller 11328 CAS, DS, UC, US; 0.9 mile west of Paradise on short-cut to Chico, April 17, 1947, Stocking 28 USC. Calaveras County: Angels Camp, April 11, 1923, Eastwood 11564 CAS, POM. Shasta County: Keswick, May 19, 1913, Smith 234 CAS: 4, MARAH FABACEUS (Naud.) Greene, Leafl. Bot. Obs. 2: 36. 1910. Aerial stems from 1 the first year to 20 in older plants, 3—7 m. long, 1-3 mm. in diam., slightly pubescent to sub-glabrous, internodes 7-10 cm. long; leaf blade sub-orbicular, 5-10 cm. long and wide; more or less deeply 5—7-lobed, the lobes less than one-half the leaf length, their apices acute or obtuse, occasionally mucronate, the basal sulcus of blade 1-3 cm. deep, one-half as broad to somewhat broader at opening, narrowing toward the base, or sometimes partially closed above, the blade surfaces glabrous to somewhat scabrous; petiole 3--6 cm. long; tendrils usually bifid, the peduncle 1.5—3 cm. long; staminate flowers in racemes or pan- icles 8-15 (—25) per raceme, the racemes 5—15 cm. long and panicles to 25 cm., the pedicels to 6 mm. long, the calyx teeth almost always lacking or if present, minute, the corolla rotate, (6—) 7-10 (-—13) mm. in diam., 5—8-merous, cream or greenish-white, its tube (2.5—) 3-4 (—5) mm. long, its lobes unequal, deltoid or somewhat lanceolate, 1.5-—2.5 (—3.5) mm. long, half as wide to not quite as wide at the base, with trichomes on mar- gins and inner surfaces, the anther head short-cylindrical, 1.5—2 mm. in diam., 1-1.5 mm. thick, the filament column 0.67—1 (—2) mm. long; pistillate flower with corolla rotate, 5-7 (—12)-merous, 10-15 mm. in diam., its lobes unequal, 3-5 mm. long, about one-half as wide at the base, sub-acute, the staminodia lacking, the stigma discoid, 2(—3) mm. in diam., 0.67—-1 mm. thick, sessile or nearly so, the ovary globose below, tapering to a tip; fruit globose, 4-5 cm. in diam., densely spinose, the spines rigid, 12 mm. long, 1-2 mm. in diam. at the base, the carpels 4 with usually 1 ovule per carpel; peduncle 3—5 cm. long; seeds lenticular, oblong-obovoid, 18-24 mm. long, 15-20 mm. wide, 12-15 mm. thick, the hilum ridged, 5 mm. long, brownish-tan. KEY TO VARIETIES Spines less than 5 mm. long, somewhat soft; locules with mature seeds often 2 or 3; seeds seldom flattened laterally . .. os Pa Val « CEKeSizS Spines more than 5 mm. long, rigid; locules ete rte eds usually 4; seeds com- monly flattened laterally... = sa...5,.0G¢, Soe Se a Ve eo aeeres 4a. M. FABACEUS var. fabaceus. Echinocystis fabaceus Naud. Ann. Sci. Nat. Bot. ser. 6, 12: 154. 1859. Type: Paris Botanical Garden from seeds obtained in California by Dr. Aube, Naudin. Megarrhiza californica Torr. in Wats. pro parte, Bot. Calif. 1: 241. 1876, excl. specim. Bigelow, Coco- mungo. Micrampelis fabacea (Naud.) Greene, Pittonia 2: 129. 1890. Fruits quite globose, thick walled; locules with mature seeds usually 4; spines more than 5 mm. long, rigid; seeds somewhat asymmetrical, flat- tened laterally. Near coast of central California from Marin to Monterey counties, chiefly in Coastal Transition Life Zone (fig. 5). 1955] STOCKING: MARAH 129 M. fabaceus Som var. fabaceusS ‘s » ie) e var. agrestis © intermediates M. macrocarpus @e var. macrocarpus O var. major M, Reecidus. <2 © var. micranthus Fic. 5. Distribution maps of species of Marah. 130 MADRONO [Vol. 13 Representative specimens seen. Marin County: wooded slope, Lone Pine Beach, Tomales Bay, July 25, 1932, Schreiber 662 LA. Monterey County: Whaler’s Knoll, Point Lobos State Park, Feb. 9, 1935, Lee & Mason 9213 UC. Santa Clara County: foothills west of Los Gatos, March 12, 1904, Heller 7262 GH, UC; Stanford Univer- sity, April, 1903, Elmer 4863 ARIZ, DS, POM. Santa Cruz County: 4 miles west of summit on Los Gatos-Santa Cruz highway, elev. 1,700 feet, May 11, 1947, Stocking 34 USC. There is definite intergradation between var. fabaceus and var. agrestis (fig. 5). Further collections of material bearing mature fruit will undoubt- edly bring to light even more intermediates than now known. Intermediate specimens seen. Contra Costa County: 1 mile west of Martinez, March 29, 1947, Stocking 18 KMS; Nortonville hills, March 29, 1933, Howell 10902 CAS. Santa Clara County: Milpitas, March, 1906, Smith ARIZ. Sonoma County: 1 mile west of Petaluma, March 29, 1947, Stocking 19a KMS. 4b. M. FABACEUS var. agrestis (Greene) comb. nov. Micrampelis faba- ceus var. agrestis Greene, Fl. Fran. 236. 1891. No type specimen for Greene’s variety has been located but he cites the plant as occurring on “open plains... of the valley of the San Joaquin.”’ Until some other speci- men, historically with better claim to this status may be located, the author designates the following specimen as lectotype: 114 miles east of Linden, San Joaquin County, 25 April 1947, Stocking 31 USC, (fig. 4). Echinocystis inermis Congdon, Zoe 5: 134. 1901. Type: Sherlocks, Mariposa County, Congdon (UC 131976! photograph KMS!) Marah inermis Dunn, Kew Bull. 4: 153. 1913. Echinocystis fabacea var. inermis Jepson, Fl. Calif. 2: 554. 1936. Echinocystis scabrida Eastw., Bull. Torrey Club 30: 500. 1903. Type: ... collected by Mrs. T. S. Brandegee in Fresno Co., Calif., at Zapato, [and| Chino Creek, Mar. 26, 1893, and also at Alcalde in the same re- gion.” Mrs. T. S. Brandegee (CAS! type number 354, photographs KMS! ). Similar to variety fabaceus except spines less than 5 mm. long, soft; seeds often 1-3, symmetrical, lateral edges rounded. Common in Upper and occasional in Lower Sonoran Life zones of Cali- fornia, especially in the interior; along the coast only between points in San Luis Obispo and Ventura counties (fig. 5). c¢ Representative specimens seen. Butte County: 9.6 miles from Paradise on Oro- ville road, elev. 250 feet, April 17, 1947, Stocking 29 USC. Calaveras County: 3-5 miles east of Milton, April 25, 1947, Stocking 32 USC. Fresno County: 2 miles west of Coalinga, April 2, 1947, Stocking 20 USC. Kern County: between Rosamund and Mojave, April 30, 1927 Abrams 11793 DS. Lake County: Clear Lake Park, May 16, 1938, Eastwood & Howell 5572 CAS, GH. Santa Barbara County: 5 miles south of Surf, April 14, 1929, Ferris 7579 DS, UC. Ventura County: Murietta Canyon, Ven- tura River basin, elev. 2,500—-3,000 feet, April 26, 1947, Pollard CAS. The Bigelow (Knights Ferry) collection cited by Watson (1876, p. 241) has not been located, but one can assume that it belongs in M. faba- ceus var. agrestis. The Santa Clara River valley in Ventura County, the southern side of the Tehachapi Mountains, and the Mojave Desert form both the southern 1955] STOCKING: MARAH 1s)! boundary for M. fabaceus var. agrestis and the northern boundary for M. macrocar pus var. macrocar pus. 5. MArAH HorRRIDUS (Congdon) Dunn, Kew Bull. 4: 151. 1913. Echin- ocystis horrida Congdon, Erythea 7: 184. 1900. Type: “. . . foothills of Mariposa County, California,” Congdon (UC 28933! photographs KMS! fig. 6: isotypes UC 131939! UC 131940! ). Aerial stems from one the first year to 15 on older plants, 1—4 m. long, 2—4 mm. in diam., sparsely puberulent; leaf blade orbiculate, 10—15 cm. broad, not quite as long, usually rather deeply 5- or 7-lobed, the apices of lobes acute or obtuse, the basal sulcus of blade to 4 cm. deep, much nar- rower or as broad, blade margin sinuate-dentate to sub-entire, the upper surface more or less scabrous, lower surface less so; petioles 3—8 cm. long; tendrils bifid, the peduncle (2—) 3-5 cm. long; staminate flowers in racemes or panicles, with 5—12 flowers per raceme, the racemes 7—16 (—23) cm. long, pedicels 5-15 mm. long, the calyx lobes 0-3, subulate-lanceo- late, to 0.7 mm. long and 0.33 mm. wide at base, the corolla campanulate, 10-12 (—15) mm. in diam., its tube 5—7 mm. long, its lobes as long, 3—4 mm. wide at base, obtuse, white, the anther head cylindrical, 2—2.5 mm. in length and diam., the filament column 2—2.5 mm. long, 1 mm. in diam.; pistillate flower 13-17 mm. in diam., with calyx lobes 1 mm. long, fili- form, the corolla lobes broadly lanceolate, 7-9 mm. long, 3—4 mm. wide at base, the staminodia 1—1.5 mm. long, the stigma to 3.5 mm. in diam., to 2 mm. thick, the style 1—1.5 mm. long, the ovary oblong-ellipsoidal, slight- ly pointed at each end, with spines conspicuous, the carpels 4 (—6 or —8), the ovules usually 3 per carpel; fruit oblong-ellipsoidal, 9-15 cm. long, 6—9 cm. in diam., very spinose, the spines 5—35 mm. long, 3-7 mm. wide at the base, variable lengths on one fruit; peduncle 4—6 cm. long; seeds lenticular, oblong-obovoid, 26-32 mm. long, 15—18 mm. wide, 13-15 mm. thick at thicker end, light olive, some seeds encircled by slight ridge, the ridge grooved at maturity, the hilum ridged, 5 mm. long. Upper Sonoran Life Zone, foothills of Sierra Nevada and Tehachapi Mountains, California, from Tuolumne River to Lake Hughes in Los Angeles County (fig. 5). Representative specimens seen. Fresno County: about 2 miles east of Dunlap, April 4, 1947, Stocking 22 USC; Pine Flat Dam, 1 mile west, Stocking 53 KMS. Los Angeles County: Lake Hughes, Johnstone USC. Mariposa County: 2 miles west of El Portal, elev. 2,000 feet, April 9, 1941, Rose 41105 CAS, GH. Tuolumne County: chaparral, Rawhide Hill, elev. 1,300 feet, April 11-16, 1919, Ferris 1469 DS, POM, UG US: 6. MARAH MACROCARPUS (Greene) Greene, Leafl. Bot. Obs. 2: 36. 1910. Aerial stems from 1 the first year to 10 in older plants, 1-7 m. long, 2—4 mm. in diam., deeply striated, the internodes 5—10 cm. long; leaf blade sub-orbicular, 5-30 cm. wide, not quite as long, the lobes deep, the apices acute or obtuse, surfaces of blade more or less scabrous above, his- 132 MADRONO [Vol. 13 pid beneath, the basal sulcus deep; petiole 2-10 (15) cm. long; tendrils usually bifid, sometimes trifid, the peduncle to 12 cm. long; staminate flowers in racemes or panicles, 5—15 (—25) flowers per raceme, the inflor- escences 5—20 (—40) cm. long, the pedicels 2-10 (—20) mm. long; stam- inate flower with calyx lobes 5 up to 2.2 mm. long, to 0.8 mm. broad at base, deltoid to linear-lanceolate, sometimes obsolete, the corolla cup- shaped, 3-30 mm. in diam., white, its lobes ovate to oblong-ovate (1-) 3-10 (—12) mm. long, 2-3 (—5) mm. wide at base, with apices obtuse, the anther head sub-globose, 1—-1.5 (—2) mm. in diam., the filament col- umn 0.8—1.5 mm. long; pistillate flower with calyx lobes sub-obsolete or deltoid and to 0.67 mm. long, the corolla cup-shaped, 5—24 mm. in diam., its lobes oblong-ovate, 5-8 (—10) mm. long, 3-5 mm. wide at base, with apices obtuse, the staminodia scale-like or lacking, 0.6 mm. in diam., not quite as wide on some, the stigma 2—3.33 mm. in diam., not quite as thick, the style to 0.6 (—1.33) mm. long, the ovary oblong with carpels 4; fruit cylindrical, beaked, (5—) 8-12 cm. long, (4—) 6—9 cm. in diam., densely spiny, the spines flattened, 5-30 mm. long, 1-3 mm. wide at base, with trichomes; peduncles to 4 cm. long; seeds oblong, somewhat flattened and angular-ovoid or sub-globose; 12-33 mm. long, (8—) 9-25 mm. wide, 6-14 mm. thick, olive-brown (some races reddish-brown) with a conspicuous, more or less dark equatorial line. KEY TO VARIETIES Staminate flowers 14-30 mm. in diam.; seeds 28-33 mm. long . ._ b. var. major ~ Staminate flowers less than 14 mm. in diam.; seeds less than 21 mm. long. Staminate flowers 8-13 mm. in diam.; seeds 15-20 mm. long . a. var. macrocar pus Staminate flowers 3-6 mm. in diam.; seeds 10-13 mm. long . c. var. micranthus 6a. M. MACROCARPUS var. macrocarpus. Megarrhiza californica Torr. in Watson, Bot. Calif. 1: 241. 1876, as to the Bigelow, Cocomungo [Cuca- monga, San Bernardino County?| specimen. Echinocystis macrocarpa Greene, Bull. Calif. Acad. 1: 188. 1885. Micrampelis leptocarpa Greene, Pittonia 2: 282. 1892. Type: ‘“‘the Colorado Desert, in southern California,” W. G. Wright. (None of the original specimens from which Greene drew his descriptions have been positively located. This is a very questionable entity.) Marah leptocarpa (Greene) Greene, Leafl. Bot. Obs. 2: 36. 1910. Echinocystis macrocarpa var. leptocarpa (Greene) Wright, Muhlenbergia 3: 125. 1907. The following topotype represents the author’s approximate concept of the species and may be taken as a lectotype until such time as some other specimen, historically with better claim to this status, may be located: vicinity of San Bernardino, California, S. B. Parish 3633 (US! photo- graphs KMS! fig. 6, isotype GH! ). Leaf blade 5-10 cm. wide, petiole 2—7 cm. long; tendrils to 5 cm. long; staminate racemes 5-15 cm. long, pedicels 4-10 mm. long; staminate flower 9-12 mm. in diam., the calyx lobes obsolete to short subulate or deltoid, the corolla lobes 3-4 (—5) mm. long, the anther head diam. 133 : MARAH STOCKING 1955 | (WY4stU) SQ FEgE PARK ty EDX LAB <2S% ysiavg ‘9dA4}0439] ‘snd ADIOAIDUM IPA SA GADIOAIDU * PT £ ¢ G Ja) ©6872 O eS EE og ee e Q “vopsuo0Z ‘od Ay ‘snp A AOY YDAD IY 0 S| 134 MADRONO [Vol. 13 0.8-1.25 mm., filament column 0.6—1.25 mm. long; pistillate flower 15—20 mm. in diam. with calyx lobes obsolete or very reduced, the corolla lobes 5—7 mm. long, the staminodia lacking to 0.4 mm. long, the stigma 2-3 mm. in diam., the style to 0.6 mm. long; fruit 8-10 cm. long, 6—9 cm. in diam.; seeds 15-20 mm. long, 12—18 mm. wide, 11—14 mm. thick, brown- ish tan. Southern California, Northern District of Baja California, Mexico, and coastal islands to the west (fig. 5). Representative specimens seen. CALIFORNIA. Los Angeles County: 2 miles north of Claremont, March 3, 1932, Wheeler 455 DS. Ventura County: Foothill Trail, Ven- tura River basin, March 25, 1945, Pollard CAS. Mexico. Baja CatirorniA. Northern District: Cabo Punta Banda, south arm of Todos Santos Bay, Feb. 24, 1930, Wiggins 4231 ARIZ, CAS, DS, GH, US; 2-5 miles north of Hamilton Ranch, Santo Domingo, March 2, 1930, Wiggins 4296 CAS, LA, POM, US; Santa Catarina Landing, March 10, 1930, Wiggins 4436 LA, POM, UC, US. Intergradation between var. macrocarpus and var. major is suggested by several specimens from the Channel Islands. Collections have not been extensive enough to show complete intergradation clearly. In Baja Cali- fornia var. macrocarpus may intergrade with var micranthus (Wiggins 4436). It is quite interesting to note that a line along the Santa Clara River in Ventura County, through Lake Hughes in Los Angeles County and thence eastward marks not only the northern limit of M. macrocarpus, but also the southern limits of both M. fabaceus var. agrestis and M. horridus. 6b. M. MACROCARPUS var. Major (Dunn) comb. nov. M. major Dunn, Kew Bull. 4: 151. 1913. Type: “S. California Islands. San[ta] Catalina, San Clementi[e], San Nicholas ... common in moist canyons .. . white flowers . . . roots as large as small barrels,” Trask 281 (US! photographs KMS!, fig. 7; isotype US! ). Leaf blades (10—) 15-25 (—30) cm. in diam.; staminate flowers (14—) 16-30 mm. in diam., the calyx lobes linear-lanceolate, 2—2.2 mm. long, 0.6-0.8 mm. broad at the base; pistillate flowers 20-24 mm. in diam., the calyx lobes deltoid, 0.67 mm. along each side; seeds (22—) 28-33 mm. long, (17—) 21-25 mm. wide, 12-14 mm. thick; other vegetative and reproductive parts of this variety correspondingly large. On all of the larger Channel Islands of California from San Miguel in the north to San Clemente in the south (fig. 5). Representative specimens seen. San Clemente Island, Middle Ranch, Feb. 16, 1941, Moran 573 DS; shaded slope, east coast, April 11, 1923, Munz 6779 GH. San Miguel Island: April 10, 1930, Munz & Crow 11,810 POM. San Nicholas Island: April, 1901, Trask US. Santa Catalina Island: Avalon, March, 1901, Trask US. Santa Cruz Island: Pelican Bay, April 26, 1930, Abrams & Wiggins 64 CAS, GH. Santa Rosa Island: June, 1888, Brandegee UC. 6c. M. MACROCARPUS var. micranthus (Dunn) comb. nov. M. micran- thus Dunn, Kew Bull. 4: 150. 1913. Type: Cedros Island, Rose 16159 (US! photographs KMS! fig. 7). 135 : MARAH STOCKING 55 “(YSI) SQ OSI9L as0¥ . ‘ ad Ay é SHY FUDAIIUM IBA SNGADIOAIDU * PY ‘ (391) SA 182 yso4Z adAy 6 40(pu “eA sng svIO : AIDUL YDAD PY L orgy 136 MADRONO [Vol. 13 SRA A Re SMNGIAHOM LABEL Revision af Echinseygotix, Marah, Eebinapcpan Merah Bilersis (Greene) Creene KM. Steckin 3949 Fic. 8. Marah gilensis, type, Greene ND. Staminate flowers 3-6 (—8) mm. in diam.; pedicels 2—4 mm. long; fruit 3-5 (-6) cm. long; seeds 12-13 mm. long, 9-10 mm. wide, 6 mm. thick; other dimensions in proportion. Known only by the two following collections from Cedros Island, off the west coast of Baja California, Mexico, 28° N.: Rose 16159, March 12, 1911: US; and Brandegee, April, 1897, UC: 102131,)102133) 102134 (fig. 5). 7. MARAH GILENSIS (Greene) Greene, Leafl. Bot. Obs. 2: 36. 1910. Megarrhiza gilensis, Greene, Bull. Torrey Club 8: 97. 1881. Type: “the canyon of the Upper Gila in Grant County, New Mexico, within fifteen miles of the dividing ridge between the Pacific and Atlantic slopes.” . . . 1955] STOCKING: MARAH U3 Canon of the Gila, April 30 [20?], 1881, Edward L. Greene (photographs ND! KMS! fig. 8; possible isotypes GH! photographs ND! KMS!). Echinocystis gilensis Greene, Bull. Calif. Acad. 1: 189. 1885. Micram- pelis gilensis Britt. Trans. N. Y. Acad. 8: 67. 1889. Stem slender, 2(—3) mm. in diam., puberulent to sub-glabrous, inter- nodes 5—15 cm. long; leaf blade orbicular, 4-10 cm. in diam., deeply 5—7- lobed, the lobes to 5—7 times as long as broad, central apex acute, others acute to obtuse; basal sulcus of blade to 25 mm. deep., 0.33-0.5 as broad, the margins entire or sometimes irregularly and broadly dentate, upper surface scabrous, lower slightly so; petioles 2-4 cm. long; tendrils unbranched or bifid, peduncle 2—2.5 (—3) cm. long; staminate flowers in racemes, seldom in panicles, 10—20 flowers per raceme, the racemes 15-25 cm. long, the pedicels 2-5 mm. long, the calyx lobes almost obsolete, the corolla rotate, 6-10 (—11) mm. in diam., the corolla tube 2—4 mm. in diam., its lobes deltoid to lanceolate, 3—3.5 mm. long, 2-3 mm. wide at the base, obtuse or acute, the margins and inner surfaces glandular-punctate, the anther head somewhat triangularr in cross-section, 1—1.6 mm. in diam., about one-half as thick, filament column 1 mm. long, slender; pistillate corolla rotate, 10-12 mm. in diam., the lobes 3-4 mm. long, 1—-1.5 mm. wide at base, with apices obtuse or acute, the staminodia very inconspicu- ous, less than 1 mm. long, the stigma 2—2.5 mm. in diam., 1—1.25 mm. thick, the style 0.5—0.8 mm. in diam., not quite as long, the ovary globose with carpels and ovules usually 4, spines conspicuous; fruit globose with a rather persistent beaked apex, 26-32 mm. in diam., the spines rather dense, stout, 2-3 (—5) mm. long; peduncle 8-12 mm. long; seeds ovoid- lenticular, 14-16 mm. long, 10-12 mm. wide, 7-10 mm. thick, dark brownish-olive, encircled by a grooved ridge, bottom of groove dark. Arizona and southwestern New Mexico (fig. 5). Representative specimens seen. Ar1IzonA. Mojave County: Chloride, elev. 4,500 feet, April 14, 1903, Jones DS, POM, US. Pima County: Ajo Mountains, Ajo, April 19, 1942, Cooper 636 AHFH; shallow gorge, Coyote Mountains, elev. 4,000 feet, Feb. 4, 1945, Phillips 2597 ARIZ, CAS, GH, US. Yavapai County: Congress Junction, elev. 3,000 feet, May 2, 1903, Jones CAS, DS, POM, US. New Mexico. Grant County: Burro Mountains, May, 1880, Rusby 141 UC, US. College of the Pacific, Stockton, California LITERATURE CITED Essic, E. O. 1926. Insects of western North America. New York. Hanna, G. D. and A. W. AntHony. 1923. A cruise among desert islands. National Geographic Magazine 44: 70-99. McKay, J. W. 1931. Chromosome studies in Cucurbitaceae. Univ. Calif. Publ. Bot. 16: 339-350. Science News LETTER. 1948. 53: 103. [Photograph of Marah tuber. ] Scott, F. M. 1943. Survey of anatomy, ergastic substances, and nuclear size in Echinocystis macrocarpa and Cucurbita Pepo. Bot. Gaz. 104: 394-408. STOCKING, K. M. 1955. Some considerations of the genera Echinocystis and Echino- pepon in the United States and northern Mexico. Madrono 13: 84-100. 1955. Watson, SERENO. 1876. Botany of California, Volume 1. WuHuitAke_r, T. W. 1949. Polyploidy in Echinocystis. Unpublished paper. 138 MADRONO [Vol. 13 OBSERVATIONS ON PRASIOLA MEXICANA, A FRESHWATER ALGA OF UNKNOWN RELATIONSHIPS HERBERT F. COPELAND The observations upon Prastola mexicana J. Agardh which are here presented were made in the expectation that they would clarify the rela- tionships of this organism. This expectation has not been fulfilled. Prasiola mexicana produces thin green thalli, of small to moderate size, on stones in mountain streams. In conventional classification it is placed in or near the family Ulvaceae, to which plants, however, it is not closely similar in microscopic structure. The cells of Praszola (their dimensions are of the order of 5 to 10y.) become separated in a colorless matrix. They divide in three planes. A cross section of the thallus shows ordinarily two layers of octettes of cells. Continuing division may produce columns of several cells. These features of the structure are shown in the illustrations of Setchell and Gardner (1920). In living material the protoplast ap- pears to consist principally of a single bright green plastid lying in the middle of the cell and connected to the cell membrane by radiating strands (fig. 1, a). Setchell and Gardner suggested the possibility that Prasiola may not belong naturally with the green algae, but with the prim- itive red algae. According to Kylin (1930), however, the pigmentation is that of green algae and higher plants. The material studied was collected in Butte Creek at Jonesville, Butte County, California, at an altitude of about 1500 m. Since nuclear and cell division occur in many organisms at particular times of the day, material was fixed at intervals of from one to two hours during one day and night. Portions of each collection were fixed respectively in iron acetocarmine and in FAA. A part of the material in FAA was stained as whole mounts with Heidenhain’s haematoxylin; another part was dehydrated, imbed- ded, and sectioned at 10y.. Some of the sections were stained with Heiden- hain’s haematoxylin, and others with other stains, including basic fuchsin applied after several minutes exposure to warm normal hydrochloric acid. No difference was observed in the condition of the cells fixed at differ- ent times; in particular, an abundance of newly divided cells was present in every collection. The sectioned material stained with Heidenhain’s haematoxylin showed the internal structure most clearly: other tech- niques gave the same results as this, or else failed to show the internal structure. As protoplasts stained in Heidenhain’s haematoxylin are destained in iron alum, four concentric parts become distinguishable. In order from outside to inside, these are as follows (fig. 1, b, c, d). 1. A narrow superficial layer is destained promptly. 2. Within the foregoing there is a thicker layer which is more resistant to destaining. After very brief destaining it appears as a solid black mass 1955] COPELAND: PRASIOLA 139 Fic. 1. Prasiola mexicana. a, living cells in surface view; ), c, d, stained cells, b, c, in surface view, d, in a cross section of the thallus; all X 1000. nearly filling the cell. Further destaining decolorizes it gradually from without inward; eventually, it becomes indistinguishable from the super- ficial layer. This body is believed to be the plastid. 3. Inside of the plastid there is a fairly extensive area (here called the “clear space’’) which is more promptly destained. No definite membrane has been seen at the boundary of the clear space. 4. In the middle of the clear space there is a granule (here called the “central granule”) about 0.5y. in diameter. It is the most resistant to de- staining of all parts of the cell. It is‘sometimes slightly irregular in out- line. In some preparations one can see vague strands radiating from it. No internal structure has been seen. At least one clear space and one central granule are present in every cell. In many cells, the central granule is elongate. Rarely, two central granules can be found in one clear space; usually, if a cell contains two or more central granules, even quite close together, they lie in separate clear spaces. It appears that the central granule divides by constriction and that the clear space divides very promptly after it. Constriction of the protoplast is less prompt, and may be delayed until four or more clear spaces and central granules are present. According to these observations, cells of Prasiola mexicana have no nucleus in the proper sense of the word: no structure limited by a mem- brane, and whose division involves the appearance and division of def- inite chromosomes, was seen. Nothing was seen of the nature of the parietal granule, apparently a primitive nucleus, which has been de- scribed in the lower red algae, in Porphyridium by Lewis and Zirkle (1920), in Porphyra by Ishikawa (1921), and in Bangia by Dangeard (1927). The structure is not that of the blue-green algae. Except for the plastid, it is very much like bacterial structure as it is currently under- stood (Robinow, 1942, 1949; Tulasne and Vendrely, 1947; Hillier, Mudd, 140 MADRONO [Vol. 13 and Smith, 1949). The relationships of Prastola remain altogether ob- scure. Sacramento Junior College, Sacramento, California. LITERATURE CITED DANGEARD, P. 1927. Recherches sur les Bangia et les Porphyra. Le Botaniste 18: 183-244. HIvuier, J.,S. Mupp, and A. G. SmitH. 1949. Internal structure and nuclei in cells of Escherichia coli as shown by improved electron microscopic techniques. Jour. Bact. 57: 319-338. Isuikawa, M. 1921. Cytological studies on Porphyra tenera Kjellm. I. Bot. Mag. Tokyo 35: 206-218. Kyun, H. 1930. Some physiological remarks on the relationship of the Bangiales. Bot. Notiser 1930: 417-420. Lewis, I. F. and C. Zirke. 1920. Cytology and systematic position of Porphyridium cruentum Naegeli. American Jour. Bot. 7: 333-340. Rosinow, C. F. 1942. A study of the nuclear apparatus in bacteria. Proc. Roy. Soc. Bot. 130: 299-324. . 1949. Cytological observations on bacteria. Proc. Sixth Internat. Congr. Exp. Cytol. 204-207. SETCHELL, W. A. and N. L. Garpner. 1920. Phycological contributions I. Univ. Cali- fornia Publ. Bot. 7: 279-324. TuLAsNeE. R., and R. VENpDRELY. 1947. Demonstration of bacterial nuclei with ribo- nuclease. Nature 160: 225-226. A NEW SPECIES OF BOUVARDIA (RUBIACEAE) FROM BAJA CALIFORNIA, MEXICO ANNETTA CARTER During the past sixty or seventy years the flora of Baja California, Mexico, has become reasonably well known in the vicinity of anchorages along the Pacific Ocean and Gulf of California, in the mountains of the north and some of those of the Cape Region, and along “‘E] Camino Na- cional,” as the main route of travel the length of the peninsula is now officially known. Whenever a botanist manages to get off the beaten track, however, he is quite apt to find plants of special interest. Arroyo del Salto, which empties into the Gulf east of La Paz near Las Cruces in the Cape Region, is such a place. Here, where a granite dike forms a dam across the upper reaches of the deep and narrow canyon, is an oasis of tall palms (Erythea Brandegeei Purpus) and an abundance of such mois- ture-loving herbaceous plants as Samolus ebracteatus H.B.K., Bacopa Monnieri (L.) Pennell, and Cyperus. On the drier talus slopes and cliffs, in addition to the usual Lysiloma candida Brandegee, Jatropha cinerea (Orteg.) Muell., Cyrtocarpa edulis (Brandegee) Standl., and Fouquieria peninsularis Nash, I found the Bouvardia described below as well as the 1955] CARTER: BOUVARDIA 141 Fic. 1. Inflorescences of Bouvardia Alexanderae: upper, valvate buds; lower, from plant having fiowers with styles exserted and stamens included. All x 2. 142 MADRONO [Vol. 13 delicate annual, Drymaria debilis Brandegee, a species not previously re- ported from south of La Purisima (ca. latitude 26°N.), and Dudleya nubigena (Brandegee) Br. & Rose, known before only from the type lo- cality in the Sierra de la Laguna. Bouvardia Alexanderae sp. nov. Planta perennis lignosa scabrida foliis oppositis vel plerumque ternatis lineari-lanceolatis apice acutis 15-45 mm. longis 3-10 mm. latis floribus in cymis terminalibus 5—12-floris dis- positis hypanthio obconico 2—3 mm. longo 3 mm. lato calycis lobis lance- olato-linearibus acutis crassis patentibus 4-5 mm. longis corolla hypo- craterimorpha alba tubo gracili 2.5—3.5 cm. longo ejus lobis oblongis plus minusve acutis 8-10 mm. longis 4—5 mm. latis staminibus inclusis vel exsertis heterostylis stylo exili stigmate bifido 4-5 mm. longo ovario biloculari placentis peltatis crassis prope septi basem affixis ovulis nu- merosis ad perpendiculum dispositis capsula globosa fissura septo ad perpendiculum dehiscente seminibus exalatis angulatis diametro 1 mm. vel minoribus. Type. Steep granite talus, Arroyo del Salto (latitude 24°12’N., long. 110°7.5’W.), east of La Paz, Baja California, Mexico, March 30, 1949, Annetta Carter 2577 (UC 985926). Woody perennial 3-6 dm. tall, stems terete, scaberulous, epidermis exfoliating on older parts, the internodes 10-25 mm. long; leaves 15-45 mm. long, 3-10 mm. wide, opposite, or occasionally in whorls of three, linear-lanceolate, acute at apex, sessile or gradually tapering to a short petiole, glabrous or slightly scaberulous on margins and midrib, the vena- tion, except for the midrib, obscure; stipule sheath hyaline-membranous bearing several slender, hyaline teeth up to 2 (-—3) mm. long; flowers in a terminal 5—12-flowered cyme on pedicels up to 1 cm. long, the pedicels with a hyaline, toothed scale at the base and occasionally below the hypanthium; hypanthium obconic, slightly quadrangular, 2-3 mm. long, 3 mm. broad, glabrous or slightly scaberulous; calyx lobes 4, lance-linear, acute, 4-5 mm. long, 1—1.5 mm. wide near base, 0.75—1 mm. thick, spread- ing, glabrous or slightly scaberulous, the sinus between the calyx lobes bearing one or two short hyaline teeth; flowers fragrant; corolla salver- form, white (fading to pale rose) with a tint of rose dorsally on the mid- vein of the lobe, the tube slender, 2.5—3.5 cm. long, throat ca. 2 mm. in diameter, the lobes oblong, acutish, 8-10 mm. long, 4-5 mm. broad; stamens borne either at mouth of throat and exserted or at the base of the throat where they are included and nearly sessile, the filaments of the exserted stamens 1.5—2.5 mm. long; anthers 2-3 mm. long, versatile, yellow or black; heterostylic, style slender, stigma 4—5 mm. long, bifid; ovary inferior, two-celled, placenta swollen, peltate, attached near base of septum and bearing numerous crowded and vertically placed ovules; cap- sule globose, slightly two-lobed, 4 mm. broad and as high, dehiscing by an apical slit at right angles to the septum; seeds up to 1 mm. in diameter, angular, finely reticulate, wingless. 1955] CARTER: BOUVARDIA 143 Fic. 2. Bouvardia Alexanderae: a, habit, * %; b, seeds, * 153; c, dehisced cap- sule, X 5; d and e, corollas split longitudinally to show heterostyly, x 1%; f, calyx (showing thick spreading sepals with hyaline teeth in the sinuses), young capsule, and hyaline tooth on pedicel, X 2. Drawings by Emily Patterson Reid. 144 MADRONO [Vol. 13 Because of its wingless seeds, B. Alexanderae keys to the genus Hous- tonia where it is not at all at home; in all other characters it fits into Bouvardtia, where it falls in the section of the genus having large, white, salverform corollas with long tubes. Its 5—12-flowered compact cymes readily distinguish it from all of the closely related species: B. induta (Robinson) Standley, B. Langlasse: Standley, B. erecta (DC.) Standley, B. Karwinskyi Standley, B. glabra Polak, B. latifolia Standley, and B. longiflora (Cav.) HBK. In addition, the linear-lanceolate leaves distin- guish B. Alexanderae from all of its close relatives except B. Karwinskyi and B. erecta. Of these, B. Karwinskyi is described as a shrub 4—5 feet tall with long (15-26 mm.), linear calyx lobes and narrow, seemingly whorled leaves, while B. erecta is a divaricately, often rigidly branched, low shrub with 1—3-flowered cymes and winged seeds. Bouvardia glabra and B. longiflora also have winged seeds. This character has not been as- certained for the other above-mentioned species because of the lack of fruiting material. No other species of Bouvardia are known to occur in Baja California, nor have the above-mentioned, closely related species been collected in the states of Sonora and Sinaloa on the opposite shore of the Gulf of Cali- fornia. Bouvardia Alexanderae is named in memory of Miss Annie M. Alexan- der, who in 1947 invited me to accompany her and Miss Louise Kellogg on an expedition covering the length of the peninsula, thus initiating my field work of successive years in Baja California. She assisted generously in the financing of the 1949 trip during the course of which I collected this plant. In 1947 Miss Alexander, a keen and painstaking collector and an inspiring field companion, contributed her full share of work to the expedition and, although in her eightieth year, endured with cheerful equanimity the rigors of three months of rough travel and camping in the peninsula. For the loan of herbarium specimens I wish to express my appreciation to the curators of the United States National Museum and the Chicago Natural History Museum. For fresh flowering material which served as a basis for the illustrations and certain details of the description, I am in- debted to Mr. Frank F. Gander of Escondido, California, who, in his na- tive plant nursery, grew plants from the seed that I provided. Mr. Gander (correspondence) reported that the plants bloomed profusely the first year, thrived in the hot, dry summer climate, withstood winter tempera- tures as low as 26°F., exhibited a tolerance to “hard” water, and did not seem to be attractive to rabbits. Because of its abundance of fragrant white flowers, its long blooming period, and its hardiness, Bouvardia Alexanderae may well prove to be a desirable garden plant in the South- west. Department of Botany, University of California, Berkeley. INFORMATION FOR CONTRIBUTORS Manuscripts submitted for publication should not exceed 20 pages when printed, or contain more than 20 per cent illustrative material including tabular matter, unless the author agree to bear the additional cost. Subject to the approval of the Editorial Board, manuscripts may be published ahead of schedule as additional pages to the issue, provided the author assume the complete cost of publication. Shorter items, such as range extensions and other biological notes, will be published in condensed form with a suitable title under the general heading, ‘“‘Notes and News.” Institutional abbreviations in specimen citations should follow Lanjouw and Stafleu’s list (Index Herbariorum. Part 1. The Herbaria of the World. Utrecht. Second Edition, 1954). Articles may be submitted to any member of the Editorial Board. MADRONO A WEST AMERICAN JOURNAL OF BOTANY A quarterly journal devoted to the publication of botanical re- search, observation, and history. Back volumes may be obtained from the Secretary at the following rates: Volume I, 1916-1929 ss) oc 2, og 1 $5200 Volume IT, 1930-1934. . . . . 5.00 Volume TIT, 1935-1936 . . ... % 5.00 Volume IV, 1937-1938 . . . . 5.00 Volume V, 1939219405"). y, SUSce Volume VI, 1941-1942 . . . . 5.00 Volume VII, 1943-1944 . . .. . 5.00 Volume VITI, 1945-1946 . . . . 5.00 Volume IX, 1947-1948 . ere 2 6.00 Volume X, 1949-1950. . . .... 7.00 Volume XJ, 1951-1952. . tt 8.00 Volume XII, 1953-1954 . . . . 8.00 Single numbers, Volumes land 2) 9; 9. 9 2) = 50 Succeeding numbers, | (except Volume XI, No.2)... 1.00 Volumesxa NO. 20345" ae ae 1.50 The subscription price of MApRONo is $4.00 per year. If your institution does not now subscribe to MADRONO, we would be grateful if you would make the necessary request. Since there is no extra charge for institutional subscriptions, an individual may hold membership in the California Botanical Society on the basis of his institution’s subscription. Address all orders to: G. THomMAS RossBins, Corresponding Secretary Department of Botany University of California, Berkeley 4, California ADRONO VOLUME 13, NUMBER 5 JANUARY, 1956 Contents BIOSYSTEMATICS OF HELIANTHUS DEBILIS, Charles B. Heiser, Jr. 145 A NEw TRIFOLIUM FROM OREGON, Helen M. Gilkey 167 A New REcorD oF DwarF MISTLETOE ON LODGEPOLE AND WESTERN WHITE PINE, Job Kuijt 170 PROSOPIS GLOBOSA GILL. IN BAJA CALIFORNIA, Robert L. Dressler “TZ Review: W. C. Muenscher, Weeds (Helen K. Sharsmith) 175 Notes AND NEws: EUPHORBIA ANTISYPHILITICA DIS- COVERED IN SOUTHERN NEw Mexico, Philip J. Ley- endecker and Cecil A. Kennedy 176 A WEST AMERICAN JOURNAL OF BOTANY Ce MADRONO A WEST AMERICAN JOURNAL OF BOTANY Entered as second-class matter at the post office at Berkeley, California, January 29, 1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price $4.00 per year. Published quarterly and issued from the office of Madrofio, Herbarium, Life Sciences Building, University of California, Berkeley 4, California. BOARD OF EDITORS Hersert L. Mason, University of California, Berkeley, Chairman Epcar ANDERSON, Missouri Botanical Garden, St. Louis. LyMAN BENSON, Pomona College, Claremont, California. HERBERT F. CoPELAND, Sacramento College, Sacramento, California. Joun F. Davinson, University of Nebraska, Lincoln. Ivan M. Jounston, Arnold Arboretum, Jamaica Plain, Massachusetts. MitpreD E. MATHIAS, University of California, Los Angeles 24. Marion Ownsey, State College of Washington, Pullman. IrA L. Wiccrns, Stanford University, Stanford, California. Secretary, Editorial Board — ANNETTA CARTER Department of Botany, University of California, Berkeley. Business Manager and Treasurer — RicHARD W. Ho_m Natural History Museum, Stanford University, Stanford, California. CALIFORNIA BOTANICAL SOCIETY, INC. President: Lincoln Constance, Department of Botany, University of California, Berkeley, California. First Vice-president: John Thomas Howell, California Academy of Sciences, San Francisco, California. Second Vice-president: Mildred E. Mathias, Department of Botany, University of California, Los Angeles, California. Recording Secretary: Mary L. Bowerman, Department of Botany, University of California, Berkeley, California. Corresponding Secretary: G. Thomas Robbins, Department of Botany, University of California, Berkeley, California. Treasurer: Richard W. Holm, Natural History Museum, Stanford University, Stanford, California. Annual membership dues of the California Botanical Society are $4.00 for one person or $5.00 for two members of the same family. Other classes of membership in the Society are: Life ($100), Sustaining ($25 annually), and Honorary (elective). All types of membership carry with them one subscription to Madrofio and all other privileges of the Society. Dues should be remitted to the Treasurer. General corre- spondence, changes of address, and applications for membership should be addressed to the Secretary. 1956] HEISER: HELIANTHUS 145 BIOSYSTEMATICS OF HELIANTHUS DEBILIS! CHARLES B. HEISER, JR. A study of the gulf coast sunflower, Helianthus debilis, was begun in 1946, and field work was carried on in both Florida and Texas. From the analysis of population samples and the subsequent garden cultivation of plants, it was possible to recognize nine morphological races. A systematic series of hybridizations between the races, as well as some hybridizations with other sunflower species, was carried out. Material was borrowed from several herbaria in order to supplement the author’s collections. The first part of the present paper summarizes the results of the cytogenetic study, the second part deals with the taxonomy, and the final part comprises a discussion of subspeciation in these sunflowers. HYBRIDIZATIONS 1. Intraspecific crosses. Forty-eight reciprocal crosses were made in- volving the races listed in Table 1. The F cross was readily secured and the resulting progeny were vigorous. In general, twenty plants of each re- ciprocal combination were grown. As can be seen from the polygonal crossing charts (fig. 1), two distinct groups can be recognized with re- spect to the fertility of the F, hybrids. Hybrids between any two of the first six races (the ‘‘debilis” assemblage) and hybrids between any two of the last three races (the ‘“‘praecox” assemblage) as a rule give highly fer- tile F; plants with around 90 per cent pollen fertility and nearly 100 per cent good seed set. Occasionally, plants with somewhat reduced fertility were encountered both in hybrids within races and between races, but their occurrence followed no consistent pattern. In general, it can be stated, then, that crosses within either the ‘“‘debilis” or “praecox”’ assemblage of races produce fertile hybrids. However, any crosses between races of these two assemblages gave F, plants with pollen fertilities from 20 to 50 per cent, averaging around 30 per cent, and the seed set was one third to one half the normal amount. Only a few F. generations have been grown, and it has been found that those involving crosses within either assemblage give vigorous fertile plants, whereas those between the two assemblages give plants with fertilities varying from 30 to 100 per cent. In the second I’. group, although the majority of the hybrids were vigorous, some weak- ness was observed in a few of the plants. 2. Interspecific crosses. Representatives of the “debilis” and ‘“praecox” assemblages (Table 1) discussed above were crossed with several other 1 The present work has been supported in part by a grant from the National Science Foundation. A grant for travel in Florida was awarded by the Society of Sigma Xi. Mrs. Barbara Norby and Mr. Dale M. Smith, graduate students at Indiana University, have served as most faithful assistants. I would also like to thank Dr. Charlotte Avers, Dr. T. R. Fisher, Dr. Jasper Garner, and Miss Rose Overstreet who made special collections for me in certain areas of Texas and Florida. Maprono, Vol. 13, No. 5, pp. 145-176, January 25, 1956. 146 MADRONO [Vol. 13 silvestris tordiflorus cucumerifolius vestitus weed race debilis praecox hirtus runyonii silvestris tardiflorus cucumerifollus \ \ Vat ya) vestitus : , \ \ weed race a oS us | \ debilis me i SS a ce M ae praecox hirtus Cc runy onii annuus oa tof +E me "debitis" x | od oh| x oe a “praecox" argophyllus Py eee ca petiolaris F, fertile Paglia ent F, with reduced fertility cet ta ate F, with highly reduced fertility Fic. 1. Results of crosses: top, crosses of races of Helianthus debilis giving fer- tile Fy hybrids; center, crosses of races of H. debilis giving F ’s with reduction in fertility ; bottom, crosses involving other species of annual sunflowers. Further ex- planation in text. 1956] HEISER: HELIANTHUS 147 species, and the results obtained are shown in fig. 1 (bottom). Other spe- cies crosses previously reported (Heiser, 1951) are also shown. Most in- terspecific crosses result in rather highly sterile F,; hybrids with pollen fertilities generally below 20 per cent and seed set of generally less than 10 per cent. The exceptions are those hybrids between H. annuus and H. argophyllus, H. petiolaris and “‘praecox,” and the previously mentioned hybrid between ‘‘debilis” and ‘“‘praecox.” Both H. annuus and H. argo- phyllus are partially sympatric with H. debilis, and natural hybrids be- tween H. annuus and the latter have been found, but no natural hybridi- zation is known between H. argophyllus and H. delibis. Helianthus bo- landeri of California and Oregon does not particularly concern us in the present discussion, but is included to show that the taxa recognized as species in this group generally show considerable reduction of fertility in the F, hybrids. Helianthus petiolaris, on the other hand, and H. debzilis are vicarious species, with H. petiolaris replacing H. debilis to the west (fig. 3). Morphologically the “‘praecox”’ assemblage appears to be inter- mediate between H. petiolaris and the “‘debilis” assemblage. Crosses be- tween H. petiolaris and two representatives of the “‘debilis” assemblage give F, plants which fall into the highly sterile category of 20 per cent or less pollen fertility whereas crosses between H. petiolaris and ‘“‘praecox”’ give F, hybrids with about 40 per cent pollen fertility (fig. 1). Thus it can be seen that the “praecox” assemblage, which appears to be intermediate morphologically between H. petiolaris and “‘debilis,” gives moderately fertile hybrids when crossed with these other taxa. On the other hand, hybrids between “‘debilis” and H. petzolaris show a rather great reduction in fertility, similar to those obtained in hybrids between most species of annual sunflowers. CHROMOSOME NUMBERS AND PAIRING All of the annual sunflowers previously studied have been shown to possess a haploid chromosome number of 17 (Heiser, 1948). The same number has been found in all the races included in the present study. In most hybrids within races and between races belonging to either the “debilis” or “‘praecox” assemblages, 17 bivalents are formed at meiosis. Exceptions were noted in two crosses between races and two crosses within races of the ‘“‘debilis” assemblage where some cells occurred with 17 pairs and some with 15 pairs and a chain of 4 chromosomes. On the other hand, the hybrids resulting from crosses of races of “debilis” and “praecox”’ show the variable type of chromosome pairing previously reported for interspecific hybrids in Helianthus (Heiser, 1951). Twenty-five cells were examined in one hybrid plant from each cross and the number of pairs varied from 13 to 17. In the cells with 13 pairs, the remainder of the chromosomes generally formed two chains of four chro- mosomes each. Univalents were rarely observed. These configurations sug- gest that the two crossing groups differ by a minimum of two transloca- 148 MADRONO [Vol. 13 tions, and in all probability the sterility observed in the hybrids is the direct result of the abnormal pairing relations. Hybrids between “‘praecox” and H. petiolaris showed the same sort of configurations as observed in the “‘debilis”-“‘praecox”’ hybrid, whereas in hybrids between “‘debilis” and H. petiolaris and all of the other interspe- cific hybrids studied, with the exception of H. annuus X H. argophyllus, a still greater reduction in pairing at meiosis was observed. TAXONOMIC CONSIDERATIONS If sterility barriers develop along with morphological differences, there is no great problem in the delimitation of taxonomic categories (see Clau- sen, 1951), but sterility and morphological difference do not go hand in hand in H. debilis. The greatest morphological difference within this spe- cies is between the Atlantic Coast race of Florida and the other races of the species, whereas the gap based on crossing relationships falls between the “‘debilis’” assemblage and the “‘praecox”’ assemblage. In Texas there is considerable difficulty in distinguishing certain ‘“‘debilis” races from ‘“‘prae- cox ” races. If a purely morphological species definition were employed, two species might be recognized—the “‘debilis” race of east Florida con- stituting one species and all the other races constituting the second species. If reduction in fertility in the F, hybrid were used as the major criterion of species separation, two quite different species would be recognized— debilis and praecox. A compromise proposal might result in the creation of either one or three species in this complex. However, if only one spe- cies were to be recognized, then H. petzolaris would have to be included in it both on morphological grounds and crossability relations. Any one of the above proposals could be justified and there appears to be no one com- pletely satisfactory solution to species delimitation in this group. What- ever course is adopted does not alter the true evolutionary relation, and for the present it is proposed to recognize the whole assemblage of “‘de- bilis” and “‘praecox”’ races as a single species. This solution, fortunately, does not result in any radical departure from current taxonomic treat- ments of this group. The problem of the proper treatment of H. petiolaris will have to be deferred until a more thorough analysis of its variation has been undertaken, and for this reason no formal taxonomic treatment of it will be attempted here. A second problem is what taxonomic recognition, if any, should be given to the races. Thus far in the discussion the term race has been used to refer to populations or series of populations which replace each other geograph- ically and at the same time show conspicuous morphological differences. Within each of these races it is possible to recognize ‘“‘sub-races.”’ The best example is in “‘tardiflorus” in which three large, geographically isolated populations were studied and showed recognizable morphological differ- ences. It is possible also that races which appear very distinct to one who has been working with a group intensively may appear of minor signifi- cance to others. As Dobzhansky and Epling (1944) point out, it is a mat- 1956] HEISER: HELIANTHUS 149 ter of expedience, judgment, and conventions which prevail among stu- dents at a given time as to how a species is broken up taxonomically. It seems desirable at present to give recognition to eight of the races listed in Table 1. TABLE 1. CRossEs ATTEMPTED BETWEEN Races oF H. debilis Herbarium Crossed with Race No. Source specimen* race number “DEBILIS” ASSEMBLAGE debilis (Florida) 21 Marineland, St. John’s Co. 3179 6. a5 3 20 ~+#Flagiler Beach, Flagler Co. Godfrey 50894 17, 21 19 Melbourne, Brevard Co. 3185 2, 4, 10; 11 17. + Surfside, Dade Co. 3189 205 138 vestitus (Florida) 16 Long Beach, Sarasota Co. 3198 13 15 Anna Maria Key, Manatee Co. 52A5 13 14. Indian Rocks Beach, Pinellas Co. 3207 13 13. Clearwater, Pinellas Co. 3209 Dex Oh Oued eo al On nie £47 15.°16;-21 tardiflorus (Florida) 12 Sarasota, Sarasota Co. 3200 10 11 Cedar Key, Levy Co. 3211 3705-7, 9). .10,. 135.19 10 Carabelle, Franklin Co. 3221 Delo One Ove ts 13, 19 weed race 8 Augusta, Richmond Co., Ga. 4931 5, Age De. O. 9 Belvedere, Aiken Co., S.C. Sires Pose ipa be silvestris (Texas) 2 Nacogdoches, Nacogdoches Co. 3086 123, 476 lio 9 1 Grapeland, Houston Co. 3048 25 10 cucumerifolius (Texas) 4 Frio, Dilley Co. 3068 Bs Sy 0y Ones 19 3. Westhoff, DeWitt Co. 3074 War: Oe (ear em kG Mae I 13 “PRAECOX” ASSEMBLAGE praecox (Texas) 5 Galveston Island, Galveston Co. 3080 D 3+5.0 01, Os On Ont tee 13.021 runyonii (Texas) 6 Yturria, Willacy Co. Runyon 4365 A. Se it WOve Tt? L hirtus (Texas) 7 Carrizo Springs, Dimmit Co. 3064 Sy ae Oy fo Op 40, 1S *All specimens except where otherwise noted were collected by the writer. ooaaRaeae=IeaoaQaaQaaoaaaaaaaanaaoaannmuou0uuuunummE————ee oo 150 MADRONO [Vol. 13 TAXONOMIC HISTORY Helianthus debilis was described in 1841 by Nuttall from material col- lected in Florida, and a year later Torrey and Gray described H. cucum- ertfolius from Texas. In 1847 Engelmann and Gray described H. praecox, also from Texas. In his Synoptical Flora, Gray reduced H. praecox to synonymy under H. debilis and treated H. cucumerifolius as a variety of the species. Watson (1929) in his study of the genus recognized H. debilis and H. cucumertfolius and described another species, H. vestitus, from western Florida without indicating its relationship to either of the other species. In the recent edition of Gray’s Manual, Fernald (1950) treats H. debilis as a species with one variety, H. debilis var. cucumerifolius. Material from the following herbaria was seen during the course of the study: University of California, Chicago Natural History Museum, Uni- versity of Georgia, University of Florida, Gray Herbarium of Harvard University, Michigan State College, Missouri Botanical Garden, New York Botanical Garden, Pomona College, Southern Methodist Univer- sity, University of Texas, and Texas A. & M. College. I should like to extend my thanks to the curators of these herbaria for the privilege of studying this material. In the taxonomic treatment which follows, only the type specimen, one widely distributed specimen, and such others as are necessary for the discussion are cited. The distribution of the various taxa is shown in the maps. A mimeographed list of exsiccatae has been pre- pared and will be sent upon request to those interested. The sunflowers to which H. debilis appears most closely related are for the most part annual species with alternate leaves and purple disk flowers. In the key given here all of the species in this group which are found in the area from east central Texas to Florida are included; however, only H. debilis will be treated in detail in the part which follows. Key TO SPECIES Phyllaries over 4 mm. wide; heads 2.5 cm. in diameter or greater; leaves usually cordate; plants commonly 1.5 m. or more tall. Leaves, phyllaries, and stems with dense silvery white tomentum; in sandy soils; east coastal Texas, naturalized in Florida and also in cultivation ioe) te bc AM ieee SMUD Ae at een Siccy & aku E le ek ee SC ee H. argophyllus T. & G. Leaves, etc., merely hispid; rarely on sand; wide-spread weed and cultivated 10, EA eel Ren Re TIRE cM IML Deere Bina is Sil a att Le a. H. annuus L. Phyllaries 4 mm. or less wide; heads less than 2.5 cm. in diameter; leaves cuneate, truncate, or cordate; plants 2.0 m. or less tall. Leaves lanceolate with a few prominent white hairs near base; plants commonly 1.0 to 2.0 m. tall; in wet soils; south-central Florida........ H. agrestis Pollard Leaves deltoid-lanceolate without prominent white hairs; plants commonly less than 1.5 m. tall; in sandy soils. Leaves usually entire, generally twice as long as broad, bluish-green; tips of pales in center of disk densely villous; Western United States and Northern Mexico,naturalized? ini Hast > tre seas ome eae eee H. petiolaris Nutt. Leaves entire or more often serrate, generally less than twice as long as broad, not bluish-green; tips of pales villous to merely hispid........ H. debilis Nutt. 1956] HEISER: HELIANTHUS 151 Key TO SUBSPECIES OF HELIANTHUS DEBILIS Stems prostrate or nearly so; peduncles usually less than 20 cm. long; coastal regions. Leaf serration irregular; stems usually conspicuously white-hirsute; rays seldom over 1.6 mm. long; disk diameter less than 1.2 cm. (at anthesis of the outer row of disk flowers) ; west Florida ...................-...- 2. H. debilis subsp. vestitus Leaf serration fairly regular, shallow; stem glabrous, hispid, or hirsute. Stems completely prostrate, glabrous or hispid; rays less than 2 cm. long, yellow; Cas beh ORLA Site eed ass RA te ISR a es 1. H. debilis subsp. debilis Stems semi-prostrate, hispid-hirsute; rays usually 2 cm. or more long, orange- yellow; bracts about 3 cm. wide; Galveston Island and adjacent mainland, PING Kil Site Menem nale f 7 A cee MNES i APY teat soe) 6. H. debilis subsp. praecox Stems erect; peduncles usually over 15 cm. long; coastal and inland. Phyllaries mostly 2.5-4.0 mm. broad, rarely long attenuate; stems hispid to hir- sute; leaves usually with constriction near middle (fig. 6). Branches horizontal or ascending; stems not densely hispid-hirsute; tips of central pales of the chaff seldom villous; Galveston Island and adjacent matin Anse WEX AS GR foes ele ies aie iss eccesiete cae 6. H. debilis subsp. praecox Branches ascending; tips of central pales of the chaff usually villous. Stems hirsute but not densely so; peduncles seldom longer than 30 cm.; leaves seldom longer than 8 cm.; rays usually fewer than 14; southern IS: RBA ae Oe anit Re ee Sele ae eS APO 7. H. debilis subsp. runyonii Stems densely hirsute or hispid-hirsute; peduncles usually 30-40 cm. long; leaves frequently 8-9 cm. long; rays usually 14 or more; Carrizo Springs, PUNKS eee nL ny ret Pe OR Deca Rs 8 Cen i 8. H. debilis subsp. hirtus Phyllaries usually 2.0-2.5 mm. broad, usually long-attenuate; stems glabrous or sparingly hispid; leaves gradually tapering from base to tip (fig. 6). Lower leaves usually large, blade over 8 cm. long; peduncles very slender, 20—40 em. long; disks relatively small, seldom over 1.5 cm. at anthesis; north- eastern Texas chiefly in oak-pine woods.............. 4. H. debilis subsp. silvestris Lower leaves smaller, blade seldom over 9 cm. long; peduncles not conspicu- ously slender, disks 1.4-2.0 cm. broad at anthesis. Peduncles 25-50 cm. long; leaf serration usually shallow, fairly regular; rays usually 2 cm. or more long; disks usually 1.6 or more in diameter at anthesis; south central Texas.................. 5. H. debilis subsp. cucumerifolius Peduncles seldom over 25 cm. long; leaf serrations usually deep, irregular; rays usually less than 2 cm. long; west and north coastal Florida. eee ae, ee Te ROR LP SR MME Cd 3. H. debilis subsp. tardiflorus 1. HELIANTHUS DEBILIS subsp. debilis.? Helianthus debilis Nutt. Am. nilesoc. Nis, 77367, 1841. Annual or perennial, prostrate, stem and branches subglabrous to densely hispid; leaves irregularly shallowly serrate to nearly entire, blade glabrous to hispid, 3.0-10.0 cm. long, 2.0-8.0 cm. wide; peduncles gener- ally 10-20 cm. long; disk 1.1 to 1.4 cm. in diameter at anthesis; phyllaries 2 Helianthus procumbens Raf. (Fl. Ludovic. 70. 1817) has been provisionally con- sidered as synonymous with H. debilis Nutt. (Merrill, E. D. Index Raf. 237. 1949). This plant was observed by Robin presumably in what is now Louisiana or from the neighborhood of Pensacola. Helianthus debilis subsp. debilis does not occur in this region and H. debilis subsp. tardiflorus, which might be found near Pensacola, is not procumbent. Moreover, H. debilis does not have the opposite leaves described by Rafinesque for his H. procumbens. In any event H. procumbens Raf. is a later homonymn of H. procumbens Pers. (Syn. ii. 475. 1807). Rafinesque’s species might be Spilanthes repens Michx. 152 MADRONO [Vol. 13 about 2 mm. wide, short-attenuate, sometime squarrulose; rays 11-17, 1.2—2.0 cm. long, 0.5—1.0 cm. wide, yellow; central pales slightly hispid. Distribution and citation. Common on sandy beaches from St. John’s County to Dade County, Florida (fig. 2). Apparently an introduction at Key West and Lake City, Florida and in Cuba. Brevard County: Cur- tis 1441. Type. Baldwin, “E. Fla.,’ BM, not examined (photograph from GH, and US examined). oats 2 C) H. debilis a ss hah @ ssp. debilis Mh o/ @® ssp. vestitus ae j @® ssp. tardiflorus Red ip { —-—+ 5 os atthe: 2°11 pote 7 OH Lhe Fic. 2. Distribution of Helianthus debilis in Florida. (Base map through the cour- tesy of Dr. H. B. Sherman.) This subspecies is very distinct and well set off from its nearest rela- tive which appears to be subsp. vestitus. Helianthus debilis subsp. debilis is geographically isolated from other sunflowers, except H. argophyllus, which is naturalized in Volusia County. No evidence of hybridization be- tween these two taxa was found. The susbpecies is quite common on the keys of eastern coastal Florida and is rare even a short distance inland, although it is planted in yards at a number of localities in eastern Florida. The subspecies forms more or less continuous large populations in both the northern and southern parts of its range. However, in Indian River, St. Lucie, and Martin counties the populations are smaller and it seems possible that gene flow is cut down 1956] HEISER: HELIANTHUS 153 in this area. If populations from the northernmost part at Marineland are compared with southern ones from Surfside (fig. 6), the plants appear different enough to warrant subspecific distinction, but if a series of popu- lations from north to south is studied it is found that the variation in disk diameter, leaf length (fig. 4), and pubescence (Table 2) is in the nature H. debilis ssp. silvestris ssp. cucumerifoliws ssp. praecox Ssp. runyonii ssp. hirtus e H. pefiolaris ©0@P? Cc Fic. 3. Distribution of Helianthus debilis in Texas. The distribution of H. petio- laris (small dots) is given for Texas, Oklahoma, and a part of New Mexico. The latter species extends from Manitoba and Minnesota to Washington and south to northern Mexico. (Base map of Texas through the courtesy of Dr. Lloyd Shinners.) of acline. The leaves of the northern populations are more generally entire and cuneate whereas those from the southern counties are serrate and more frequently truncate (fig. 6). However, certain other characters, such as number and length of rays and peduncle length, do not show any signifi- cant trends. That this variation pattern is not merely ecological modifica- tion is clearly evident when comparisons are made with plants grown in the experimental garden (second column, fig. 4). The same general trend of variation is seen in the experimental garden populations, although the 154 MADRONO [Vol. 13 measurements for certain characters do not agree with those from plants in nature. In view of these data, however, it appears that the analysis of variation in nature in Helianthus debilis is reliable, although Clausen (1951) has implied that the analysis of natural variation is untrustworthy because of ecological modifications in many environments. In Florida subsp. debilis flowers throughout most of the year. At Bloomington plants started indoors in March come into bloom the first of July. TABLE 2. VARIATION IN LEAF PUBESCENCE IN H. DEBILIS SUBSP. DEBILIS FROM NorTH TO SOUTH FROM NATURAL AND GARDEN POPULATIONS* Collection Number of plants number* Source glabrate hispid 3179 Marineland, St. Johns County 20 0 5221 Marineland, St. Johns County 18 0 5220 Flagler Beach, Flagler County 23 0 3185 Melbourne Beach, Brevard County 20 0 5219 Melbourne Beach, Brevard County 20 2 3187 Palm Beach, Palm Beach County 18 2 5218 Palm Beach, Palm Beach County 11 8 3189 Surfside, Dade County 9 11 5217 Surfside, Dade County 4 18 * The 3100 series was collected in nature, and the 5200 series was grown in the Experimental Garden at Bloomington, Indiana. 2. HELIANTHUS DEBILIS subsp. vestitus (KE. E. Wats.) stat. nov. H. ves- titus E. E. Wats. Pap. Mich. Acad. 9:347. 1929. Annual or perennial, stem rather densely hirsute with whitish hairs, main stem erect or semi-erect to 30 cm. tall, lateral branches becoming decumbent; lower leaves deltoid-lanceolate to deltoid-ovate, usually cor- date or truncate, rarely cuneate at base, irregularly serrate, occasionally somewhat undulate, 4.0-8.0 cm. long, 2.5—6.0 cm. wide; peduncles 9-15 cm. long, usually hirsute; disk 1.1-1.2 cm. in diam. at anthesis; phyllaries 1.5—2.0 mm. wide, short-attenuate; rays 12—15, 1.2—1.6 cm. long, 0.6—0.7 cm. broad, yellow-orange; central pales hispid. Distribution and citation. Known only from the keys off west central Florida from Pinellas County to Sarasota County (fig. 2). Manatee County: Cuthbert 1445. Type. Hog Island (now Caladesi), Pinellas County. Tracy 6919, MSC. Variation. This subspecies has at maturity a prostrate habit and in that respect resembles subsp. debilis and subsp. praecox. However, it does EXPLANATION OF FIGURE 4 Fic. 4. Variation in leaf length of Helianthus debilis subsp. debilis from north to south in Florida. On the left are shown histograms from four population samples from Florida; on the right are shown histograms from seven populations grown in the experimental garden at Bloomington. The leaf length is given on the horizontal axis. The number of individuals in each population ranges from 17 to 24. The col- lections or seeds come from the following localities: 3179, 5221, Marineland, St. John’s County; 5220, Flagler Beach, Flagler County; 5423, Daytona Beach ,Volusia County ; 3185, 5219, Melbourne Beach, Brevard County; 5425, Stuart, Martin County; 3187, 5218, Palm Beach, Palm Beach County; 3189, 5217, Surfside, Dade County. 1956] HEISER: HELIANTHUS LEAF LENGTH FLORIDA 3179 ¥.35 2 8 cm. 3185 x 43 3187 x Ov 3189 x 59 Fic. 4. Variation in leaf length of Helianthus debilis subsp. debilis. GARDEN 522 x 52 4 12 cm 5220 x 56 e © 5423 x 5. ’ 2219 x6 © i oO 5425 Kod r 5218 XF, O : 6) 217 Xa w : 155 156 MADRONO [Vol. 13 not seem unlikely that such a habit could have originated independently in these three taxa. Both geographically and morphologically this sub- species is intermediate between subsp. debilis and subsp. tardiflorus, al- though in the sum total of its morphological features it appears to be more closely related to subsp. tardzflorus. Four populations were studied in detail both in the experimental garden and from population samples in nature, and although there is considerable variability there is no evidence of clines or of hybridization with other subspecies. This subspecies is, like all other taxa in this species, characterized by red-purple disk flowers. It is of interest to note that one plant (5214-15) grown in the experimental garden had yellow disk flowers. Yellow disk flowers are known in H. annuus and H. Bolanderi and this character is recessive to red-purple disk flowers. The origin of the yellow disk flowers in subsp. vestitus is probably more readily explained on the basis of in- dependent mutation rather than hybridization. Flowering material has been collected in the field from January through October. In the experimental garden it comes into flower late in July. 3. HELIANTHUS DEBILIS subsp. tardiflorus subsp. nov. Herba annua ramosissima caulibus erectis glabris vel sparse hispidis maculatis vel ru- bris 60-75 cm. altis foliis inferioribus deltoideo-lanceolatis raro ovalibus cordatis irregulariter serratis interdum crispatis laminis 5-10 cm. longis 3-9 cm. latis pedunculis 10—25 cm. longis disci diametro 1.4—1.5 cm. phyllaribus 2-3 mm. latis ligulis 13-19, 1.4-1.6 cm. longis 6—8 mm. latis paleis hispidis. Distribution and citation. In isolated colonies on sandy beaches in west- ern and northern coastal Florida in Franklin, Levy, and Sarasota counties (fig. 2). Single plants of this subspecies were found in Alachua County (Heiser 3218) and Taylor County (Heiser 3219), and it was collected by Small and De Winkeler in 1919 (9169) in Okeechobee County, although I failed to find it in this county in 1951. Type. Sarasota County, Sarasota Key, Curtiss 1455, MO. This subspecies is most readily recognized by its usually deeply and irregularly serrate leaves (fig. 6). It also has a more bushy habit than the other subspecies because of its tendency toward abundant branching near the base. It appears to be most closely related to subsp. sdvestris, and at one time the distribution of these two subspecies may have been contin- uous across the gulf coast, although today they are separated by a consid- erable distance (fig. 2). Crosses between subsp. sélvestris and subsp. ves- titus give F, plants which are erect and very similar to subsp. tardiflorus in appearance. The three populations of this subspecies which have been studied in detail show rather conspicuous differences in type of branching, leaf length, degree and nature of leaf serration, ray number and peduncle length (fig. 5). Although these differences are maintained in cultivation, they do not seem to be sufficient basis for recognizing three taxa. 1956] HEISER: HELIANTHUS RAY NUMBER LEAF LENGTH FLORIDA GARDEN FLORIDA GARDEN 3198 x'3 5216 x 14 9 23cm. 10 x13 52l2 3200 x48 5212 x 72 32 x14 52t x 14 3211 x 74 5211 x 94 3221 ¥ 19 5210 x 18 3221 x82 5210 R92 a re el PEDUNCLE LENGTH FLORIDA GARDEN FLORIDA GARDEN 3198 x90 5216 % 136 3198 x17 5216 x12 3 3A-33cm 46 40-42cm. Re 2Acm. .9 18cm. x19 5212 x 161 5212 x 25.0 al 3211 x93 5S2il x15 321 x22 Sell x14 3221 x174 5210 x 22.9 3221 x 21 5210 ee rt | DISK DIAMETER per ee Fic. 5. Variation in four characters in Helianthus debilis subsp. vestitus (black) and H. debilis subsp. tardiflorus (white). The collections and seeds come from the following localities in Florida: 3198, 5216, Long Beach, Sarasota Connty ; 3200, 5212, Sarasota, Sarasota County; 3211, 5211, Cedar Key, Levy County; 3221, 5210, Cara- belle, Franklin County. 158 MADRONO [Vol. 13 Although both subsp. tardiflorus and subsp. vestitus occur in the Sara- sota region, the two do not come into contact and there is no clear-cut evidence of hybridization. However, it can be seen from figure 5 that the Sarasota population of subsp. tardiflorus closely approaches subsp. ves- titus in leaf length, and disk diameter. Such evidence might suggest that there had been past hybridization between these two taxa in this region, but the measurements of peduncle length do not support this, for the Cedar Key population of subsp. tardiflorus which is well isolated today from subsp. vestitus shows the short peduncles similar to those of subsp. vestitus, whereas the Sarasota population of subsp. tardiflorus has long peduncles. In nature flowering material of this species has been collected from March to September. In the experimental garden it matures extremely late (August and September) and several plants from Cedar Key (5211) were killed by frost before flowering. It seems possible that at times it may live through the winter in Florida, a fact which could account for specimens collected in flower in March. 4. HELIANTHUS DEBILIS subsp. silvestris subsp. nov. Herba annua caulibus erectis atro-rubris vel maculatis plerumque glabratis superne ramosis ramis aliquantum pendulis foliis inferioribus deltoideo-ovatis cordatis plerumque regulariter serratis laminis 6—14 cm. longis 4-13 cm. latis pedunculis gracilibus 16-30 (—40 raro) cm. longis disci diametro plerumque 1.5 cm. phyllaribus ca. 2 mm. latis ligulis 12-14 (—20 raro) 1.5—2.3 cm. longis 0.7—1.2 cm. latis paleis plerumque hispidis. Distribution and citation. In the pine and oak regions of northeastern Texas from Shelby County to Wood County south to Waller County and Milam County (fig. 3). Upshur County: Reverchon 2581. Type. Nacogdoches County: 3 miles south of Nacogdoches on U. S. Highway 59, Heiser 3086, IND. This subspecies is distinguished from the others by the large, almost invariably cordate, leaves (fig. 6), the pendulous branches which arise in the upper half of the plant, and the usually rather small heads. However, one population from Houston County (Hezser 3048) has rather large heads and many rays. The plants of this subspecies from the northern part of its range appear to be most closely related to subsp. tardiflorus, and like subsp. tardiflorus this subspecies matures late in the experimental garden. However, in nature it has been collected as early as May. The southern limit of the subspecies has been drawn somewhat arbitrarily, in part because herbarium material from this area is scanty and also the specimens are somewhat fragmentary. Some of the specimens from the southern part of the range approach subsp. cucumerifolius, and it may be that hybridization between the two subspecies occurs. Such might be ex- pected since there are no known barriers to interbreeding and the hybrids are fertile. Another factor which may influence the variation pattern is hybridi- zation with H. annuus. Hybrids have been found in Henderson County 1956] HEISER: HELIANTHUS 159 and H. annuus is rather common in the southern part of the range of subsp. silvestris (Heiser, 1951). Although no evidence for introgression of H. annuus was cited previously (Heiser, 1951), it now appears that such introgression might account for the high ray number, large disks, and $4644 DEBILIS MARINELAND DESILIS SURFSIDE Ad VESTITUS AA ADA TARDIFLCRUS SARASOTA RAAAS TARDIFLORUS CARABELLE SILVESTRIS Fic. 6. Leaf outlines of Helianthus debilis from plants grown in the experimental garden. Leaves were selected from five plants in each population in order to show the range of variation. H. debilis subsp. debilis, Marineland, St. John’s County, Fla., 5150; H. debilis subsp. debilis, Surfside, Dade County, Fla., 5217; H. debilis subsp. vestitus, Indian Rocks Beach, Pinellas County, 5214; H. debilis subsp. tardiflorus, Sarasota, Sarasota County, Fla., 5211; H. debilis subsp. tardiflorus, Carabelle, Frank- lin County, Fla., 5210; H. debilis subsp. silvestris, Grapeland, Houston County, Texas, 5162. < 1/5. 160 MADRONO [Vol. 13 perhaps the large leaves found in some populations of subsp. szvestris (Heiser, 1951, Table 2). 5. HELIANTHUS DEBILIS subsp. cucumerifolius (T. & G.) stat. nov.? H. cucumerifolius Torrey and Gray, Fl. N. Am. 2:319. 1842. H. debilis var. cucumerifolius A. Gray, Syn. Fl. 17:273. 1884. Annual, much branched, branches ascending, stem hispid below, usually conspicuously purple mottled, 55-65 cm. tall; lower leaves deltoid-ovate to deltoid-lanceolate, cuneate, truncate, or cordate at base, fairly regu- larly serrate, slightly wavy, margined, blade 4.0—9.0 cm. long, 3.0-8.0 cm. wide; peduncles not conspicuously slender, 25-50 cm. long, disk 1.6— rarely 2.0 cm. long, 7-12 mm. broad; central pales of the chaff with hispid or slightly villous tips. Distribution. In open sandy soils in southeastern Texas from Travis County, south to Victoria and Frio counties (fig. 3). Guadalupe County: EJ. Palmer 11650. | Type. Texas. Exact locality unknown. Drummond 171, NY. On the whole this subspecies appears more similar to subsp. runyonii than to any other subspecies of H. debilis, although it may intergrade with subsp. sé/vestris in the northern part of its range. There is within the sub- species considerable variation which remains to be analyzed carefully. In the northern part of its range the chaff tips are generally merely hispid, in the central part of the range both hispid and slightly villous chaff tips are found, whereas in the southern part of the range generally all of the plants have slightly villous chaff tips, approaching the condition found in subsp. runvonii and subsp. Airtus. Hybridization between the subspecies, of course, is one possible explanation of this variation. This subspecies also hybridizes with H. annuus and it is possible that introgression from this species is another source of variability, although as yet we have no clear evidence for it. This subspecies matures early, coming into bloom in the experimental garden in the middle of June. Flowering specimens from nature have been collected from April to October. 6. HELIANTHUS DEBILIS subsp. praecox (Engelm. & Gray) stat. nov. H. praecox Engelm. & Gray, Boston Jour. Nat. Hist. 5:221. 1847. H. debilis var. praecox A. Gray in Torr., Emory, Rep. U.S——Mex. Bound. (Bot.), 90. 1859. (The Schott specimen cited by Gray is subsp. cucumert- folius.) 3 Helianthus lindheimerianus of Scheele (Linn. 22:159. 1849) has been placed in synonymy with H. cucumerifolius by Watson (1929). Scheele’s description might refer either to H. annuus or to H. cucumerifolius. His statement that this plant is perennial further complicates matters, but this is probably an error. For his type he cites Lindheimer, Neubraunfels, Texas, August, 1846, but does not give a collection number. Lindheimer’s No. 96, which is H. debilis subsp. cucumerifolius, was collected at this locality, but does not agree as to date. His H. annuus No. 259 does agree, how- ever, and therefore I am inclined to consider H. lindheimerianus as syncnyanous with H. annuus subsp. texanus Heiser. 1956] HEISER: HELIANTHUS 161 Annual, usually somewhat prostrate or erect with branches horizontal, stem greenish, not conspicuously mottled, hispid to hirsute, 40-50 cm. tall; lower leaves deltcid-ovate, usually cuneate to truncate at base, finely and fairly regularly serrate, 3.0—8.0 cm. long, 2.0—7.0 cm. wide; peduncles 15-30 cm. long; disk about 1.4 cm. in diameter; phyllaries 3-4 mm. broad, short-attenuate; rays generally 14-15, 1.7—2.7 cm. long, 7-12 mm. broad; central pales of the chaff hispid or occasionally slightly villous. Distribution and citation. In coastal sands on Galveston Island, Texas, and adjacent mainland in Galveston and Chambers counties (fig. 3). Chambers County: Cory 51022. Type. Texas. Galveston County, Galveston Island, Lindheimer 97, MO. This taxon was originally described as a species, but later Gray reduced it to synonymy under H. debilis. Watson (1929, p. 355) states that the type of H. praecox should be referred to H. petiolaris, and the description to H. debilis. Watson has ‘‘completely misinterpreted” (to use his own words) the type specimen. The description matches the type perfectly. Watson was under the impression that the central pales of the receptacle were villous only in H. petzolaris. The semi-prostrate habit and the pubescence of this subspecies suggest an affinity to subsp. vestitus, but the similarities here are in all probabil- ity the result of parallel development of coastal ecotypes in Texas and Florida. When all the morphological features are considered, the rela- tionships of subsp. praecox are clearly seen to be with the southern Texas race, subsp. vunyont, with which it produces fertile hybrids. Helianthus annuus appears as a weed on Galveston Island, and hybrids with subsp. praecox have been found in several localities. This subspecies flowers fairly early (late June) in the experimental garden, and has been collected in nature in flower from July to November. 7. HELIANTHUS DEBILIS subsp. runyonil, subsp. nov. Herba annua caulibus erectis plerumque rubris hispidis vel hirsutis 50-60 cm. latis foliis inferioribus deltoideo-ovatis basi cuneatis vel truncatis plerumque denti- bus parvis acutis regulariter serratis laminis 3.0—7.5 cm. longis 2.0—6.0 cm. latis pedunculis 20-30 cm. longis disci diametro ca. 1.4 cm. phyllari- bus 3—4 mm. latis abrupte attenuatis ligulis plerumque 11—13, 2.2—2.4 cm. longis 1.0—1.1 cm. latis paleis florium disci centralium villosis. Distribution and citation. Coastal prairies from Aransas County to Cameron County, Texas (fig. 3). San Patricio County: Wiitehouse 18133. Type. Willacy County, Yturria, Robert Runyon 4365, IND. Morphologically this subspecies is intermediate between subsp. praecox and subsp. Azrtus, the two races with which it produces fertile F; hybrids. In the northern part of its range there is some difficulty in distinguishing this subspecies from subsp. cucumerifolius, and additional field study needs to be undertaken. The specimens from Brooks County (Ferris and 162 MADRONO [Vol. 13 Duncan 3200, Tharp 3834) are somewhat more robust than the other specimens and have thicker leaves. Hemsley (Biol. Cent. Am. 2:180. 1881-1882) lists H. praecox from northern Mexico based on the collec- tion of Berlandier, no. 2354. I have not seen this specimen, but it may well be H. debilis subsp. runyonit. This subspecies flowers fairly early (late June) in the experimental garden, and has been collected from April to November in Texas. 8. HELIANTHUS DEBILIS subsp. hirtus, subsp. nov. Subspeciei runyonii similis sed hirsutior laminis 3.0—9.0 cm. longis 2.0—7.0 cm. latis pedunculis 30-40 cm. longis disci diametro 1.5—1.7 cm. ligulis 14-16, 2.1—2.7 cm. longis 0.7—1.3 cm. latis. Distribution. Known only from near Carrizo Springs, Dimmit County, Texas (fig. 3). Type. Texas. Dimmit County, Carrizo Springs, Heiser 3064, IND. This local race is most distinctive because of the hirsute pubescence. It has previously been confused with H. petzolaris, and Watson has anno- tated specimens of it as such. Its relationship to H. debilis is most clearly seen in the leaves (fig. 7), and in its crossing relation it has been shown to belong to the “praecox” assemblage of H. debilis. In its chaff pubescence, however, it is almost identical to H. petiolaris. After three generations of crossing sister plants from this population (3064), progeny has been ob- tained with almost glabrous chaff. The latter species does not occur in this area, but H. annuus does and hybridizes with subsp. dirtus. However, the peculiar features of subsp. /ivtus almost certainly are not derived from H. annuus. This subspecies is apparently confined to a single population. It is of interest to note that the Carrizo Springs sand region is also the home of a number of other endemics (Shinners, oral communication). The pocket gophers also have a subspecies limited to this same small area (Davis, 1940). Subspecies /zrtus flowers relatively early in the experimental garden, and flowering plants have been collected in nature from April through July. HELIANTHUS DEBILIS AS A WEED Helianthus debilis makes its appearance as a weed in a number of locali- ties in the eastern United States outside of Florida and Texas. Apparently in most localities where it has been reported, it is merely locally estab- lished, frequently as an escape from cultivation. However, in Georgia and South Carolina it forms extensive colonies along roadsides and is well es- tablished as part of the weedy flora. The decision as to whether these plants are indigenous in these states or represent introductions from either Florida or Texas rests on the following evidence. In Texas and in Florida H. debilis occurs in a number of places which are relatively undisturbed by man, whereas in Georgia and South Carolina the plants occur only as 1956] HEISER: HELIANTHUS 163 Py YY. CUCUMERIFOLIUS FRIO Anas CUCUMERIFOLIUS WESTHOFF bAba & PRAECOX a4 RUNYONI | adbs HIRTUS Fic. 7. Leaf outlines of Helianthus debilis from plants grown in the experimental garden (cont.). H. debilis subsp. cucumerifolius, Frio, Dilley County, Texas, 5161; H. debilis subsp. cucumerifolius, Westhoff, DeWitt County, Texas, 5156; H. debilis subsp. praecox, Galveston Island, Galveston County, Texas, 5159; H. debilis subsp. runyoni, Yturria, Willacy County, Texas, 5160; H. debilis subsp. hirtus, Carrizo Springs, Dimmit County, Texas, 5158. « 1/5. roadside weeds. Secondly, H. debilis was not collected in the latter states until 1899 (Cuthbert) whereas the species was collected in both Florida and Texas before 1850. Moreover, the Cuthbert collection from Georgia has “escape” written on some of the sheets. From their geographical position one might suppose that these weed races in Georgia and South Carolina were introduced from Florida, but all in all the weed specimens seem to have more in common with subspe- cies cucumerifolius of Texas. However, the weed specimens differ from this subspecies in their larger disks (1.7—2.0 cm. in diam. at anthesis), and larger leaves (9-12 cm. long, 8—11 cm. wide) which are almost invariably 164 MADRONO [Vol. 13 cordate at the base. For the present I do not feel that it is necessary to give the weeds a distinct epithet, but simply refer them to H. debilis subsp. cucumerifolius. Both the large heads and leaves could have been derived through introgression from H. annuus. Furthermore, segregates with a yellow disk have appeared among plants grown in the experimental garden; this also suggests influence from H. annuus. Distribution. Connecticut: Fairfield County, New Haven County, New London County. Massachusetts: Barnstable County. Pennsylvania: Lehigh County. Virginia: Isle of Wight County. North Carolina: Samp- son County, Pender County. South Carolina: Florence County, Aiken County. Georgia:+ Taylor County, Richmond County. Alabama: Houston County. Louisiana: Calcasieu Parish, Rapides Parish. ORNAMENTAL RACES Helianthus debilis is grown as a garden ornamental and is carried by a number of seed companies in the United States and Europe. Without much doubt it is the most attractive annual species of the genus. A number of strains have been grown both at Bloomington and in St. Louis. There is considerable variation between strains, particularly in regard to ray color, which varies from a pale primrose to a deep orange yellow; mor- phologically most of them resemble either subsp. cucumerifolius or subsp. sélvestris. Some of the ornamental strains are almost identical to the weeds of South Carolina and Georgia. Some seeds offered as “‘H. cu- cumerifolius” when grown proved to be H. annuus, and other strains show evidence of having been derived from hybrids between H. annuus and H. debilis subsp. cucumertfolius. DISCUSSION AND CONCLUSIONS For purposes of discussion we may think of H. petiolaris, the ‘“debilis”’ assemblage (subspecies debilis, vestitus, tardiflorus, silvestris and cucu- merifolius), and the ‘‘praecox”’ assemblage (subspecies praecox, runyonii and /irtus) of H. debilis as three incipient species. It seems highly prob- able that they all have come from a recent common ancestor. The morpho- logical differentiation that has occurred can perhaps most readily be ex- plained by geographical isolation which has permitted the development of different gene combinations in response to different environmental condi- tions. Partial reproductive barriers leading to some sterility in F, hybrids appear to have developed through structural changes in the chromosomes. The incipient speciation in these sunflowers then can be readily explained in neo-Darwinian terms. In attempting to reconstruct the past history of this group we are hampered by the lack of any fossil evidence and of any indication of the ages of the various racial components, factors which make it exceedingly difficult to explain the present distribution. It is known, however, that the region inhabited by H. debilis is for the most part quite young geologically. In the following discussion certain assump- 4 Thorne (Castanea 16:29. 1951) also reports it from Clay and Seminole counties. 1956] HEISER: HELIANTHUS 165 tions are made. The principal ones are rather obvious: (1) races having the same structural arrangement of chromosomes as indicated by pairing at meiosis come from the same progenitor, and (2) in general, morpho- logical similarities indicate closeness of relationship. Moreover, we shall speak of certain types as the progenitor or original type, although it is fully realized that in all probability no existing race is similar to the progenitor. It has been shown on morphological grounds that H. debilis subsp. hirtus connects “praecox” to H. petiolaris and that subsp. runyoni con- nects “‘praecox” to “debilis” by way of subsp. cucumertfolius. From the crossing studies it has also been shown that “praecox”’ forms a bridge between “‘debilis” and H. petiolaris. Assuming that one of these three is similar to the progenitor we have the following possibilities: (1) H. petio- laris could have been the progenitor which gave rise to “‘praecox”’ which in turn gave rise to “debilis.”” This would imply a western origin and then a gradual spread over the newly arisen land in the gulf coast region some- time in the Miocene or later. In this connection it should be pointed out that five other annual sunflowers today are centered in the western United States, whereas only two species are native to the gulf region. (2) The second possibility is that ‘‘praecox”’ is the original type which in turn gave rise to H. petiolaris to the west and “‘debilis” to the north and east. (3) From the evidence we have it is equally likely that a subspecies of the “debilis” assemblage could have been the ancestral type which gave rise to “praecox” and that “praecox”’ in turn gave rise to H. petiolaris. Thus it can be seen that no definite conclusions can be reached in regard to the progenitor. A second problem is to explain the origin of subsp. debilis, which has been shown to differ greatly morphologically from the remainder of the species. Such great morphological differentiation might be explained by an early isolation of the progenitor of this subspecies from the ancestral species. One is tempted to postulate long-distance dispersal of H. debilis into Florida in the lower Oligocene when Florida appeared as an island isolated from the mainland (Schuchert, pl. 11, 1935) and later a gradual spread of subsp. vestitus and subsp. tardiflorus into peninsular Florida from the northwest. In view of the fact that the differences between subsp. dedilis and the other subspecies of the “debilis” assemblage are as great as, or greater than, the differences between ‘‘debilis,”’ “praecox,” and H. petiolaris in Texas we might suppose that the isolation of the ancestor of subsp. debilis took place before the differentiation of the other taxa. Then at some later date H. petiolaris, ‘““praecox,” and “‘debilis,” excepting subsp. debilis, dif- ferentiated with the latter assemblage retaining the original structural arrangement of the chromosomes. If this is actually the case it means that all the other subspecies of ‘‘debilis’” are more closely related to “praecox” and H. petiolaris in terms of a recent common ancestor than to subsp. debilis. If these crossing groups were interpreted as species it 166 MADRONO [Vol. 13 can be seen that it might give us an unnatural classification. To follow this to its logical conclusion we might point out that the use of sterility or crossing limits, which is the primary prerequisite in most genetic species definitions, may actually obscure understanding relationship at times. Thus far, in order to complicate the discussion as little as possible, no consideration has been given to the possibilities of hybridization in at- tempting to explain the evolution of these sunflowers. It is difficult to evaluate the role of hybridization, particularly since it may go hand in hand with the other mechanisms promoting evolution, but it has almost certainly played a role. It has been pointed out repeatedly that “praecox”’ is intermediate be- tween H. petiolaris and “‘debilis,” and although the latter two are not known to hybridize today it is possible that past hybridization could have taken place which gave rise to ‘‘praecox.”’ Unfortunately, few crosses have been made between H. petiolaris and “‘debilis,’’ but in those which have been made, the F, progeny would most nearly key out to “‘praecox”’ although they do not exactly match any of the existing three subspecies. The artificial F,; has reduced fertility, but it seems possible that stable, fertile types might be selected from hybrid derivatives of H. petiolaris and “debilis” which would differ both morphologically and in chromosome structure from the parental types. There also exists the possibility that certain of the subspecies owe their origin at least in part to hybridization. Helianthus annuus is known to hybridize naturally with several of the subspecies. Although in an earlier paper (1951), wherein all Texas races of H. debilis were referred to H. debilis var. cucumertfolius, it was stated that there was no evidence of introgression of H. annuus into “‘debilis,” it now seems probable that genes from H. annuus might account for the large disks, numerous rays, and large cordate leaves of certain Texas populations of subsp. sévestris. It has also been pointed out that subsp. ¢ardzflorus in some respects is inter- mediate between subsp. sdlvestvis and subsp. vestitus, although from present-day distribution one can hardly postulate hybridization to explain the origin of subsp. tardiflorus. Other possibilities of hybridization modi- fying the subspecies have been pointed out in the section on taxonomy. The foregoing discussion has explored some of the possibilities which might account for the present pattern of subspeciation in H. debilis. How- ever, we cannot satisfactorily point to one taxon as the definite progenital type, nor can we evaluate fully the role of hybridization. In a group of closely related diploid organisms in which nothing is known of the past history of the taxa, and many times when it is, it is practically impossible to construct a phylogenetic ‘‘tree.’’ On the other hand, it has been pos- sible to show the relationships of the various taxa as they exist today using the criteria of comparative morphology and crossability. Department of Botany, University of Indiana, Bloomington. 1956] GILKEY: TRIFOLIUM 167 LITERATURE CITED CLAUSEN, J. 1951. Stages in the evolution of plant species. Ithaca. Davis, W. B. 1940. Distribution and variation of pocket gophers (genus Geomys) in the southwestern United States. Texas Agr. Expt. Sta. Bull. No. 590. DoszHANSKY, TH. and C. EprLinc. 1944. Contributions to the genetics, taxonomy and ecology of Drosophila pseudoobscura and its relatives. II. Chromosomal races in Drosophila pseudoobscura and Drosophila persimilis. Carnegie Inst. Wash. Publ. 554:49-144. FERNALD, M. L. 1950. Gray’s Manual of Botany. 8th Ed. New York. HEIser, C. B. 1951. Hybridization in the annual sunflowers: Helianthus annuus x H. debilis var. cucumerifolius. Evolution 5:42-51. . 1954. Variation and subspeciation in the common sunflower, Helianthus annuus. Am. Midl. Nat. 51:287-305. ScCHUCHERT, C. 1935. Historical geology of the Antillean-Caribbean Region. New York. Watson, E. E. 1929. Contributions to a monograph of the genus Helianthus. Pap. Mich. Acad. 9:305-—475. A NEW TRIFOLIUM FROM OREGON HELEN M. GILKEY In May of 1954, Mrs. Bessie Fleischman Murphy, an analyst at the Seed Laboratory located at Oregon State College, discovered in south- eastern Oregon a colony of a conspicuous large-headed clover, specimens of which she brought to the writer for identification. It proved strikingly different not only from any species previously known in the state but, so far as could be determined, from any thus far described. Consequently it is offered here as new to science. In size of head, this species is comparable, in northwestern United States, only to Trifolium macrocephalum (Pursh) Poir. and T. Thomp- sont Morton. But resemblance to either of these species ceases with this character. In fact the new clover is barred, by its possession of only three leaflets, from the section Macrocephala to which these two species are assigned. The single colony located by Mrs. Murphy represents practically a pure stand, though the plants are scattered. It occurs in Malheur County on a bluff above Sucker Creek, a tributary of Snake River, in a section where fingers of coarse-textured blue-gray diatomaceous earth protrude into the darker soil of typical sagebrush plains. The new clover occurs only upon these light-colored areas, its glaucous herbage blending closely with the blue-gray soil. Except upon this slope which is pre-empted by the clover, Artemisia tridentata is the predominant species in the region. 168 MADRONO [Vol. 13 Fic. 1. Owyhee Clover. Trifolium owyheense: habit, flowering stem and root, xX 34; a, flower, x 2%; b, flower at later stage, showing withered corolla and in- flated calyx, X 214; c, pistil from b, & 2%; d, lower leaf, including stipules, X 34. 1956] GILKEY: TRIFOLIUM 169 The partial chemical report of the soil, made by the Soils Department at Oregon State College, is as follows: Phosphorus Potassium Sodium Total salts Soil pH Ibs./A. level Ibs./A. level ppm. ppm. 6.8 7.6 ib 2000 VH 5000 140 The precipitous nature of the slope is favorable to a rapid run-off of melting snow, and unfavorable to the establishment of a foot-hold on the part of any except the most tenacious of plants. The large fleshy deeply- penetrating root with widely spreading and extensively developed branch- es, together with the glaucous and somewhat succulent herbage, indi- cates adaptations to a difficult environment. Mrs. Murphy and members of her family secured a good series of speci- mens in flowering and fruiting stages, also kodachrome slides of the col- ony and of individual plants in the field. The specimens and kodacolor prints are filed in the Oregon State College Herbarium. The specific epithet chosen for the species is derived from Owyhee, the Indian name for river and dam which are predominating factors in giving agricultural status to this section of the state. Trifolium owyheense sp. nov. Herba glauca ut videtur perennis e radice carnosa caulibus diffusis, 6-8 cm. longis; foliolis crassis glaucis obovato-emarginatis vel orbiculatis, raro denticulatis; stipulis crassis glaucis rotundatis; capitulo ebracteato, pyramidato, 3.5-4 « 3.5—4 cm.; calyce membranaceo tubo breviter obconico subvilloso demum inflato; laciniis glabris, subaequalibus, anguste subulatis apice aristulatis; corolla calyce duplo longiore; vexillo roseo, angustato apice truncato basi albido; alis roseis; carina saturate rosea; semina 2-3 elliptica maculata. Trifolium owyheense. Owyhee Clover (fig. 1). Plant probably peren- nial, glaucous; stems several, spreading, reaching 2 dm. in length; leaflets thick, broad, overlapping, more or less emarginate, rarely sparingly and minutely dentate, glaucous-green with white crescents; stipules thick, glaucous, slightly lobed, the members of a pair sometimes fused wholly or in part; head without an involucre, pyramidal, 3.5-4 « 3.5-4 cm.; calyx tube membranaceous, short-conical, somewhat villose, at length inflated, the teeth subequal, narrowly subulate, deep green, glabrous, the apices soft-aristulate; corolla twice as long as the calyx, vexillum narrow, the apex truncate, rose-colored, the base translucent-white, wings rose- colored, keel deep rose; seeds 2-3, elliptic, spotted. Collected in Malheur County, Oregon, May 1954, by Bessie Fleisch- man Murphy. Type in Oregon State College Herbarium. Oregon State College, Corvallis. 170 MADRONO [Vol. 13 A NEW RECORD OF DWARF MISTLETOE ON LODGEPOLE AND WESTERN WHITE PINE JosB Kurjt From various collections examined at the University of British Colum- bia the writer has for some time suspected the occurrence of dwarf mistle- toe, Arceuthobium campylopodum Engelm., on lodgepole pine (Pinus contorta Dougl. ex Loudon) in the Pacific Coast region of British Colum- bia. Observations in the field in 1954 confirmed these suspicions, this mistletoe being found not infrequently in stands or small groups of lodge- pole pine on the eastern part of Vancouver Island as far north as Comox, and also on this host near Sechelt on the mainland. Both the occurrence and the symptoms of this host-parasite combina- tion indicate some interesting parallels with those of the larch mistletoe- lodgepole pine combination.' Brooming is almost completely absent, ex- cept where the infections are so abundant that branches appear stunted in part of the tree. The swellings produced are fusiform, the older ones becoming markedly elongated and attaining a considerable size. Near Horne Lake, on Vancouver Island, a single small tree of western white pine (Pinus monticola Dougl.) also was found to bear dwarf mistle- toe shoots protruding from fusiform to spherical swellings on the branches and main stem. The mistletoe is believed to be the same as that on the lodgepole pine, which locally formed a heavily infected overstory. Although its morphology leaves no doubt as to its inclusion in Arceu- thobium campylopodum, the intraspecific identity of the mistletoe occur- ring on these two species of pine is questionable. This species is at present subdivided into forms, such as f. tsugensis, f. laricis, f. blumert, etc., ex- clusively on the basis of host relationships, no consistent morphological differences as yet having been found. No one form is known to be restricted to lodgepole pine; both f. /avicis and f. campylopodum have been found on this host. From the very nature of the present classification, it is there- fore impossible to identify, below the specific level, the mistletoe here reported. The plants on Pinus monticola could possibly be referred to f. blumert, which is said to be restricted to white pines. The conditions under 1 Kuijt, J. Some notes on the larch mistletoe in British Columbia. Can. Dept. Agr. Forest Biol. Div., Bi-monthly Progress Rept. 10(6). 1954. EXPLANATION OF FIGURE 1 Fic. 1. Arceuthobium campylopodum infections. A. On lodgepole pine (Pinus con- torta) at Horne Lake, Vancouver Island. Although literally hundreds of infections are present on the tree, and practically all larger branches are infected, only the lower ones show any signs of brooming. B. Typical swellings on lodgepole pine near Quali- cum Beach, Vancouver Island. C. On lodgepole pine. Close-up of female plant, with mature, recurved berries and young flowering shoots. August. D. On Western white pine (Pinus monticola) near Horne Lake, Vancouver Island. Swellings on this tree seemed more severe than those on lodgepole pine. Arrows indicate small plants pro- truding from the swellings. Nga DWARF MISTLETOE KUIJT 1956] infections. pylopodum mum cam Arceuthobi 1 Fic 172 MADRONO [Vol. 13 which it was found, however, very strongly indicate that this is the same mistletoe as that on the overstory lodgepole pine. The writer prefers there- fore not to assign this mistletoe to a definite form of Arceuthobium cam- pylopodum. There nevertheless remains the possibility that the mistletoe involved is no other than that on western hemlock. In some localities infected hem- lock and lodgepole pine were indeed found together; in others, however, no hemlock was present. Moreover, in several mixed stands either the hemlock or the lodgepole pine was heavily infected, while no infections were found on the other host. If cross-inoculations will eventually estab- lish that we are here concerned with hemlock mistletoe only, some ex- planation will be required for the apparent freedom from mistletoe in these cases. Most of the collections of dwarf mistletoe checked in 1954 as occur- ring on lodgepole pine on the Pacific Coast had previously been identified as A. americanum Nutt. ex Engelm. However, no reliable collection of A. americanum from west of the Coast Range has as yet been seen by the writer. There is a distinct possibility that, at least in Canada, no 4. americanum is present west of these mountains. This would parallel the western limits of distribution of the Douglas fir mistletoe (A. douglasi Engelm.), although the ranges of neither species are continuous east of this mountain system. University of British Columbia, Vancouver PROSOPIS GLOBOSA GILL. IN BAJA CALIFORNIA RoBERT L. DRESSLER An attractive yellow-flowered, mimosoid shrub was one of the most interesting plants found on a trip to the /dria forest area beyond Rosario, Baja California, Mexico, and on west from San Agustin to the Pacific Coast at Santa Catarina Landing. The plant in question proved, on our return, to baffle all attempts at identification; to be sure, it keyed to Prosopis or Neptunia, but it did not resemble any North American rep- resentative of either genus. Had fruiting material been available, I fear the plant would have been described as a new genus. For several years this troublesome shrub continued to resist the efforts of taxonomists far more competent than myself; then, by chance, I happened across a photograph of the Argentine Prosopis globosa Gill. (P. striata Benth.) and recognized it as being very similar to the Mexican plant. Prosopis globosa is the only species in the section Lomentaria Speg. (Burkart, 1940). While our Mexi- can plant differs in some respects from P. globosa, the differences are small, and it seems best to assign it to varietal status in this species. Thus, we add yet another species to the series showing close relationship 1956] DRESSLER: PHOSOPIS TS PLANTS OF MEXICO Raja California Preenpis globosa G41]. var. mexicana Dregealer Senta Catarina Landing. April 16, 1943. Wash in rocky brushland. lenge bush 1.3 m tall, fla, yellow, fragrant. Gollected by: Robert L, Dressler Ho. Gti: Determined by. BLD, Fic. 1. Type specimen of Prosopis globosa Gill. var. mexicana Dressler (GH). 174 MADRONO [Vol. 13 between the arid regions of North and South America. I had planned to write a summary of this interesting floristic relationship and include it with the description of P. globosa var. mexicana, but that “‘brief” consid- eration has got quite out of hand and cannot be subordinated to this tax- onomic note. For the full synonymy of P. globosa var. globosa, see Burkart (1945). Prosopis GLoBosA Gill. var. mexicana var. nov. A specie differt foliis maioribus (pinnis 20-25 mm. longis, foliolis 4-8 mm. longis) et pedun- culis longioribus (3.5—5 cm.) et lignescentibus. Dense hemispheric shrub ca. 1.3 m. tall; stems green with straw-colored longitudinal striations; stipules linear, erect, 2-3 mm. long, tomentulose; leaves bipinnate, unijugate; petioles (1)—3—7 mm. long, with rudimentary, tomentulose rachis (0.5—2 mm.) usually projecting beyond pinnae, and an erect, truncate gland ca. 0.5 mm. high between pinnae; pinnae 10—25 mm. long (petiolules 4-12 mm.), the rachillae with a tomentulose mucro pro- jecting ca. 1 mm. beyond leaflets; leaflets 2 or 3 pairs per pinna, glab- rescent, oblong, obtuse, 4-8 mm. long, 2—4 mm. wide; peduncles axillary, stout, striate, 3.5-5 cm. long; flowers in dense, spherical heads ca. 1.5—2 cm. in diameter; bracts inconspicuous, tomentulose, linear-lanceolate, ca. 2 mm. long; flowers sub-sessile; calyx campanulate, strigose-tomentulose, 1.5—2 mm. long, lobes broadly deltoid, ca. 0.5 mm. long; petals ca. 4 mm. long, lanceolate, obtuse, glabrous; filaments 5-7 mm. long; anthers ca. 1 mm. long, with prominent, stipitate, globose gland at apex; ovary sparsely strigose, stipitate (stipe ca. 1-1.5 mm. long) 2—2.5 mm. long, style 4-5 mm. long. Fruit unknown, but apparently a loment as in the South American variety. Mexico: Baja California, Santa Catarina Landing (ca. 29° 30’ N., 115° 15’ W.), wash in rocky brushland, April 16, 1949, R. L. Dressler 604 (type GH, isotype UC). This variety differs from the nearly leafless South American plants only in degree. The leaves are larger in all parts, but the xtreme measurements overlap. The stipules are not markedly spinescent as in var. globosa. The peduncles are not only longer (3.5—5 cm. as compared to 0.7—2 cm.), but are much stouter and more persistent. The flowers of var. mexicana are not so clearly pedicellate, but otherwise tend to be a little larger in all parts. Many, but not all, South American plants are more strongly pu- bescent. A fragment of the persistent margin of the fruit which is present on one old peduncle of the type specimen (fig. 1, right center) indicates that the fruit is probably similar to that of the Argentine plant. Gray Herbarium, Harvard University LITERATURE CITED BurKART, ARTURO. 1940. Materiales para una monografia del género Prosopis (Leguminosae). Darwiniana 4:57-128. . 1945. La intricada sinonimia del “manca-caballo”, Prosopis globosa Gill. Physis 20:53-55. 1956] REVIEW L75 REVIEW Weeds. By W. C. MUENSCHER. xvi + 560 pp., 135 figs. (Second Edition). Mac- Millan Company, New York. 1955. $10.00. Public service, an important function of tax-supported institutions, represents an integral part of the work at the University of California Herbarium. The Her- barium staff daily serves the people of the state by answering numerous requests for information about plants of horticultural or agricultural importance. Many of these requests have to do with the identification of weeds. Thus we who do this work are particularly appreciative of the recently published second edition of W. C. Muenscher’s well-known “Weeds.” “Weeds” was originally published in 1935. Although the second edition follows the same pattern, Part I (‘““General: Weeds and Their Control’) is more or less com- pletely rewritten, and Part II (“Weeds Arranged According to Family, Together With Key’) adds 71 new weeds to the original 500. The most outstanding change in the Second Edition is the deletion from Part I of the discussion on chemical control. This change was effected because of the recognition that chemical control has become a subject too complex and too significant to be treated concurrently with weed iden- tification within a single treatise. The prime purpose of the second edition of “‘Weeds”’ thus becomes their identification. Control in the broad sense is not neglected, how- ever, for its various manual aspects are carefully treated, the importance of biological and chemical control are mentioned, the most significant references to chemical control are given in the text, and in addition many references to the various phases of con- trol are included in the over 300 titles listed at the end of the book under “Literature References.” It is to be regretted, however, that a few outstanding references to the still experi- mental field of biological control were not included in the body of the text as is done for chemical control, nor are such references singled out from the terminal list of “Literature References” via footnotes in the text. Considered as out-of-place, unwanted plants, weeds are discussed as to their economic importance, their alarmingly successful ability for seed production, their multifarious methods of dissemination, their amazingly effective methods of vegeta- tive reproduction, and consequently their frequently ubiquitous and omnipresent occurrence. All these are subjects of vital importance to the intelligent application of eradication or control methods. Prevention, eradication, and control, and the prac- tical methods for obtaining each of these objectives are discussed. Control is of course the most limited objective, but often the only possible one. In Part II (“Weeds Arranged According to Family, Together With Key”) the key leads directly to species, rather than first to families and then to genera. The single key thus runs through 40 pages, a feature which cannot help but lead to some cumbersomeness although this is minimized by skillful handling. The arrangement of genera and species is alphabetical under the families which are in the usual Englerian sequence. Three hundred of the species are illustrated by line drawings. Of special interest in the careful treatment of each species is the short paragraph on control. California, being a state of highly varied topography and climate and consequently of great agricultural diversity, has attracted a plethora of alien plant species. For the identification of these California aliens, which number well over 700, there is already available the semi-technical but very readable and highly useful ‘Weeds of California” by W. W. Robbins, M. K. Bellue, and W. S. Ball (Dept. of Agri. State of California, second edition, 1951). Many of Muenscher’s 571 weeds do not occur in California, and conversely many weeds known from California are outside of Muenscher’s territory and are not listed in his book. In many ways, however, his book complements “Weeds of California,’ particularly in the much fuller general discussion and in the specific statement of control for each species listed. It is unfor- tunate that the price of such a useful book could not be lower. HELEN K. SHARSMITH, Department of Botany, University of California, Berkeley. 176 MADRONO [Vol. 13 NOTES AND NEWS EUPHORBIA ANTISYPHILITICA DISCOVERED IN SOUTHERN NEw Mexico. A very un- usual species, Euphorbia antisyphilitica Zucc., has been found in southern New Mexico. Candelilla, as it is commonly called, had only been reported in the United States from the Big Bend area of Brewster County, Texas, and 5 states in Mexico: Zacatecas, San Luis Potosi, Coahuila, Durango, and Nuevo Leon. The discovery of the species in New Mexico extends its known range some 350 miles north of the Texas locality. Fic. 1. Large clump of Euphorbia antisyphilitica on the east slope of the San Andres Mountains showing the vigorous and closely spaced branches characteristic of the species. This large clump of E. antisyphilitica, which is 30 feet across (fig. 1), was dis- covered by Mr. Cecil Kennedy of the U.S. Fish and Wildlife Service. It is located on Bennett Mountain on the eastern slope of the San Andres Mountains, Dona Ana County, at an altitude of 4,200 feet where it grows on a gravelly ridge and appears to be thriving as the stems are about 30 inches high. The rainfall here is limited, averaging 10 to 12 inchesa year. A very thorough search of the San Andres Mountains and vicinity by Mr. Kennedy, who has worked in the area for over 13 years, has not revealed the presence of additional plants of E. antisyphilitica. This newly discovered locality is possibly relictual from a more continuous and extensive range of the species which existed many years ago. Other desert plants associated with it are Tridens pulchellus, Enneapogon des- vauxii, Setaria macrostachya, Ephedra trifurca, Fouquieria splendens, Dasylirion wheeleri, Opuntia macrocentra, Dalea formosa, Larrea divaricata, and Parthenium incanum. Of interest is that, in Mexico, candelilla usually grows in association with another species of Parthenium known as P. argentatum or guayule. Candelilla, a Mexican name for the plant, refers to its very characteristic small reed-like leafless stems. The stems or branches, which are extremely numerous, are covered with a kind of wax which is extracted for commerce. To obtain the wax, the stems are boiled in water whereupon the grayish crude wax rises and is skimmed off. It is primarily used to make candles which burn with a bright light and an agree- able odor. The wax has also been used for phonograph records, shoe polish, floor and furniture polish, lubricants, and waterproofing. The plant is claimed to have purga- tive properties and is used in Mexico as a remedy for venereal diseases. Recent in- formation from Texas indicates that candelilla is no longer abundant in the Big Bend region where it has been exploited for its commercial properties. PHrrip J. LEYENDECKER and Crecit A. KenNEpy, New Mexico College of A. & M. A., State College, New Mexico and U. S. Fish and Wildlife Service, Box 791, Las Cruces, New Mexico. INFORMATION FOR CONTRIBUTORS Manuscripts submitted for publication should not exceed an estimated 20 pages when printed unless the author agree to bear the cost of the ad- ditional pages at the rate of $15 per page. Illustrative materials (includ- ing “typographically difficult” matter) in excess of 30 per cent for papers up to 10 pages and 20 per cent for longer papers are chargeable to the author. Subject to the approval of the Editorial Board, manuscripts may be published ahead of schedule, as additional pages to an issue, provided the author assume the complete cost of publication. Shorter items, such as range extensions and other biological notes, will be published in condensed form with a suitable title under the general heading, ‘“‘Notes and News.” Institutional abbreviations in specimen citations should follow Lanjouw and Stafleu’s list (Index Herbariorum. Part 1. The Herbaria of the World. Utrecht. Second Edition, 1954). Articles may be submitted to any member of the Editorial Board. MADRONO A WEST AMERICAN JOURNAL OF BOTANY A quarterly journal devoted to the publication of botanical re- search, observation, and history. Back volumes may be obtained from the Secretary at the following rates: Volume I, 1916-1929"*.- ©. «2. 0) °° $5.00 Volume IT, 1930-1934. . . . . 5.00 Volume ITI, 1935-1936 . . . . 5.00 Volume IV, 1937-1938 . . .. . 5.00 Volume V, 1939-1940. . . .. . 5.00 Volume VI, 1941-19427 0 5.00 Volume VII, 1943-1944 . . .. 5.00 Volume VIII, 1945-1946 . . ..., 5.00 Volume IX, 1947-1948 Tarte dite 6.00 Volume X, 1949-1950. . . .. . 7.00 Volume XI, 1951-1952. wi wt, 8.00 Volume XII, 1953-1954 . . .. 8.00 Single numbers, | Volumes:i and 25. ¢25. (08 .50 Succeeding numbers, — (except Volume XI, No.2)... 1.00 Volume:XT | No.2) els: We ay 1.50 The subscription price of MaproNo is $4.00 per year. If your institution does not now subscribe to MApRONO, we would be grateful if you would make the necessary request. Since there is no extra charge for institutional subscriptions, an individual may hold membership in the California Botanical Society on the basis of his institution’s subscription. Address all orders to: G. THomMAS Rossins, Corresponding Secretary Department of Botany University of California, Berkeley 4, California ADRONO VOLUME 13, NUMBER 6 APRIL, 1956 Contents PAGE STUDIES IN THE CAPPARIDACEAE II. THE MEXICAN SPECIES OF CLEOMELLA: TAXONOMY AND EVOLU- TION, Hugh H. Iltis 177 Two Funct ASSOCIATED WITH A Microcyctic Rust, COLEOSPORIUM CROWELLII CUMMINS ON NEEDLES OF PINUS EDULIS ENGELM. IN ARIZONA, Paul D. Keener 189 A NATURAL Hysrip, X ADIANTUM TRACYI C. C. HALL, : Warren H. Wagner, Jr. 195 | DOCUMENTED CHROMOSOME NUMBERS OF PLANTS 205 NOTES AND NEws: Terms USED TO DESIGNATE TYPE MATERIAL, S. Ff. Blake; NEws 207 CALIFORNIA BOTANICAL SOCIETY, REPORT OF THE TREASURER FOR 1955 208 A WEST AMERICAN JOURNAL OF BOTANY ?UBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY $F ee MADRONO A WEST AMERICAN JOURNAL OF BOTANY Entered as second-class matter at the post office at Berkeley, California, January 29, 1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price $4.00 per year. Published quarterly and issued from the office of Madrofio, Herbarium, Life Sciences Building, University of California, Berkeley 4, California. BOARD OF EDITORS HERBERT L. Mason, University of California, Berkeley, Chairman EpcAr ANDERSON, Missouri Botanical Garden, St. Louis. LyMAN BENSON, Pomona College, Claremont, California. HERBERT F. COPELAND, Sacramento College, Sacramento, California. Joun F. Davinson, University of Nebraska, Lincoln. IvAN M. JoHNsTON, Arnold Arboretum, Jamaica Plain, Massachusetts. Mivprep E. Marutas, University of California, Los Angeles 24. Marion OWNBEY, State College of Washington, Pullman. Ira L. WiccINs, Stanford University, Stanford, California. Secretary, Editorial Board — ANNETTA CARTER Department of Botany, University of California, Berkeley. Business Manager and Treasurer—MAtLcotm A. Noss Carnegie Institution of Washington, Stanford, California CALIFORNIA BOTANICAL SOCIETY, INC. President: Wm. M. Hiesey, Carnegie Institution of Washington, Stanford, Cali- fornia. First Vice-president: Roger Reeve, Western Regional Research Laboratory, Albany, California. Second Vice-president: Lewis Rose, California Academy of Sciences, San Francisco, California. Recording Secretary: Mary L. Bowerman, De- partment of Botany, University of California, Berkeley, California. Corresponding Secretary: G. Thomas Robbins, Department of Botany, University of California, Berkeley, California. Treasurer: Malcolm A. Nobs, Carnegie Institution of Wash- ington, Stanford, California. 1956] ILTIS: CLEOMELLA 177 STUDIES IN THE CAPPARIDACEAE II.1 THE MEXICAN SPECIES OF CLEOMELLA: TAXONOMY AND EVOLUTION Hucu H. ILTIs The Mexican species of Cleomella, a small North American genus of the Capparidaceae, are poorly known despite the fact that the genus was recently monographed by Payson (1922) and has received some attention since by West Coast floristic workers. During the preparation of a treat- ment of the Capparidaceae of Nevada (Iltis, 1955) and a phylogenetic study dealing with some North American Cleomoideae (Iltis, 1956), many specimens of Cleomella were examined. In the present study, a new species (Cleomella perennis) is described, the already-described taxa are more accurately delimited, some nomenclatorial problems are clarified, and certain aspects of the evolution of the taxa are discussed. An examination of the specimens labelled C. longipes Torr. at the United States National Herbarium revealed that these could be separated easily into two groups on the basis of both morphological characters and geographic distribution. The specimens from the area including northern Mexico (Chihuahua), southwestern Texas, and southeastern Arizona (C. longipes Torr., sensu stricto) are all rather robust annuals with an erect stem or leader and with relatively large leaves, while those from Durango, Zacatecas, San Luis Potosi and Guanajuato in central Mexico (C. perennis) are all perennials with strongly thickened caudices, slender, decumbent or ascending branches, much smaller, narrower leaves, and fewer-seeded silicles. Despite these rather marked differences, the north- ern and southern populations have been classified under the same name for over one hundred years, largely because the shape of the fruit and length of the gynophore are nearly identical, two criteria usually of critical im- portance in other Cleomoideae. The only other Mexican species, the rare Cleomella mexicana Moc. & Sessé in DC., of the south-central Mexican plateau, is easily distinguished from its two northern relatives by its very short gynophore, smali leaves, and prostrate branches. As was recognized by de Candolle (1824), it, too, is perennial, with a woody caudex and rootstock which is well illustrated in the “type-drawing” (cf. p. 182). This condition, occurring in two out of the three Mexican species of Cleo- mella, was completely overlooked by Payson (1922), who considered all species of the genus as annuals. 1 The present study was made while the author was on the staff of the Depart- ment of Botany, University of Arkansas, Fayetteville, where he has had support from the Institute of Science and Technology, College of Arts and Sciences, and a grant-in-aid from the Society of Sigma Xi, to both of which he would like to express his thanks. Maprono, Vol. 13, No. 6, pp. 177-208, April 11, 1956. 178 MADRONO [Vol. 13 I. TAXONOMY CLEOMELLA DC. Prodr. 1:237. 1824. (Type: C. mexicana Moc. & Sessé in DC.). Jsopara Raf. Ati. Jour. 1:144. 1832. (Based on C. mexicana Moc. & Sessé in DC.). Hyponema Raf. Good Book 40. 1840. (nom. nov. Cleomella DC. 1824). Cleome sect. Cleomella (DC). Baill. Hist. Plant. 3:149. 1872. Erect to decumbent, unarmed, slender to robust, annual or perennial herbs, glabrous (except C. obtusifolia). Leaves 3-foliolate; leaflets entire, mucronulate, less than 4 cm. long, more or less thick, stipulate, the stip- ules minute, hyaline, filiform, usually less than 3 mm. long. Racemes ter- minal, usually bracteate, the bracts either 1-foliolate or the lower 3-folio- late, or flowers borne singly in the axils of cauline leaves. Sepals very small, free or basally connate, tardily deciduous. Corolla closed (convo- lute) in the bud; petals yellow, sessile or subsessile. Disk small, fre- quently expanded adaxially. Stamens 6, the anthers tightly coiled when dry. Fruit a small obdeltoid, rhomboidal, deltoid or ovoidal silicle, usually wider than long, often compressed contrary to the placenta, the valves usually conical. Silicles borne on short to elongate gynophores, the indu- rate, persistent styles usually setulose. Seeds 2-12 (—19) in each silicle, free-falling, the two claws formed by the invagination of the testa fused nearly their whole length, with only a trace of an internal sinus between them; testa usually smooth and often delicately marked (colliculate). A genus with ten species, endemic to the western United States and Mexico; very closely related to Cleome sect. Peritoma (DC.) Baill., from which it differs principally in the smallness and shape of the fruit. Key TO THE MEXICAN SPECIES OF CLEOMELLA A. Leaflets of principal leaves 15-30 mm. long, 3-10 mm. wide; plants annual, from a usually slender taproot, with a branched or unbranched main stem (leader) usually simple at the base, the branches strongly ascending ; Chihuahua to south- western Texas and southeasterni:Arizona: =... ee ee 1. C. longipes AA. Leaflets of principal leaves 5-14 mm. long, 1-5 mm. wide; plants perennial, much branched at the base from a greatly thickened, short woody root and caudex, the branches ascending to decumbent or procumbent. B. Gynophore of fruit 11-16 mm. long; seeds 2.2—2.6 mm. long; branches ascend- ing to decumbent, to 7 dm. long, the plants up to 5 dm. tall; Durango to San als, Potosivand Guanajuato oe oe ee le ees 2. C. perennis BB. Gynophore of fruit 0.4-2.0 mm. long; seeds 1.9-2.0 mm. long; branches de- cumbent or prostrate, to 3 dm. long, the plants rarely over 1 dm. tall; Mexico, DAF and: northern, Pueblos sate Fe 3. C. mexicana 1. CLEOMELLA LONGIPES Torr. in Hook. Jour. Bot. & Kew Gard. Misc. 2:255. 1850; in Gray, Plantae Wrightianae 1:11. 1852 [Smithson. Contr. Knowledge 3 (Art. 5):11], pro parte, including the lectotype, Gregg 544!; excluding the Gregg San Luis Potosi specimen [herein cited under C. perennis Iltis]. Robust, glabrous, erect annual from an usually unbranched or little- branched taproot, 3—8 dm. tall, unbranched or more frequently with sev- 1956] ILTIS: CLEOMELLA 179 eral, strongly ascending branches, these usually scattered above the lower one-fourth of the erect main stem (leader), with only few branchlets, the main stem 2—6 mm. in diam.; petioles 7-30 mm. long, the petiolules 1—2 mm. long; leaflets oblanceolate to oblong-oblanceolate, acute to rounded or rarely emarginate, attenuate to the base from near the middle, 15-30 mm. long, 4-10 mm. wide, flat or conduplicate; racemes well defined, dense, greatly elongating in fruit, to 30 cm. long, bracteate or the upper portion ebracteate, the bracts 3-foliolate below, unifoliolate and reduced above; sepals connate at the base, triangular-lanceolate, acuminate to caudate-acuminate, 1.5-2.5 mm. long, tardily deciduous; petals oblong or lanceolate-oblong, obtuse to rounded, usually abruptly contracted into a short claw (5—)6-9 mm. long, 2-3 mm. wide; staminal filaments 8—12 mm. long; anthers 2.0—-3.5 mm. long; ovary exserted on long gynophore; mature silicles obdeltoid to rhomboidal, 4-7 (—11) mm. long, 6—12 mm. wide, the style 1-2 mm. long; gynophore (6—)10-17(—21) mm. long, usually yellowish; pedicel (5—)7—15(—18) mm. long; seeds (4—)8—16 (—19) in each silicle, obovoidal, 1.9-2.4 mm. long, 1.4-2.0 mm. wide, 1.0-1.2 mm. thick, when mature dark brown, with the prominent lip of the shorter claw yellowish. Chisos Mountains of southwestern Texas to the Chiricahua Mountains of southeastern Arizona, south to southern Chihuahua, Mexico, at eleva- tions of 2500-5000 feet, in saline, alkaline soils or sands of semi-deserts, at edges of playas, old lake beds, thickets, ravines, and valleys, sometimes in Tamarix-Prosopis association, flowering and fruiting from late April to late September. UNITED STATES: Arizona. County unknown: D. T. McDougal, s.n. (US) .? Cochise County: west of the Chiricahua Mountains, Wright 857 (MO) ; 3 miles west of Willcox, Darrow, Phillips & Pultz 1045 (US). NEw Mexico. Hidalgo County: Dog Spring, Dog Mountains, Mearns 2379 (US), Hershey s.n. (ILT, NMC). Texas. County unknown: Near J. Davis’ ranch, West Texas, Havard 135 (US). Brewster County: Chisos Mountains area, in ravine of Rio Grande valley, Sperry 136 (US) ; near Chisos Mountains, Young 101 (MO); Rio Grande valley near Terlinguas, Pal- mer 34215 (MO) ; 3 miles southeast of Chilocal Mountain, Marsh s.n. (F). Hudspeth County: Rio Grande near Indian Hot Springs, Waterfall 4856 (MO). Presidio County: Penitas Ranch, 24 miles south of Marfa, Hinckley 1050 (F); near Chinati Mountains, Hinckley 824 (F). MEXICO. CurtHuanua. Near San Pablo, April 29, 1847, Gregg 544 (MO); Rio Palotat near Janos, Schott s.n. (F); Casas Grandes, Goldman 436 (US); Colonia Diaz, Nelson 6434 (US). Torrey, in his original description of C. longipes, lists only two speci- mens, both collected by Dr. Gregg, one from Chihuahua and the other from San Luis Potosi. In the Missouri Botanical Garden Herbarium there are three Gregg collections, all of which can be considered isotypic, two of 544 labeled as from Chihuahua and clearly belonging to the northern taxon (C. longipes sensu stricto) and a third without any data except the 2 The herbarium abbreviations used in the species citations, except for ILT, which refers to specimens in my own study collection, are those proposed by Lanjouw and Stafleu, Index Herbariorum, Regnum Vegetabile 21:106—-117. 1952. 180 MADRONO [Vol. 13 collection number (579). The latter is evidently the specimen from San Luis Potosi mentioned by Torrey, for it agrees in every respect with the other collections from that state, which belong to the southern entity mentioned in the introduction (C. perennis). Torrey thus established C. longipes on two collections belonging to two distinct though closely re- lated taxa. His very generalized description offers no clue as to which of the two collections was primarily used by him in establishing C. longipes. Payson (1922), in his revision of Cleomella, likewise did not distinguish between the two taxa included in Torrey’s C. longipes, undoubtedly be- cause of the small number of specimens available to him for study. He did, however, choose the Chihuahuan Gregg collection as the lectotype, thus permanently associating the name Cleomella longipes with the plants from north-central Mexico and the adjoining United States. Kearney and Peebles (1942, p. 373) cite Wright 857, from the Chiricahua Mountains, as the type of C. longipes. This is an error, for the specimen was not cited by Torrey in the original description. The specimens from the Chisos Mountains, the easternmost station of C. longipes, apparently represent a local race characterized by wider, emarginate leaflets, stronger connation of sepals, shorter anthers, and lower number (4—5) of seeds, which are larger than average. 2. Cleomella perennis sp. nov. Herbae glabrae vel minutule papil- losae, perennes basi ramosae e caudice lignoso crasso; rami annui plus minusve suffruticosi, interiores erecti et ca. 15-50 cm. alti, exteriores de- cumbentes adscendentes, longitudine ad 70 cm.; folia densa, petiolis bre- vibus (2-10 mm.) ; foliola oblanceolata-cuneata 5-14 mm. longa, 1—4 mm. lata, apice rotundato vel emarginato; bractae racemi uni- vel tri-folio- latae, superne graduatim decrescentes; petala spatulata, 4.5-7.2 mm. longa, sepala triplo superantia; stamina et ovarium corollam duplo super- antia; siliculae maturae obdeltoideae vel deltiodeae, 6-12 mm. latea; gynophorum gracillimum 11-16 mm. longum; semina subgloboso-reni- formia, 2.2-2.6 mm. longa, 1.7—2.2 mm. lata, 3—7 per siliculam. Holotypus. C. G. Pringle, Plantae Mexicanae 3089, in U.S. Nat. Herb. No. 1418485 (‘“‘Saline Plains, Salinas, San Luis Potosi, Mexico, June 30, 1890’). Isotypi in Herb. BR, F, ILT, MO, MSC, SMU, et W. Slender, more or less suffruticose and bushy, glabrous or minutely pa- pillose perennial herbs, much branched mainly from the apex of the greatly thickened, woody caudex (short and to 2 cm. in diam.) the clus- tered and branched annual stems 1—2(—3) mm. in diam., the inner erect and 15—50 cm. tall, the outer ascending to decumbent and up to 70 cm. long, densely leaved and frequently with many, very short branchlets; petioles 2-10 mm. long; petiolules 1 mm. long or less; leaflets narrowly oblanceolate-cuneate, rounded to emarginate, mucronulate, gradually at- tenuate to the base from near the apex, 5—12(—14) mm. long, 1-5 mm. wide, usually strongly conduplicate; racemes well-defined, dense, greatly elongating in fruit, up to 25 cm. long; lower bracts trifoliolate and like the foliage leaves, much reduced and either trifoliolate or unifoliolate 1956] ILTIS: CLEOMELLA 181 above; sepals joined at the base, broadly to narrowly triangular, acumi- nate to caudate-acuminate, 1.3—2.2 mm. long, tardily deciduous; petals oblong to lanceolate-oblong or oblanceolate, obtuse, gradually tapering to the base, 4.5—7.2 mm. long, 1.8—-2.8 mm. wide; staminal filaments 10-14 mm. long; anthers 1.6—2.7 mm. long; ovary exserted on long gyno- phore; mature silicles obdeltoid, rhomboidal or deltoid, (3—)5-7 mm. long, 6-12 mm. wide; style 0.9-1.5 mm. long; gynophore 11-16 mm. long, usually dark purplish; pedicel 7-14 mm. long; seeds (3—)5—7 in each silicle, broadly obovoidal to subglobose, 2.2—2.6 mm. long, 1.7—2.2 mm. wide, 1.3-1.7 mm. thick, when mature brown to brownish-black, sometimes mottled, shiny, with the usually prominent lip of the shorter claw yellow-brown to whitish. Mexico, from central Durango east to central San Luis Potosi and northern Guanajuato, at elevations near 6000 feet, in saline plains, cién- ega bottomlands in grama grasslands (Gentry), steppes, ditches in grassy plains, and in sunny uncultivated fields; flowering and fruiting from late June into October (December). MEXICO. Duranco. City of Durango and vic., E. Palmer 1896-326 (F, ILT, MO, US) ; 40 miles north of Ciudad Durango, Gentry 8587 (US). ZAcATECAS. 9 miles south of Villa de Cos, on road from Sierra Hermosa southwest to Zacatecas, John- ston 7437 (F, US), Shreve 8623 (US). Saw Luts Porost. Salinas, Pringle 3089 (BR, F, ILT, MO, MSC, SMU, US type, W); Charcas, Lundell 5587 (US); San Fran- cisco, Gregg 579 (MO, cited by Torrey as C. longipes). GUANAJUATO. Jaral, Schu- man sit Clio IMEOUS) Schnee san) (ILY, P).. The Pringle collection was chosen as the type because of the widely distributed isotypes and the excellence of the material which includes not only flowers and nearly mature fruits, but also well developed caudices, so often left behind by most “hay-baling” collectors. Its bracts are 1- foliolate. Palmer’s Durango collection, also widely distributed, supple- ments the type by its mostly 3-foliolate bracts and fully mature seeds and fruits. The two collections from Zacatecas cited above have unusually large caudices. Cleomella perennis is intermediate between, and very closely related to, the preceding and following species geographically as well as morpho- logically. Thus the fruit and gynophore are nearly identical with those of C. longipes, the caudex and leaves with those of C. mexicana. In habit, size, erectness and most other attributes, C. perennis is exactly interme- diate. Only the larger seeds and the lower seed number are out of line. As will be discussed presently, it is probable that C. longipes gave rise to C. perennis, which in turn is ancestral to C. mexicana. 3. CLEOMELLA MEXICANA Moc. & Sessé in DC. Prodr. 1:237. 1824; A. DC. Calq. Dess. Fl. Mex. Moc. & Sessé, t. 19. 1874. Isopara mexicana (Moc. & Sessé in DC.) Raf. Atl. Jour. 1:144. 1832. Cleome mexicana (Moc. & Sessé in DC.) D. Dietr. Syn. Pl. 2:1068. 1840, non Cleome mexi- cana Hemsl. Bio. Centr. Amer. Bot. 1:41. 1879. 182 MADRONO [Vol. 13 Cleomella medicagineae Turcz. in Bull. Soc. Nat. Mosc. 277:313. 1854, ex char. (Type: Galeotti 7216). Physostemon medicagineum (Turcz.) Briqu. in Ann. Conserv. & Jard. Bot. Genéve, 17:390. 1914. Slender, herbaceous or somewhat suffruticose perennials from a stout taproot and caudex, 1—2 dm. tall, profusely branched, particularly from the base, the clustered branches strongly decumbent, 10—25 cm. long, 1—2 mm. in diameter; petiole 5—13(—19) mm. long; petiolules 1 mm. long or less; leaflets thick, cuneate-obovate, truncate to emarginate, gradually attenuate to the base from above the middle, 3-11 mm. long, 2—5 mm. wide, strongly conduplicate; racemes lax and ill-defined, not greatly elon- gating in fruit, the flowers borne near the tip of the branches in the axils of full-sized or slightly reduced 3-foliolate leaves; sepals barely joined at the base, lanceolate, acuminate, 1.3-1.7 mm. long, deciduous; petals broadly oblanceolate, rounded, gradually attenuate to a broad base, 3.5— 4.2 mm. long, 1.5—1.9 mm. wide; stamens included, the filaments 2-3 mm. long; anthers 1.2 mm. long; ovary included; mature silicles obdeltoid to rhomboidal, 3-5 mm. long, 6-9 mm. wide; style 0.3-0.5 mm. long; gyno- phore 0.4—2.0 mm. long, purplish; pedicel 8-11 mm. long; seeds 6—8 (—10?) in each silicle, obovoidal, 1.9-2.0 mm. long, 1.5—1.6 mm. wide, 1.1-1.2 mm. thick, brown, the lip of shorter claw not conspicuous. In saline plains, in the vicinity of Mexico City and Tepeyahualco, Pueblo, at elevations of 7000—7700 feet, flowering and fruiting from May to late September (December). MEXICO. Sessé, Mocino, Castillo & Maldonado 3356 (MA) ; between Vera Cruz and Mexico City, Halstead s.n. (MO). Mexico, D. F. Mexico City, Rutten & Rutten- Pekelharing 438 (ILT, U). Puresio. Prope Tepeyahualco, Schiede & Deppe s.n. (M.). Cleomella mexicana, the type of the genus, is fairly well illustrated in Sessé and Mocifio’s unpublished “Icones Florae Mexicanae,” which were copied by the ladies of Geneva for the elder de Candolle before the plates had.to be returned to Madrid (cf. Standley, 1920, p. 16). Tracings of these copies (A. de Candolle, 1874) are in the library of the Missouri Botanical Garden, Cleomella mexicana being plate 19. A Macbride photo- graph (No. 30454) of the original water color copy now at Geneva is in the herbarium of the Chicago Natural History Museum. Since the only thing de Candolle had at hand when he described the species was this Sessé and Mocino plate, it must be taken in lieu of the type. However, in the Instituto Botanico “Antonio José Cavanilles,’ Madrid, there is a sheet with abundant specimens of Cleomella mexicana collected by Sessé, Mocino et al. Judging from the description, Cleomella medicaginea Turzc. is clearly the same as C. mexicana. It is based on a collection of Galeotti’s from “planitie salsa prope urbem Mexico.” Cleomella medicaginea was later transferred by Briquet (1914, p. 390) to Physostemon (as P. medicagi- neum), a quite unrelated group, because he thought Turczaninow’s spe- cies to be identical with Hemsley’s Cleome mexicana (1879), and wanted 1956] ILTIS: CLEOMELLA 183 to conform to the rules of priority, for Hemsley’s Cleome mexicana indeed does belong to Physostemon, a Cleome segregate with unifoliolate leaves and apophysate stamens.* The exact localities where Cleomella mexicana has been collected have never been cited with the collections. It is almost certain though that the “salt-lake” Lago de Texcoca (alt. 2236 m.), east of Mexico City, which was drained in 1920, was the location for the specimens labelled as com- ing from Mexico City, and that the series of large salt-flats south and southwest of Tepeyahualco are likewise the origin of the collection la- belled as coming from that town. As a matter of fact, these are the only saline lakes reported on the Hoja Puebla (lat. 19°N., long. 97° 30’W.) topographic map (1:500,000; 1944) of the Direccion de Geografia, Me- teorologia e Hidrologia of Mexico. II. PHYLOGENY Evolution in the subfamily Cleomoideae seems to have progressed in many instances from types inhabiting mesic habitats to those growing in deserts or semi-deserts. Certain recurrent evolutionary trends represent not only evolution from the more primitive to the specialized morpho- logical type but, more specifically, the evolution of structures more close- ly in harmony with arid and/or alkaline habitat requirements. Similar trends occur in the New World not only in Cleomella but also in quite unrelated groups of species (e.g., the unifoliolate species of Cleome, in- cluding Physostemon), as well as in various groups of the Old World. The following list of evolutionary tendencies in the Cleomoideae ap- plies mainly to the genus Cleomella, especially the Mexican species, but would hold true for all its close relatives (viz. Cleome sect. Peritoma, Wislizenia, Oxystylis) and, with a little modification, for most other xero- phytic Cleomoideae as well. Some of these conclusions are at variance with present day phylogenetic points of view. A full explanation of the reasons for considering characters primitive or specialized in the Western North American Cleomoideae (exclusive of Polanisia) are given in a forthcoming study (Iltis, 1956) dealing with the morphology and phy- logeny of the species of Cleome Section Peritoma, Cleomella, Wislizenia, and Oxystylis. In Table 1 characters that occur in Cleomella longipes are indicated by (1), those that occur in C. perennis by (2) and in C. mexicana by (3). 3 Hemsley’s Cleome mexicana of 1879 (in Biologia Centrali-Americana, Botany 1:41) was invalid because preoccupied by Cleome mexicana (Moc. & Sessé in DC.) D. Dietr. 1840. (= Cleomella mexicana). To correct this, Bullock (in Kew Bull. Misc. Inf. 1936:388) renamed Hemsley’s species Cleomella hemsleyana Bullock, thus assign- ing it to the wrong genus. To place this taxon among its relatives, Foster (in Contr. Gray Herb. 155:58. 1945) transferred it to Physostemon [as P. hemsleyanum (Bul- lock) Foster]. In my opinion, it is not possible to segregate Physostemon from the unifoliolate New World species of Cleome. It therefore becomes necessary to return the invalid Cleome mexicana Hemsley to Cleome as C. hemsleyana (Bullock) Iltis, comb. nov. (Type: Galeotti 3194, K !). 184 OL ~r jig 238 24. MADRONO [Vol. 13 TABLE 1. COMPARISON OF PRIMITIVE AND SPECIALIZED CHARACTERS IN MEXICAN SPECIES OF CLEOMELLA AND RELATED TAXA A. Primitive . Annual with little branched taproot. (1) . Plants large (1-6 m.). . Stems simple at the base with unbranched main leader. (1) . Stems erect or strongly ascending. (1-2) . Leaves estipulate. . Leaves large, 3- to 13-foliolate. (1) . Leaves long-petioled. (1, 3) . Leaflets at apex caudate-acuminate to acute, at base long-attenuate. (1) . Leaflets lanceolate-elliptic. (1) . Leaflets thin, with many raised lateral nerves. . Leaflets more or less flat. (1) . Flower and fruit production after much vegetative growth. (1-2) . Racemes bracteate. (1-3) . Flowers in well defined terminal racemes. (1-2) . Bracts 1-foliolate. (1-2) . Post-floral elongation of raceme axis very pronounced. . Nectary small, smooth. . Petals large. (1-2) . Stamens and ovary long-exserted. (1-2) . Stigma large, capitate, sessile. . Fruits terete, linear-cylindric, large and long (5-25 cm.), many-seeded. Fruits with deciduous valves, and free-falling seeds. (1-3) Gynophore long. (1-2) Seeds with large internal sinus. B. Specialized Perennial with roots from a woody caudex or rootstock. (2-3) Plants small (1-100 cm.). (1-3) Stems much branched from the base without a main leader. (2-3) Stems decumbent to prostrate. (2-3) Leaves with minute “stipular” struc- tures. (1-3) Leaves small, 3- or 1-foliolate. (1-3) Leaves short-petioled to sessile. (2-3) Leaflets obtuse to emarginate at apex, cuneate to rounded at base. (1-3) Leaflets oval to obovate. (1-3) Leaflets thick, with only the midrib and few immersed, nearly obsolete lateral nerves. (1-3) Leaflets strongly conduplicate. (2-3) Flower and fruit production after little vegetative growth. (2-3) Racemes ebracteate. (rarely in 1) Flowers in the axils of cauline leaves (associated with 4B). (3) Bracts 3-foliolate or like cauline leaves. (2-3) Post-floral elongation of raceme axis small. Nectary relatively large, variously sculp- tured, etc. (1-3) Petals small. (3) Stamens and ovary included or nearly so. (3) Stigma minute, truncate or pointed, on slender style. (1-3) Fruits compressed, various, small and short (1-3 cm. or less), few-seeded. (1-3) Fruits with persistent valves separating only at apex, or with valves that per- manently enclose seeds. Gynophore short. (3) Seeds with obsolete internal sinus. (1-3) The somewhat unusual conclusion that the perennial woody caudex of the Cleomoideae is a derived or specialized structure is based primarily on the fact that in the Mexican species of Cleomella, where all traits ex- cept for minor exceptions (see below) follow interrupted clines from north to south, all of many characters (except for the problematic one of dura- tion) are in the primitive state in the northern C. longipes, in the special- 1956 | ILTIS: CLEOMELLA 185 C. LONGIPES C. MEXICANA oT Fic. 1. Principal cauline leaves of the Mexican species of Cleomella. Natural size. ized state in the southern C. mexicana, while the geographically interme- diate C. perennis is morphologically intermediate also (cf. figs. 1 and 2). Secondly, the reputed ancestors of Cleomella (see below) are all annuals. Thirdly, parallel situations exist in Wislizenia (Greene, 1906) and in 186 MADRONO [Vol. 13 C. LONGIPES ye aEsaey RE TES Fic. 2. Mature silicles and bracts of the Mexican species of Cleomella. Natural size. Each fruit from a different collection, except the lower left two, where the first shows the usual dimerous silicle, and the second the rare, abnormal trimerous fruit. several groups quite unrelated to Cleomella which leave little doubt that in certain arid climes evolution may proceed from the annual to the perennial habit. For example, in the phyletic sequence of the unifoliolate Cleomoideae (and relatives) of the New World, the primitive taxa (Cleome tenuis, C. stenophylla) are annuals, while the more specialized forms had evolved the perennial habit in at least two separate lines: 1) in the West Indian semi-desert species (C. wrighti, C. macrorhiza, etc.) and 2) in the highly specialized taxa commonly classified under the genus Physostemon, which are native to the arid “‘Caatinga” of northeast Brazil and to the southern Mexican grasslands. 1956] ILTIS: CLEOMELLA 187 Therefore, the perennial woody rootstock of these plants which, accord- ing to Bessey (1915) and other workers, would indicate a primitive struc- ture, must be thought of rather as a derived condition which had devel- oped in the arid, alkaline inter-montane plateau region of central and southern Mexico, and in climatically similar regions, as an adaptation to that particular habitat type, a development which occurred in various groups at various times through convergent evolution. .- perennis Fic. 3. Distribution of the Mexican species of Cleomella. (Base map Goode’s copy- righted map 112 with permission of the University of Chicago.) We can detect only two minor exceptions to the uniformly north-south sequences of nearly all morphological modifications: 1) in seed size, that of C. perennis being greater on the average than that of either of the two other species, and 2) in petiole length, the petioles of C. perennis being often shorter than those of either C. longipes or C. mexicana. From this evidence, it seems reasonable to assume that in Mexico Cleo- mella migrated from north to south (fig. 3). To emphasize this, it should be pointed out that Cleomella as a genus is clearly derived from Cleome sect. Peritoma (DC.) Baill., of which the annual, erect, 5-foliolate Cleome lutea of the western United States is probably most like the ancestor of Cleomella. Cleomella angustifolia of eastern Texas and Oklahoma, a tall, erect, 3-foliolate annual, is very similar to Cleome lutea, particularly veg- etatively. It is also very similar to Cleomella longipes, allopatric to the southwest, but has more acute, larger leaflets and larger, rugose seeds. These two appear to be the most primitive of all the 10 species of Cleomella. The allopatric distribution of the Mexican taxa (fig. 3) also supports the belief that these three species of Cleomella are part of a single phylad 188 MADRONO [Vol. 13 that originated in the Arizona-Texas region and migrated southward from there. This appears to me to be of considerable interest since the species of Cleomella are a typical part of the southwestern desert flora, the mem- bers of which are reputed to have migrated north from the Mexican pla- teau (Munz, 1935). While this has undoubtedly occurred in the case of the ancestors of the more primitive Cleome sect. Peritoma, the northward migration (at least here) was followed at a later and perhaps rather recent date by a southward migration of some of its more specialized descendants, namely the Mexican species of Cleomella. Two additional north-south trends are found in the Mexican species of Cleomella. One, the tendency for a southward decrease in distributional area is evidenced here by C. mexicana, the southernmost of the species of Cleomella, as compared with the more northern species (fig. 3). It occurs in the only two available saline areas in the region around Mexico City. Because of this narrow ecological restriction, it might be considered homo- genic (Stebbins, 1942). The second north-south trend is evidenced by the fact that the species of Cleomella occur at progressively higher alti- tudes towards the south: the northern C. longipes occurring at elevations of 2500-5000 feet, the central C. perennis at about 6000 feet, and the southern C. mexicana at 7000-7700 feet. This corresponds to the fact that given vegetational zones occur at higher elevations in lower latitudes. III. SUMMARY A new perennial species of Cleomella (C. perennis) is described from north-central Mexico. It is closely related to and intermediate between C. longipes to its north, an annual with larger leaves, and C. mexicana to its south, a perennial with low decumbent branches and smaller flowers. A preliminary list comparing primitive versus specialized characters in the Cleomoideae is presented, and the view is proposed that the perennial habit may, in some cases, be the specialized rather than the primitive con- dition. All data support the contention that C. longipes is most primitive and has given rise to C. perennis and C. mexicana. These species com- prise a phylad which probably had a northern origin and a southward migration. Since the group ancestral to Cleomella (Cleome sect. Pert- toma) originated from tropical stock that moved north to the region of the southwestern United States, the Mexican branch of Cleomella thus returned to Mexico on the route of its ancestors, but probably in drier habitats and at higher elevations. The southwestern desert flora which generally is considered to have originated from the flora of the semi-arid Mexican plateau, thus seems to have returned the favors to the latter by the contribution of some of its specialized, more recent derivatives. Department of Botany, University of Wisconsin, Madison LITERATURE CITED Bessry, C. E. 1915. The phylogenetic taxonomy of flowering plants. Ann. Mo. Bot. Gard. 2:109-164. 1956] KEENER: FUNGI 189 BRIQUET, J. 1914. Decades plantarum novarum vel minus cognitarum. Ann. Conserv. & Jard. Bot. Geneve. 17:326-403. De CaANDOLLE, A. P. 1824. Prodromus Systematis Naturalis Regni Vegetabilis. 1:237. De CANDOLLE, A. 1874. Calques des dessins de la Flore du Mexique, de Mocifio et Sessé, Volume 1. GREENE, E. L. 1906. Revision of the Genus Wislizenia. Proc. Biol. Soc. Wash. 29: 127-132. Ixt1s, H. H. 1955. Capparidaceae of Nevada. Contr. toward a Flora of Nevada. Bu- reau of Plant Industry, U.S. D. A. Washington, D.C. 1-24. . 1956. The phylogeny of the Western North American Cleomoideae. Ann. Mo. Bot. Gard. (Accepted for publication.) KerarRNEY, T.H. and PEEBLES, R.H. 1942. Flowering Plants and Ferns of Arizona. U.S.D.A. Misc. Publ. 423. Muwz, P. A. 1935. A Manual of Southern California Botany. Claremont, California. Payson, E. B. 1922. A synoptical revision of the genus Cleomella. Univ. Wyo. Publ. Sci. 1:29-46. STANDLEY, P. C. 1920. Trees and Shrubs of Mexico. Contr. U.S. Nat. Herb. 23, part 1. STEBBINS, G. L. 1942. The genetic approach to problems of rare and endemic species. Madrono 6:241-258. TWO FUNGI ASSOCIATED WITH A MICROCYCLIC RUST, COLEOSPORIUM CROWELLII CUMMINS, ON NEEDLES OF PINUS EDULIS ENGELM. IN ARIZONA! PAuL D. KEENER? The microcyclic rust, Coleosporium crowellii Cummins (Cummins, 1938) is unique among species of the genus because of the occurrence of the telial stage on species of Pinus. This species of Coleosporium is also regarded as autoecious. Other species of Coleosporium are macrocyclic with needles of certain pines serving as sites for the pycnia and aecia rather than the telia. In addition, species of Coleosporium are generally heteroecious. Material of C. crowellu on Pinus edulis Engelm. was collected one mile east of Yaki Point, on the north side of State Highway 64 along the South Rim of the Grand Canyon, Grand Canyon National Park, Coconino County, Arizona, on October 6, 1953. The telial sori were of an unnatural dull yellow-brown, rather than the usual yellow-orange. The sori of the rust fungus were found to be invaded by two non-uredineous fungi, Dar- luca filum (Biv.) Castagne and Cladosporium aecidiicola Thum. Both of the fungi were in their conidial phases. No previous reports of these two hyperparasites on this rust have been found. Materials for microscopic examination were prepared in the following manner: small portions of pine needle tissue containing rust sori invaded 1 Arizona Agricultural Experiment Station Technical Paper No. 364. 2 Assistant Plant Pathologist, University of Arizona Agricultural Experiment Station, Tucson. 190 MADRONO [Vol. 13 by the hyperparasites were placed momentarily in 95 per cent ethyl alco- hol, then in lukewarm water for several minutes until softening occurred. Free-hand sections were then prepared. The sections were mounted in 0.25 per cent Orseillin BB in 3 per cent acetic acid and modified Sartory’s solution. The Orseillin BB in acetic acid was added to the Sartory’s solu- tion (phenol, 10 cc.; lactic acid, 20 cc.; glycerine, 40 cc.; distilled wa- ter, 20 cc.) in the proportion of 1 part of the former to 9 parts of the latter. The technique is essentially the same as described by Alcorn and Yeager (1937) with slight modifications including adaptation to free- hand sections. DARLUCA FILUM (Biv.) Castagne Species of Darluca and especially D. filum have been reported as asso- ciated with numerous rust fungi (Uredinales) in many parts of the world. Most of the reports indicate that D. filum is more often found in uredinial and telial sori than in aecia. Also, there are more records of the associa- tion of this hyperparasite with macro- than with microcyclic rusts. The significant literature concerning the various genera of rusts reportedly susceptible to attack by D. filum in various parts of the world has been reviewed (Keener, 1934). In addition, previous investigations involving cross-inoculations with several isolations of D. filwm from numerous rusts showed that many species of rusts are susceptible to attack by one or more forms of the hyperparasite (Keener, 1934; 1952). Isolations of D. filum from infected rusts in the field, cultured on “Difco” Lima Bean agar, were used as sources of inocula in the previous studies (Keener, 1933; 1934). The studies confirmed the fact that both macro- and microcyclic rusts are susceptible to attack by D. filum, if suitable environmental con- ditions prevail. Although not previously emphasized (Keener, 1934) it was found that no great morphological differences existed among the vari- ous isolations of D. filum from field-infected materials, and the same isolates after inoculations on rusts other than those from which they originated. The same was true of isolations of the hyperparasite from field-infected materials when inoculated in the greenhouse onto rusts similar to those from which they had been isolated. The fungus found inhabiting telial sori of Coleosporium crowelliu at Grand Canyon agrees in most morphological respects with the isolations of D. filum from various rusts studied previously (Keener, 1934). Some interesting differences were noted. In most cases previously recorded, pycnidia of the hyperparasite have been described as being superficial with respect to the area occupied by the rust sorus. In the Grand Canyon material of D. filum on C. crowellii this was not always true. Many of the pycnidia of D. filum were found at the bases of the telial columns of the rust, in close association with the basal teliospores (fig. 1). Since the basal teliospores themselves are actually embedded in the pine needle meso- phyll, the pycnidia of D. filum are frequently invisible in the field even with the aid of a hand lens. Only at certain stages were pycnidia of the hyperparasite observed in the telial columns at a level with the needle 1956 | KEENER: FUNGI 191 (oxo fos WSS SSS: Sp? as SFT OS ORES YUE 7} | ; : Sarees Sarees > POUL (Lt f] Siva : CSA Se e.g eee a SY 5 .: ~ Lh o ii, Can ae: CREO AEA NS RNS ee REELED «WN = Sy eS ete — K Vy orga’. i : Ze CEES (pt Bare psy iets Vy Ngeowwed TIE aN Saws Le Eve - BY — , Si ieee 7 Fic. 1. Schematic representation showing two pycnidia of Darluca filum associ- ated with basal teliospores in the telial column of Coleosporium crowellit. Magnifica- tion approximately x 270. epidermis or above. The invisibility of the pycnidia prevails until after the collapse of the telial column, at which time they may be observed to be superficial with respect to the rust sorus and the pine needle tissues. The invisibility of the pycnidia for considerable periods of time may ac- count in part for the absence of any previous reports of D. filum associ- ated with C. crowellit. A peculiar feature of the invasion of telial sori of C. crowellu by D. filum is the manner in which the long, twisted, dark-brown, fragile spore threads of the hyperparasite reach the needle surface. These threads issue 192 MADRONO [Vol. 13 from the region between the bases of the telial columns and the ruptured pine needle epidermis (fig. 2, B). The threads contain the typical 2-3- celled macroconidia of D. filum (fig. 3, A). Some of the fragility may be SERN A x SSO ge See BSN) SHS DNS aM NZ ‘eE D plz rey KEUNG RS \\ SON ALS Ni 1 POG SON q PISS Phir | wu > }-—_| wu >] Fic. 2. Schematic drawings of telial columns of Coleosporium crowellii with asso- ciated fungi on needles of Pinus edulis. A. Telial sorus free of any invasion by hyper- parasitic fungi. B. Spore threads of the hyperparasite, Darluca filum, issuing from between the pine needle epidermis and the base of the rust telial column. C. Conidio- phores and other fungus structures of Cladosporium aecidiicola completely over- running telia of the rust. due to the prevailing dry atmosphere in Arizona. In spite of the desicca- tion of the spore threads, macroconidia germinated readily in sterile dis- tilled water on glass slides in from 4—6 hours at room temperatures. Often the spore threads which are best observed with a dissecting microscope are the only visible evidence of rust sorus invasion. Non-invaded sori show no thread-like strands (fig. 2, A). CLADOSPORIUM AECIDIICOLA Thum. Parasitism of rust fungi by species of Cladosporium has been reported in the past chiefly from the continent of Europe and the region of the Mediterranean. Some of the pertinent literature has already been re- viewed (Keener, 1954). Of interest are two previous reports of this hyper- parasite of rusts from the western United States (Keener, 1954; Smith, 1905). In the Grand Canyon material of Coleosporium crowellu invaded by Cladosporium aecidiicola no preference by the hyperparasite for a par- ticular region of the rust telial column was noted. The brown, septate hyphae and 1-—2-celled yellow-brown conidia of C. aecidicola (fig. 3, B), appeared to inhabit all parts of the telial sorus, frequently overrunning the entire column (figs. 2, C; 4). The telial columns inhabited by the 1956] KEENER: FUNGI 193 Fic. 3. Fungus structures of two hyperparasites associated with telial sori of the microcyclic rust, Coleosporium crowellii on Pinus edulis. A. Macroconidia, two of which show germ tubes, of Darluca filum. B. Conidiophores and conidia of Clado- sporium aecidiicola. Magnification approximately x 300. hyperparasite were dark-brown to black rather than the dull yellow-brown color of sori inhabited by D. filum. Due to their dark color, telial sori of Caleosporium crowellu inhabited by Cladosporium aecidiicola are easily recognized in the field, their pigmentation contrasting strikingly with the yellow-orange color of non-invaded ones. Sori invaded by C. aeciditicola appear to collapse sooner and more completely than do those attacked by D. filum. This is probably due to the complete invasion of the entire telial column of the rust by C. aeciducola. The morphological features of the fungus referred to as C. aecidicola XD , Per S. Rare eee ~ ar) Nw r ry rae iV Hatt Fic. 4. Diagrammatic drawing showing a single telial column of the microcyclic rust, Coleosporium crowellii on Pinus edulis, completely invaded and overrun by the vegetative hyphae of the hyperparasite, Cladosporium aecidiicola. Magnification 38. 194 MADRONO [Vol. 13 agree with those described for this species by other authors as well as with those of a similar fungus found inhabiting aecial sori of Puccinia conspicua (Arth.) Mains on Helenium hoopest A. Gray in southern Ari- zona (Keener, 1954). In October 1954, over forty specimens of Pinus edulis with needles on the lower branches attacked by Coleosporium crowellii were observed within the boundaries of Grand Canyon National Park, in the vicinity of Yaki and Moran Points. In spite of the below average rainfall during the preceding months, Cladosporium aecidiuicola was widespread in many of the telial sori of the rust. This hyperparasite appeared to be more gener- ally distributed in sori of Coleosporium crowellu than was D. filum. In 1954, D. filum was virtually absent in the Grand Canyon area on C. crow- ell. Recently material of Cladosporium aeciducola has been noted in aecial sori of Cronartium quercuum (Berk.) Miyabe on cones of Pinus cembroides Zucc. The hyperparasite has been isolated into pure culture from this material. One specimen of Pinus edulis was found at Grand Canyon which had sori of Coleosporium crowellu invaded by Cladosporium aeciducola and on the same or on different needles, fruiting bodies (hysterothecia) of Elytroderma deformans (Weir) Darker. The latter fungus is an Ascomy- cete and causes a needle-cast of conifers. The assistance in securing permission to collect the necessary materials as well as the continued interest in the study profered by: Mr. Louis Schellbach, Park Naturalist for Grand Canyon National Park, is hereby gratefully acknowledged. SUMMARY 1. Association of the hyperparasites, Darluca filum (Biv.) Castagne and Cladosporium aecidicola Thum. with telial sori of the microcyclic rust, Coleosporium crowelli Cummins on Pinus edulis Engelm., is re- ported for the first time. 2. Morphological similarities and differences between these associa- tions and similar though not identical ones recorded previously are discussed. 3. The investigation is based on materials collected in the vicinity of Yaki and Moran Points, South Rim of the Grand Canyon, Grand Canyon National Park, Coconino County, Arizona. Department of Plant Pathology, College of Agriculture, University of Arizona, Tucson LITERATURE CITED ALcorn, Gorpon D., AND C. C. YEAGER. 1937. Orseillin BB for staining fungal ele- ments in Sartory’s fluid. Stain Tech. 12 (4) :157-158. Cummins, GeEorcE B. 1938. A new microcyclic Coleosporium on Limber and Pifon pines. Phytopathology (Notes) 28(7) :522-523. illus. Keener, P. D. 1933. Some characteristics of Darluca in culture. Proc. Pa. Acad. Sci. 7:130-139. illus. Contribution No. 86 from the Department of Botany, The Pennsylvania State College [now The Pennsylvania State University, University Park, Pennsylvania ]. 1956 | WAGNER: ADIANTUM 195 . 1934. Biological specialization in Darluca filum. Bull. Torrey Club 61: 475-490. illus. Contribution No. 92 from the Department of Botany, The Penn- sylvania State College [now the Pennsylvania State University, University Park, Pennsylvania]. . 1952. The occurrence of Darluca filum (Biv.) Castagne on Puccinia sorghi Schw. United States Dept. Agric. Plant Dis. Reptr. 36(10) :384—385. illus. . 1954. Cladosporium aecidiicola Thuem. and Tuberculina persicina (Ditm.) Sacc. associated with Puccinia conspicua (Arth.) Mains on Helenium hoopesii A. Gray in Arizona. United States Dept. Agric. Plant Dis. Reptr. 38(9) :690-694. illus. SmitH, Rate E. 1905. Asparagus and Asparagus rust in California. Calif. Agric. Exp. Sta. Bull. 165, pp. 1-99. illus. A NATURAL HYBRID, & ADIANTUM TRACYI C. C. HALL! WarREN H. WAGNER, JR. A plant intermediate between the two common maidenhair ferns of California, Adiantum jordanu K. Mull. and A. pedatum L., was first dis- covered over a half-century ago, but except for several brief references in British gardeners’ publications (Stansfield, 1927; Macself, 1947; Logan, 1948), notes on it have never been published. The intermediate plant has now become established in horticulture both in California and in Eng- land, and is notable for its vigorous growth. Although the spores of this fern appear to be inviable, the rhizomes are capable of bearing numerous lateral shoots, enabling single plants to form large patches in gardens and to be readily propagated. The fact that the plant described here is evidently intermediate be- tween A. jordanu and A. pedatum in its obvious features is a strong point in favor of interpreting it as a hybrid between them. This conclusion is also supported by the following facts: the intermediate plant is rare and sporadic in distribution as compared with the much more abundant and widespread parents, and it has been found wild in only three counties of California; the supposed parents were present in the immediate vicinity of the putative hybrid; the sporangia of the intermediate are abortive and its spores are of irregular sizes and shapes. With such evidence indicative of hybrid origin I believe that it would be erroneous to name and describe this fern as a normal taxonomic species. At the present time I know of only ten places where this plant is in cul- tivation but there are probably many more. In view of its luxuriant growth and graceful appearance it is likely to become widely distributed in horticulture. For this reason the information that has been gathered on the history of its discovery and spread in horticulture should be useful. The intermediate fern was first found in September, 1895, by the late Joseph Prince Tracy, a well-known botanist of Eureka, California, whose 1 Study made during the tenure of a Summer Faculty Research Fellowship of the Horace H. Rackham School, University of Michigan. 196 MADRONO [Vol. 13 collection was recently bequeathed to the Herbarium of the University of California (biography, Bacigalupi, 1954). The fern—a single plant—was growing in redwoods along the Eel River, near the town of Pepperwood. Dr. Walter C. Blasdale has written (letter, August 27, 1953) as follows: “Only about three years ago I happened to meet Mr. Tracy and talked with him about it. He assured me that he found a small patch of it not far from plants of A. pedatum and A. emarginatum [syn. A. jordani| and had never found other specimens of it anywhere else.” The specimen which was taken by Mr. Tracy was grown in his garden for many years and gave rise to new rhizomes. One of these he presented to the late Mrs. Carlotta C. Hall, who considered the plant a hybrid and had long planned to name it in honor of Mr. Tracy. She was particularly interested in the variability of the branching shown by the leaf blades. The descendants of Mrs. Hall’s plant produced a luxuriant clone in her garden at Berkeley, and she contributed living specimens to the Strybing Arboretum in Golden Gate Park, San Francisco, as well as to friends, including Dr. Blasdale, who found it a remarkably fine fern for out-of-door cultivation. There is evidence that this apparently hybrid fern has arisen independ- ently in the wild on four other occasions. For information regarding sev- eral of these I am indebted to Mr. Hugh B. Logan. He possesses living specimens from four different original localities growing in his garden at Inverness, Marin County, California, and a number of the following notes are taken from his letters about them. Mrs. Mabel Bishop, the daughter of Mrs. Frances Mason, around 1910-1915, discovered plants morphologically like Mr. Tracy’s growing near Guerneville on the Russian River in Sonoma County. Mrs. Mason subsequently presented living material of one of these to Mr. Carl Purdy of Ukiah, California, who then included this fern in the many native plants he sold. It was Mr. Purdy who introduced the plant in England; he sent it originally to Mr. Amos Perry of Enfield, Middlesex, England, who distributed it. The fern proved to be hardy there and was called “Adiantum flabellulatum,” but was considered even then to have origi- nated by hybridization—“possibly a cross between A. pedatum and A. capillus-veneris, and consequently a sort of British-American hybrid”’ (Stansfield, 1927). The fern was described as being similar in habit to A. pedatum, but the rhizome is less wide-creeping and the plant is thor- oughly evergreen. According to Mr. Logan, who has sent specimens to relatives in Wales, it can withstand the English winter. This fern was also, according to further information supplied by Mr. Logan, discovered in Bear Valley, Marin County, California, where Mrs. Slate Weston removed a plant and transferred it to her garden. From her Mr. George Syme obtained living specimens, and he subsequently re- turned to the Bear Valley area himself to search for the plant 7m situ but had no success in finding it. Mr. Syme cultivated the plant at Petaluma, Sonoma County, and sent a number of fine fronds to the Smithsonian In- stitution in 1919. Plants from the Bear Valley stock were included in Carl 1956 | WAGNER: ADIANTUM 197 as LS co Pi Luter es yi BSS SE RES oe UPSTER: STATES NS UCI SE MERRS RUE GRIPE BIBT EDS NETOANSL MUSEMBS 4 2 EGER ER GS EES een R Fic. 1. x Adiantum tracyi C. C. Hall. A perfect frond of the plant found by Mr. J. Neeman near Garberville, Humboldt County, California. Purdy’s collection, and some of these were sent, along with the Russian River plants, to Mr. Perry in England. Mr. Logan has also written me that he possesses additional plants which came from the south fork of the Gualala River, Sonoma County, and which were given by an Indian girl to Mrs. Celestine Levens of San Rafael. In 1948, Mr. James Neeman of Garberville, California, sent a small but unmistakeable frond of the hybrid to the National Herbarium (illus- trated in fig. 1) and he has recently (letter, April 25, 1954) written of the circumstances of finding this specimen as follows: “‘Regarding the Adiantum hybrid I discovered in Humboldt County, it grew about a mile and a half north of the town of Garberville. .. . The area is cut by small streams and high hills. In the deep, cool, shady gulches one finds red- woods, pepperwoods, and on the drier slopes, fir, madrone, and oak. I 198 MADRONO [Vol. 13 was following the stream upwards in a deep glen where woodwardias, Polystichum munitum, and an occasional P. californicum grew, the gold- back [Pityrogramma triangularis| with Adiantum jordaniu covering mossy slopes on the drier slopes and facing north on a sandstone cliff, I saw five-fingers [Adiantum pedatum| near by. Among them, I could dis- cern the triangular forms of the hybrids, all told, nine of them, mostly small plants.” Mr. Neeman has written also that “peculiarly, the fern is almost ever- green, unlike its parents.” Adiantum jordanu, on the contrary, holds its foliage for only a few months and then dries down completely; it is usu- ally an inhabitant of dry situations, and sends up its fronds in the spring before A. pedatum. In this respect, the intermediate plant resembles A. jordanu more and the “hybrids start in the spring before A. pedatum shows any growth.” (Logan, letter, April 10, 1954). The hybrid fern also seems to be quite resistant to injury, and Mr. Neeman informs me that three plants, each growing in a three-gallon can, were scorched to the ground when his house burned down, but since then they have begun to come up again and are now growing well once more. I wish to thank the persons mentioned for the information given above concerning this plant’s history. I am indebted also to Mr. Conrad V. Morton of the U.S. National Herbarium and Dr. Herbert L. Mason and Miss Annetta Carter of the University of California Herbarium, for lend- ing dried specimens. To facilitate this study, Mrs. R.S. Niccolls, daughter of Mrs. Hall, kindly collected from her mother’s Berkeley garden a num- ber of dried leaves and a living rhizome. Adiantum tracyi C. C. Hall, hybr. nov. Adiantum A. jordanio A. pedatoque intermedium, laminae foliorum late deltoideae, e duabus pinnis basalibus lateralibus prominentibus 16 (12-23) cm. longis atque pinna centrali paulo longiore plerumque constans, pinnis in segmenta ultima impariter lunata bis divisis; segmenta 17 (12-21) mm. lata latere anteri- ore 8 (6-10) mm. longo, latere basali 14 (8-22) mm. longo, segmenti pedicellis 2.6 (1.5-4.0) mm. longis, soris tribus ad decem omnibus in segmentis, 4 (1-10) mm. longis, sporis abortivis. A terrestrial fern of moderate size; stem a branching rhizome, 6 (4-8) mm. thick, of indefinite growth but dying off basally; roots up to at least 5 cm. long, borne around the leaf bases, and bearing numerous branches at irregular intervals, the whole root system covered with pale-yellow hairs 1 mm. long; stem apex and sides protected by shiny, dark reddish- brown, elongate-triangular paleae, 1-2 mm. long, 0.5—1.0 mm. broad at base and narrowing to a filiform apex; leaves 2—3 times divided, long- petiolate, 44 (25-60) cm. in total length, the petioles 21 (13-29) cm. long and 1.0—-1.5 mm. thick, the blades deltoid, 23 (20-30) cm. long and 25 (15-35) cm. wide; petioles lustrous purplish or maroon-black and naked except for a few paleae remaining at the base; blade subhorizontally ori- ented when alive, with 3—7 large lateral pinnae, but the two basal ones usually much larger, 16 (12-23) cm. long, than those above, the frond 1956] WAGNER: ADIANTUM 199 thus appearing to comprise usually 3 major pinnae, the two laterals roughly one-half to two-thirds as long as the whole blade and the one central pinna; basal lateral pinnae provided with one or two long basi- scopic segments each; rachis lustrous maroon-black, conspicuously angu- lar at the origins of the two large basal pinnae, but becoming nearly straight distal to them; tertiary segments unequally lunate, averaging 17 (12-21) mm. across at the widest point, the inner (or anterior) of the two sides 8 (6-10) mm. long and the lower (or basal) side 14 (8-22) mm. long; stalks of the segments hair-like but rigid, 0.1 mm. thick, 2.6 (1.5- 4.0) mm. long, lustrous black; lamina dull-green, glabrous; venation subflabellate, free, the veins forking 2—5 times; sterile margins of the lamina finely toothed with occasional broad sinuses 1—2 mm. in depth at intervals of 2-10 mm.; sori marginal, 3-10 per segment, with whitish, thick, false indusium averaging 4 (1-10) mm. in length and 1 mm. in breadth; sporangia containing spores of irregular size and shape. Type specimen. “Adiantum. Probably hybrid jordani x pedatum. Original single plant found near Pepperwood, Sept. 1895, from which this is propagated in garden at Eureka. Sept. 16, 1924.” Joseph P. Tracy 6868b (Herb. Univ. Calif. no. 249775). Specimens examined. CALIFORNIA. From near Pepperwood, Humboldt County, cultivated at Eureka, 7vacy 6868a, 6868b (UC), cultivated at Berkeley, Niccolls (UM); from near Garberville, Humboldt County, cultivated at Garberville, Neeman (US, no. 1916165); from south fork of Gualala River, Sonoma County, cultivated at Inverness, Logan (US, no. 2084573); from near Guerneville, Sonoma County, cultivated at In- verness, Logan (US, no. 2081080); from Bear Valley, Marin County, cultivated at Petaluma, Svme (US, nos. 982424, 982425, 983776, D337 1) x This fern may be distinguished from Adiantum pedatum by the over- all shape of its frond: in mature blades of A. pedatum there are normally 5—9 major divisions of the blade and the blade is broadly fan-shaped; in x A. tracyi there are usually only three major divisions, the two lateral pinnae and the somewhat longer central pinna, and the blade is more nearly triangular. Also the sori of the hybrid average two to three times as long as those of A. pedatum. From A. jordanii the hybrid is most read- ily separated by the form of the blade also: in A. jordani the leaf is broadly lanceolate and pinnate in construction and the pinnae are gradu- ally reduced in size from base to apex; in X A. tracyi the blade is rela- tively much broader at the base and it usually shows a rather sudden reduction in pinna length from the basal pinnae to those above. The pin- nae of A. jordani are almost perfectly flabellate while those of the inter- mediate have a strongly developed lower margin about twice as long as the inner margin. The fern described here appears actually to be intermediate between the two presumed parental species in practically every respect. To show this, the writer took the fifteen best-developed leaves available of the 200 MADRONO [Vol. 13 hybrid plant and an equal number of mature specimens of each parent and measured them for some obvious characteristics to give a picture of the quantitative conditions. The results of these measurements, in terms of averages and ranges for each character, are shown in Table 1. Whole TABLE 1. COMPARISON OF & ADIANTUM TRACYI WITH ITS HYPOTHETICAL PARENTS A. jordanii x A. tracyi A. pedatum Length of Blade (cm.) 19 (11-25) 23 (20-30) 22 (15-35) Length of Petiole (cm.) 14 (6-20) 21 (13-29) 29 (15-40) Blade/Stipe Length (%) 136 109 76 Length of lower side of segments (mm.) 9.3 (5-15) 13.6 (8-22) 20.0 (13-25) Length of inner side of segments (mm.) 7.1 (4-11) 7.5 (6-10) 7.6 (6-10) Lower side/inner side length of segments (%) 131 181 263 Length of segment stalks (mm.) 4.1 (1.5-8.0) 2.6 (1.5-4.0) 0.7 (0.0-1.5) Length of proximal sorus (mm.) 4.8 (2.0-10.0) 4.2 (1.5-9.0) 1.7 (0.5—4.0) Distance between segment attachments (mm.) 15.8 (11-21) 12.4 (10-16) 8.9 (5-12) leaves were measured for blade length, petiole length, and blade-petiole ratios. The differences in segment shapes of the three taxa could be ex- pressed clearly in respect to lengths of inner margins and basal margins and the ratios of these two lengths. As shown in Table I, the inner (or anterior) and lower (or basal) margins in A. jordaniu are nearly equal in length while in A. pedatum the lower margin is almost three times the length of the inner margin. The stalks of the segments in A. jordanw are much longer than those of A. pedatum. The greater remoteness of the seg- ments in A. jordanii as compared to the other parent and the hybrid was determined by measuring the distance between segment stalks along the top side of the middle of the largest basal pinna. For uniformity the sori of the three taxa were measured on two central segments of the largest basal pinna, and for each segment the proximal sorus alone was used. In all these characteristics measured and averaged, the hybrid turned out to be intermediate, except for blade length which is, however, approximately the same in well-developed leaves of all three entities. With the exception of the terminal segments of each pinna, the seg- ments (ultimate pinnules) of A. jordani and of A. pedatum, as pointed out above, are conspicuously different in their outlines (fig. 2,cf.aandc). Copeland (1947, p. 78) in describing the genus Adiantum writes as fol- lows: “lamina typically broad, and pinnately decompound with dimidiate or flabellate pinnules.”’ The present hybrid fern thus combines the two characteristic pinnule forms of the genus, the “dimidiate” (i.e., appearing as if one half of the pinnule is lacking, with a strong vein running along the basal margin) of A. pedatum, and the “‘flabellate” (i.e., fan-shaped, 1956] WAGNER: ADIANTUM 201 _ PD ‘ Yr Fic. 2. Comparison of segment structure. A. Adiantum pedatum: A;, Cheboygan County, Michigan, Ehlers 1560; As, Delta County, Michigan, Grassl 7232 (both UM). B. X Adiantum tracyi (A. jordanii K A. pedatum): By, near Garberville, Humboldt County, California, Neeman; Bo, Syme Garden, US no. 982424; Bs, Logan Garden, US no. 2081080 (all US). C. Adiantum jordanii: C1, La Jolla, California, Clements 1914; San Benito County, California, Wiggins & Ferris 9353 (both UM). without a clear costulate development at any point) of A. jordani. In the dimidiate segment of A. pedatum there is vaguely evident a costule (the “midrib” of the segment) which runs along the proximal side of the pin- nule and from which the secondary veins originate, all of the secondary veins directed toward the distal side. But in the flabellate type of A. jor- dani, no costule or main vein can be detected at all, and the segment thus appears to be entirely midribless—that is, the segment is entirely dicho- tomous in vascular structure. The question of which of these pinna types represents the more basic condition in the genus Adiantum is one which seems not to have been dealt with before, and it would be inappropriate here to discuss the problem in any detail. However, it should be noted that there are indeed species of Adiantum which do possess centrally costate segments, i.e., segments with a more or less central axis giving rise to veins equally on both sides (e.g., 4. dolosum with free veins and A. olivaceum with reticulate veins). And there are numerous species in which 202 MADRONO [Vol. 13 a progressive shift in costule position from the central position to the proximal side of a dimidiate pinna can be illustrated (e.g., the series of A. latifolium, A. villosum, A. caudatum, A. curvatum, and A. klotschia- num). It seems perfectly possible that the flabellate condition of A. jor- dani could have been derived independently of the dimidiate type of A. pedatum in evolution from a basic and primitive pinnule-type in the genus in which the costule was centrally located. On the other hand, the dimidi- ate type could equally well have given rise to the flabellate type by way of an intermediate stage such as that exemplified in the pinnule form of x A. tracy. There are other normal species in the genus in which the pinnules are similar to those of * A. tvacyi and may be looked upon either as lop-sided flabellate segments or as shortened and broadened dimidiate segments. The most interesting morphological ‘aspect of * Adiantum tracyt is its leaf architecture. This combines the pinnately-constructed leaf of A. jor- dani (an over-all leaf type which characterizes practically all of the some 200 species of the genus) with the uniquely organized leaf of A. pedatum. The latter species, which in California is commonly called the “‘five-finger fern” in reference to the division of its leaves, possesses blades distin- guished by ‘“‘stalks forked at the top, the outwardly curving forks each bearing 3 to 8 pinnae...” (Jepson, 1923). As was first pointed out by Slosson in 1906, the seemingly dichotomous leaf form of A. pedatum (an extremely rare over-all leaf form known only in this species, in the Old World A. hispidulum, and in the Mexican A. patens) is in reality a much- modified pinnately constructed leaf. I have designated this (1952) the “‘displaced-midrib type of dichotomous leaf,” because of the peculiar ori- entation of the morphological rachis. While superficially most of the leaf blades seem to have regularly dichotomous major axes, they are actually asymmetrically constructed, the true midrib being present but modified in relative development and orientation as compared with fern leaves typically in which the rachis is simply a more or less straight, central axis of the frond. In the juvenile leaves of A. pedatum, as shown by a heteroblastic series of progressively larger leaves produced on a maturing rhizome, the early ones are pinnate in form like the mature leaves of A. jordani and they have a typical strong central axis (fig. 3, c). But in pro- gressively larger leaves of A. pedatum one of the basal lateral pinnae be- comes enlarged and re-oriented, and the rest of the blade, including the midrib, is correspondingly changed so that the enlarged basal pinna finally equals the rest of the blade; it becomes practically a mirror image except for the extra pinnule on the pinna shank of the dichotomy. Trac- ings of the axes of blade bases of A. pedatum (fig. 3,a), X A. tracyi (fig. 3, b), and A. jordanu (fig. 3, c) made from dried herbarium specimens show that the intermediate fern possesses blade construction which close- ly matches an intermediate stage in the heteroblastic series of leaves of A. pedatum (cf. Wagner, 1952, fig. 3, the two middle rows of leaves). The study of this hybrid, then, emphasizes the differences between leaves of 1956 | WAGNER: ADIANTUM 203 Fic. 3. Tracings of the major axes of the blade bases in pressed herbarium speci- mens. A. Adiantum pedatum: Ay, Branch County, Michigan, McVaugh 11298; Ag, Charlevoix County, Michigan, Reis 652A; As, Houghton County, Michigan, Richards 927; A4, Alpena County, Michigan, Grassl 7230 (all in UM Herb.). B. x Adiantum tracyi (Adiantum jordanit « A. pedatum): By—-B4, George Syme Garden (original plants said to have come from Bear Valley, Marin County, Cali- fornia) Syme (specimens in US Herb.) ; Bs5—Bz, C. C. Hall Garden, Niccolls (UM) ; Bs, George Syme Garden, Eastwood (US). C. Adiantum jordanii: C1, Oakland Hills, Lemmon; Co, Cupertino, California, M. E. Jones; C4, Cupertino, California, Fisher; Cs, near San Francisco, California, Harrington (all UM). 204 MADRONO [Vol. 13 different heteroblastic stages. In A. pedatum, presumably, specific genes, different from and (or) in addition to, those of the genus as a whole, must be present which control the formation of a unique leaf type only in the mature state. In the juvenile state these controlling factors show no effect, and the basic leaf construction is then like that of A. jordanii. In the hy- brid fern, the successively larger leaves of juvenile plants must pass through a series like that of A. pedatum but the heteroblastic series lead- ing to the peculiar form of the latter can apparently proceed no further than approximately half-way because of the influence from 4. jordanit. The degree of variation of leaf form seen in X A. tracyi considerably exceeds that of the parents. Mrs. Hall some years ago called to my atten- tion that no two fronds are alike on one plant; she had discussed a plan at that time to save a large series to show the complete range of variation but it apparently never materialized. A single plant of < A. tracyi may show in the variation in architecture from leaf to leaf the same sort of irregularity that seems to characterize other hybrid ferns in which the parents differ widely in specific foliar features, as was found recently in the case of Polystichum acrostichoides X P. lonchitis from the Bruce Peninsula (Wagner and Hagenah, 1954), in which certain microscopic details of venation proved to be irregular, and in several examples of hybrids among the species of Appalachian Asplenium in which the lobula- tion of the leaves is irregular (Wagner, 1954). SUMMARY A plant intermediate between Adiantum jordani and A. pedatum has been found growing wild in three counties of California. It has proved vigorous and hardy in horticulture and reproduces readily by rhizomes. The plant is described here as a hybrid taxon, X Adiantum tracyi C. C. Hall, because of its intermediacy in a number of obvious characters, its sporadic distribution in nature and its association there with the parents, and the abortion of its spores. The leaf segments combine the ‘“‘dimidiate” type of A. pedatum with the “‘flabellate” type of A. jordani. The over-all architecture of the leaf blade is of morphological interest in combining the “‘displaced-midrib” type found in A. pedatum with the more ordinary pinnately-constructed blade type of 4. jordaniz, but the branching of the rachis is variable. Department of Botany, University of Michigan, Ann Arbor. LITERATURE CITED BacicaLuPt, Rrmo, 1954. Joseph Prince Tracy. 1879-1953. Madronio 12:190-192. CopELAND, E. B. 1947. Genera Filicum. Waltham, Mass. Jepson, W.L. 1923. A manual of the flowering plants of California. Associated Stu- dents Store, University of California, Berkeley. Locan, Hucu B. 1948. A hardy maidenhair. English Gardening Illustrated. January. Macsetr, A. J. 1947. A hardy maidenhair. English Gardening Illustrated. November. SLosson, Mary, 1906. How ferns grow. Henry Holt & Co. New York. 1956] DOCUMENTED CHROMOSOME NUMBERS 205 STANSFIELD, F. W. 1927. A new hardy Adiantum. British Fern Gazette, New Series 5 (no. 9) :195-196. Wacn_er, W.H. 1952. Types of foliar dichotomy in living ferns. Am. Jour. Bot. 39: 578-592. . 1954. Reticulate evolution in the Appalachian aspleniums. Evolution 8: 103-118. Wacner, W. H. and Date J. Hacenau. 1954. A natural hybrid of Polystichum lonchitis and P. acrostichoides from the Bruce Peninsula. Rhodora 56:1-7. DOCUMENTED CHROMOSOME NUMBERS OF PLANTS (See MaproNo 9:257-258. 1948.) SPECIES NUMBER COUNTED BY COLLECTION LOCALITY LILIACEAE Furcraea n= 30 |H. Arnott & Arnott 54, UC1/Univ. Calif. *Roezlii André M.S. Cave, UC! Campus, Berkeley, ; California Nolina *Beldingu n=19 |M.S.Cave,UC |Carter & Ferris |San Francis- Brandegee 3331, UIC quito, Baja Calif., Mexico ROSACEAE Purshia 2n—18 |M.S.Cave,UC |(Fromseed) |Janesville, *tridentata DC. B. Pearson Lassen County, California EUPHORBIACEAE Argythamnia mica 13° /F, Chisaki & Ingram 227, |Gila Mountains, *Brandegei var. J. Ingram, UC UC Yuma County, intonsa (Jtn.) Arizona Ingram SOLANACEAE Acnistus n= 12 |C.B. Heiser, Jr., |Heiser 3578, |Turrialba, arborescens IND! IND Prov. Cartago, (L.) Schlecht. Costa Rica Capsicum n= 12 ss Heiser 3483, |La Lola, Prov. macro phyllum IND Limon, Costa (HBK.) Stand. Rica maculatum Ty = 2 es Heiser 3534 ba Paz, prow. Standl. & Morton IND Heredia, Costa Rica steno phyllum i al2 Heiser 3535 7 Morton & Stand. IND stramoniifolium 17 S Hetser 3575 San Jose, Prov. San José, Costa (HBK.) Stand. *Prepared slide available. 1 Symbols for institutions are those listed by Lanjouw and Stafleu, Index Her- bariorum, Part I. Second edition, 1954, Utrecht. IND Rica (continued on p. 206) 206 SPECIES NUMBER MADRONO COUNTED BY Capsicum tetramerum Standl. & Morton Lycianthes multiflora Bitter Solanum Cervantesii Lag. ciiatum Lam. jamaicense Mill. parcebarbatum Bitter salvifolium Lam. torvum Swartz umbellatum Mill. LOBELIACEAE Clermontia *oblongifolia Gaud. Cyanea *angustifolia Hillebrand Laurentia *carnosula (H.& A.) Gray COMPOSITAE Baileva *multiradiata Harv. & Gray Brickellia * peninsularis Brandegee var. am phithalassa Robinson Perityle * SOCOrrOSENSIS Rose Tetramolopium *humile Hillebrand ni==s17 Dine 22 COLLECTION C.B. Heiser, Jr., IND S. Carlquist, UC Heiser 3470 IND Heiser 3514 IND Heiser 3491 IND Heiser 3516 IND Heiser 3518 IND Heiser 3510 IND Heiser 3509 IND Heiser 3511 IND Heiser 3632 IND Carlquist H4 UC Carlquist H5 UC Mason 14439 WiC Carlquist 323 UC Carlquist 382 UC Carlquist 368 UC [Vol. 13 | LOCALITY La Carpintera, Prov. Cartago, i Costa Rica Turrialba, Prov. Cartago, Costa Rica Robert, Prov. Cartago, Costa Rica Turrialba, Prov. Cartago, Costa Rica Mt. Olympus, Oahu, T. H. Sierra Valley, Sierra County, Cali: San Marino, Los Angeles County, Calif. Socorro Island, Mexico Carlquist H 18 |Haleakala, UC Maui, T. H. Department of Botany University of California, Berkeley 1956] NOTES AND NEWS 207 NOTES AND NEWS Terms Usep To DESIGNATE TyPE MATERIAL. Although the principal terms used to designate type material are sufficiently defined in the International Code of Bo- tanical Nomenclature as adopted in 1950 (Regnum Vegetabile v. 3, 1952, pp. 17-18), there is still laxness among botanists in the application of these terms. For those who do not have access to that volume, the definitions there given may be repeated here with some legitimate modifications in wording, and a few collateral ones added. A holotype (generally called “type”) is the single specimen on which the descrip- tion of a new taxon is based; the term is also somewhat loosely applied to the basic unit of preparation (sheet or packet of macroscopic specimens, slide of microscopic) on which a taxon is based. In the strict sense such units, if consisting of more than one specimen, are composed of syntypes. A syntype (originally cotype, a term still used by some authors) is each of two or more specimens on which a taxon is based by an author who does not designate a holotype. A lectotype is a single type selected by a subsequent author from among syntypes. By definition it has to be part of the original material. A paratype is each specimen other than the holotype (or syntypes) cited in the original description by an author who designated a holotype (or syntypes). An isotype is a duplicate of the holotype (duplicate in the sense that it is a part of the collection [e.g., Pringle 8248] on which a taxon was based). Even the most careful collectors occasionally distribute more than one entity under a given number, and an isotype should always be carefully compared with the original description before basing conclusions on it. A neotype is a substitute type selected by a later author in a case where he has convincing proof that no holotype, syntype, paratype, or isotype is in existence. Any material can be selected as a neotype, but it is best to select a topotype, if practicable. A topotype is a specimen (other than the type itself) from the type locality. It is not actually a type in any proper sense, but, especially in zoology, it may have the study value of a duplicate type. The terms isosyntype, isolectotype, isoparatype, and isoneotype are available to designate duplicative material of the specimens that served for syntype, lectotype, paratype, and neotype respectively. The most nearly complete glossary of terms relating to type material is given by D. L. Frizzell in American Midland Naturalist 14:637-668. 1933, and there is a dis- cussion of some of the commoner terms of most use in botany by the writer in Rhodora 45:481-485. 1943. With the above definitions in mind it can be seen that in the two papers on Cucurbitaceae by K. M. Stocking published in numbers 3 and 4 of this volume of Madrono, the term lectotype is used in the following places to designate specimens that are actually neotypes: on p.96 (Echinopepon minimus), p.126 (Marah wat- soni), p.130 (M. fabaceus var. agrestis), p.132 (M. macrocarpus). On p. 86 the so-called lectotype of Echinocystis lobata should have been designated instead as a neotype but has no validity as such since Michaux’s type is still in existence. The so-called “types” of Echinopepon confusus (p.90), E. nelsonii (p.92), Marah major (p.134), and apparently also Echinocystis scabrida (p.130) are lectotypes.—S. F. Brake, Agricultural Research Service, U.S. Department of Agriculture, Beltsville, Maryland. Some publications of interest follow: Drawings of British Plants, by Stella Ross-Craig. Part VII. Leguminosae. 76 pls. 1954. 12s. net. Part VIII. Rosaceae(1). 40 pls. 1955. 8s. 2d. net. G. Bell and Sons, Ltd., London. 208 MADRONO [Vol. 13 Plant Genera, Their Nature and Definition, a symposium by G. H. M. Lawrence, I. W. Bailey, A. J. Eames, R. C. Rollins, M.S. Cave, and H. L. Mason, with an intro- ductory essay on Generic Synopses and Modern Taxonomy by Theodor Just. Chron- ica Botanica, Vol. 14, No. 3, 1954. $2.00. The Chronica Botanica Co., Waltham, Mass., and J. W. Stacey, Inc., San Francisco The Ferns and Fern Allies of Minnesota, by Rolla M. Tyron, Jr. i-xx, 1-166, 207 figs, 85 maps, 2 pls. 1954. $4.00. University of Minnesota Press, Minneapolis. The Ferns and Fern Allies of New Mexico, by H. J. Dittmer, E. F. Castetter, and O. M. Clark. University of New Mexico Publications in Biology No. 6:1-139. 55 figs. 1954. This and the preceding will prove useful to both layman and botanist. CALIFORNIA BOTANICAL SOCIETY PUBLISHERS OF MADRONO REPORT OF THE TREASURER FOR 1955 RECEIPTS: Balance on hand in commercial account, January 6, 1955........ $720.50 From memberships and subscriptions .........................ccscceeeeee-- 1,817.00 From sales of back numbers of Madrono.......................0000002.----- 634.50 Receipts from annual: dinner ts.) = st es 129.50 Received. as authors’ share of publication costs!..2:.225:2) =. 349.15 Totalvrecelptsae 8s ote eee a a peje $3,650.65 DISBURSEMENTS: Credited to endowment fund from sales of back numberts........ $ 634.50 ‘Breasuner’ S"expenses 2. ™ 2600s ne eee nese a ee 75.98 Corresponding Secrétary s expenses) 9 = ee ee 76.62 BditorialSecretaryis: CxpenSes 2 = ee ee 10.00 Cost. cof annual dinner: 20 2 ie a ee ee 124.80 Cost of printing, binding, and mailing Madrono, Volume 13, Numbers: (15,2533, andar cht e ee eeeres 2,264.60 Total «disbursements: 2.28.00) 4u., So ee $3,186.50 BALANCE ON HAND IN COMMERCIAL ACCOUNT, American Trust Co., PaloVAltoue S25 30 es eee ee ee ee ke $ 464.15 ENDOWMENT FunpD: Palo Alto Mutual Savings and Loan Association, balance on hand January 921950052 2 eee ee ee $3,066.52 Accrued2interests 2647.02) 2 oN Ne ee ee eee 97.70 Realized from two United States War Savings Bonds, ! series i, matured January i 1955.28 ee ee 200.00 From sales of back numbers of Madrono........................-.------ 634.50 $3,998.72 American Trust Company, savings account, balance Janu- BEY 55, VOS Gi ee Fe A eet ee ees eee ea a eee 4305225 Accrued: interest =....:2.1 eon ee eet (hss) 372.58 Totalcendowment 2020.0) 1.) 2S eee ee ee ee $4,371.30 Accounts audited and found correct: Wo. M. Htiesey, Auditor RicHuarp W. Horm, January 23, 1956 Treasurer for 1955 INFORMATION FOR CONTRIBUTORS Manuscripts submitted for publication should not exceed an estimated 20 pages when printed unless the author agree to bear the cost of the ad- ditional pages at the rate of $15 per page. Illustrative materials (includ- ing “‘typographically difficult” matter) in excess of 30 per cent for papers up to 10 pages and 20 per cent for longer papers are chargeable to the author. Subject to the approval of the Editorial Board, manuscripts may be published ahead of schedule, as additional pages to an issue, provided the author assume the complete cost of publication. Shorter items, such as range extensions and other biological notes, will be published in condensed form with a suitable title under the general heading, ‘‘Notes and News.” Institutional abbreviations in specimen citations should follow Lanjouw and Stafleu’s list (Index Herbariorum. Part 1. The Herbaria of the World. Utrecht. Second Edition, 1954). Articles may be submitted to any member of the Editorial Board. MADRONO A WEST AMERICAN JOURNAL OF BOTANY A quarterly journal devoted to the publication of botanical re- search, observation, and history. Back volumes may be obtained from the Secretary at the following rates: Volume I,1916-1929 . . . . . $5.00 Volume IT, 1930-1934. . . . . 5.00 Volume III, 1935-1936 . . . . 5.00 Volume IV, 1937-1938 . . . . 5.00 Volume V, 1939-1940. . . . . 5.00 Volume VI, 1941-1942 sl 8 sce kaye 5.00 Volume VII, 1943-1944 . . . . 5.00 Volume VIII, 1945-1946 . . .. . 5.00 Volume IX, 1947-1948 . . . . 6.00 Volume X,1949-1950. . . . . 7.00 Volume XI,1951-1952 . . wt«i CY 8.00 Volume XII, 1953-1954 . . . . 8.00 Single numbers, Volumesland2 ... . . 50 Succeeding numbers, (except Volume XI,No.2) . .. 1.00 Volume XI,No.2 . . . . . 1.50 The subscription price of MaproNo is $4.00 per year. If your institution does not now subscribe to MADRONOo, we would be grateful if you would make the necessary request. Since there is no extra charge for institutional subscriptions, an individual may hold membership in the California Botanical Society on the basis of his institution’s subscription. Address all orders to: G. THomAs Rossins, Corresponding Secretary Department of Botany University of California, Berkeley 4, California ie PEL chix uu , d i P f . r (rod Lod C \a) ee VOLUME 13, NUMBER 7 JULY, 1956 Contents PAGE NOTES ON THE PHACELIA MAGELLANICA COMPLEX IN THE PaciFic NortHwEST, A. R. Kruckeberg 209 THE DISTRIBUTION OF QUERCUS BOYNTONI, Cornelius H. Muller 221 ON THE GENERIC LIMITS OF ERIOPHYLLUM (Com- POSITAE) AND RELATED GENERA, Sherwin Carlquist 226 NEw SPECIES OF ELATINE IN CALIFORNIA, Herbert L. Mason 239 A WEST AMERICAN JOURNAL OF BOTANY PUBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY MADRONO A WEST AMERICAN JOURNAL OF BOTANY Entered as second-class matter at the post office at Berkeley, California, January 29, 1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price $4.00 per year. Published quarterly and issued from the office of Madrono, Herbarium, Life Sciences Building, University of California, Berkeley 4, California. BOARD OF EDITORS HERBERT L. Mason, University of California, Berkeley, Chairman EpGAR ANDERSON, Missouri Botanical Garden, St. Louis. LyMAN BENSON, Pomona College, Claremont, California. HERBERT F. COPELAND, Sacramento College, Sacramento, California. Joun F. Davinson, University of Nebraska, Lincoln. Ivan M. JoHNsTON, Arnold Arboretum, Jamaica Plain, Massachusetts. Mitprep E. Maruias, University of California, Los Angeles 24. Marion Ownsey, State College of Washington, Pullman. Ira L. Wiccrns, Stanford University, Stanford, California. Secretary, Editorial Board — ANNETTA CARTER Department of Botany, University of California, Berkeley. Business Manager and Treasurer—MAatcotm A. Noss Carnegie Institution of Washington, Stanford, California CALIFORNIA BOTANICAL SOCIETY, INC. President: Wm. M. Hiesey, Carnegie Institution of Washington, Stanford, Cali- fornia. First Vice-president: Roger Reeve, Western Regional Research Laboratory, Albany, California. Second Vice-president: Lewis Rose, California Academy of Sciences, San Francisco, California. Recording Secretary: Mary L. Bowerman, De- partment of Botany, University of California, Berkeley, California. Corresponding Secretary: G. Thomas Robbins, Department of Botany, University of California, Berkeley, California. Treasurer: Malcolm A. Nobs, Carnegie Institution of Wash- ington, Stanford, California. 1956] KRUCKEBERG: PHACELIA 209. NOTES ON THE PHACELIA MAGELLANICA COMPLEX IN THE PACIFIC NORTHWEST A. R. KrRucKEBERG! The members of the Phacelia magellanica complex (Hydrophyllaceae) have defied satisfactory taxonomic delimitation ever since the traditional linneon of P. circinata Jacq. was found inadequate by E. L. Greene and later workers (Brand, 1913 and Macbride, 1917). Recently Cave and Constance (1942, 1944, 1947, and 1950) and Heckard (1954) have in- dicated the probable basis for the lack of clear-cut boundaries between the members of the group. After having made chromosome counts on a large number of field collections, it became evident to Cave and Con- stance that the taxa comprising this assemblage are elements of a poly- ploid complex. The intricate pattern of apparently reticulate phylogeny in the group has not as yet been transformed into a workable taxonomic scheme. A part of the taxonomic elucidation awaits the accumulation of additional cytological information on collections from areas in the west where phacelias have not been intensively collected. The Pacific North- west is one such area. The present paper reports on field studies and chromosome counts of phacelias from northwestern United States and western Canada. A new diploid is described from Oregon and, for the first time, the existence of hexaploids in the complex is recorded. I. A NEw PHACELIA FROM SOUTHWESTERN OREGON Phacelia capitata sp. nov. Herba caespitosa perennis, 10-15 pedunculis tenuibus, quisque ex rosula laterali; pedunculi erecti simplices longi 20- 25 cm.; folia rosulata spisse sericea angustato-lanceolata integra longa 2—3 cm., lata 0.2—0.4, sessilia; folia caulina reducta, longa 2—3 cm., lata 0.2—0.4 cm.; inflorescentia ex 2—3 cymis in unico racemo congesto subcapitato constans; racemus cymarum demum longus 2.0-—2.5 cm., latus 3.0-3.5 cm.; pedicelli longi 1-3 cm., hispidi; calycis lobi angustato-oblongi longi 3 mm., lati 0.5—0.8 mm., margines pilis longis hispidi; superficies abaxialis pilis brevibus hirsuta; corolla alba rotato-campanulata longa 4-6 mm., lata 3-5 mm., lobi simplices obtuso-rotundati; appendiculae 1 mm. supra basin tubae corollae affixae; stamina et stylus exserta 5—7 mm., filamenta glabra vel paucis pilis sparsis; capsula immatura ovoidea longa 2—3 mm. tecta setis robustis longis 2-3 mm crebe; semina non visa. Deeply taprooted cespitose perennial with 10-15 thin, wiry peduncles arising from as many rosulate tufts, the whole forming a broad and multi- cipital rosette, 25-30 cm. in diameter; stems erect, unbranched, 20-25 cm. tall, finely sericeous; silvery-gray herbage consisting of a hyaline- 1 This study was supported by funds made available through the State of Wash- ington Initiative No. 171. Maprono, Vol. 13, No. 7, pp. 209-240, July 20, 1956. We 8 0 iene 210 MADRONO [Vol. 13 Fic. 1. Phacelia capitata sp.nov. a, habit; b, basal leaf; c, flowering calyx; d, ex- panded corolla. (Habit drawing x 2/5; leaf x 4/5; calyx and corolla x 6.) shiny pubescence of two sorts: appressed bristles on a background of thinly matted hairs; basal leaves linear-lanceolate, simple, entire, 2—3 cm. long, 0.2—0.4 cm. wide, apparently sessile; the 8-12 cauline leaves gradually but not wholly reduced, 0.7—2.0 cm. long, 0.2—0.4 cm. wide; in- 1956] KRUCKEBERG: PHACELIA 211 florescence consisting of 2—3 scorpioid cymes in a congested, subcapitate erect cluster, 2.0—2.5 cm. long and 3.0—-3.5 cm. wide in early fruit, that is usually borne singly and terminally and subtended by much reduced cauline leaves; pedicels in early fruit 1-3 mm. long, hispid; calyx lobes linear-oblong, 3 mm. long, 0.5—0.8 mm. wide, two-thirds as long as corolla, margins long-hispid, abaxial surface short-hirsute; corolla white, rotate- campanulate, 4-6 mm. long, 3-5 mm. broad, the lobes entire, obtuse- rounded; appendages attached barely a millimeter above the base of the corolla tube, the free portions forming a long (3 mm.) and narrow “VW” distally; stamens and style exserted 5—7 mm., the filaments glabrous or with a few scattered hairs about mid-length along the filament; imma- ture capsule ovoid, 2-3 mm. long, densely clothed with stout bristles 2 mm. long; mature seeds not seen; n = 11 (chromosome count from buds of type collection). Type. On serpentine roadcut, 2 miles east of Bridge, above Coquille River and along State Highway 42, Coos County, Oregon, May 31, 1951, A. R. Kruckeberg 2703 (WTU No. 153940). Growing with Cheilanthes siliquosa, Sisyrinchium bellum, Eriophyllum lanatum and Eriogonum nudum. Topotypes: '% mile north of State Highway 42, Kruckeberg 3313, “stony slope near Bridge, M. EF. Peck 20324 (UC); C. L. Hitch- cock s. n. (specimen from field transplant grown in greenhouse); 2.5 miles east of Bridge, Constance 3454 (seedlings also grown in greenhouse, Heckard S-295, UC). Other collections seen. OREGON. Serpentine bluff about 1 mile northerly from Myrtle Creek, J. T. Howell 28807 (CAS) ; 2 miles south of Myrtle Creek, Eastwood and Howell 1474 (CAS). Phacelia capitata is distinguished from its nearest congeners by a num- ber of features which are summarized in Table 1. In addition to the features just tabulated, there are three others that give this new taxon specific distinction: (1) The large number of thin, wiry stems arising from the broadly cespitose rosettes; (2) the large num- ber of cauline leaves; (3) the rarity of cymes below those of the terminal capitate cluster. Some of these diagnostic characters are illustrated in figure 1. The distinctness of P. capitata is further enhanced by its diploid chromosome complement. To date, only one other diploid collection of a magellanica-type Phacelia has been reported for Oregon (Cave and Con- stance, 1942). This, the wide-ranging, rank biennial, P. heterophylla Pursh, is a very different entity. A search through herbarium material collected in southwestern Oregon which was available from several western herbaria has failed to disclose additional specimens that closely match those from the type locality of P. capitata. A specimen collected at Reston, Douglas County, by Pro- fessor Morton E. Peck (Peck 6035 WILLU) is suggestive of the new species, but differs in having tawny-green herbage, wider basal leaves and fewer cauline leaves. Collections from the Roseburg-Myrtle Creek area in Douglas County resemble P. capitata: Peck 6033, 6022 (WILLU); [Vol. 13 MADRONO 2426 —o—s/:“v—"v000—Os)989—9—000—°—€—SSSSSsms@smmn9nm@MRRRRDNDD9D09D9l93onosgyO OO “SUIeY -uno], AYIOY 0} Jsva pue eIUIOjsI[eD 07 YyNOS “SOUIAD [PUTUIIO} eIqUIN]OD YsTWIg Wor] MOTeq SuTyoueiq SIX® UBIIIIS-apeosed uaj}jo ‘SuIpusose 10 *SJIPRII snovdies JO }s¥va svare auld MOT yada apouNped ‘sautAD ¢—-T YIM AjareI 10 -d9SOB11}S -JaA Jo ysniqases usdQ pasodsip Ajeyefnotueg = atjua ‘a}epoadue’'T "Wd TI-8 "Wd SI-IT ‘AB1B-AIIATIS ZZ pycygdoona ‘U0Z9IQ }SIAMYINOS 0} VIUIOJI[VD Wo.ry ‘SUIPUIISe “eUTWIET JO aseq 3 Sasuey JSVOD YON ‘T[2} “Wd S¢—-ST Sjayeoy Jo sured 7-] pidsty-1e[npurys ‘sdoi9jno ayyWopiied =sulaj}s fsquiAiod uedo = yyIM udazJo ‘8u0TqGo :ysed snoo ZZ pue sutjuedies Aq IO asuap UT sauIAd -d0UR] 0} IBIUTT "Wd ¢-Z "wd /-Z -UISNAIIJ-AUME TL TI DSOquikso9 "U0Z3910 “Od sood ‘AzTeI0], VdAy "Wad9 ‘prdsiy wio1y A[UO UMOUY ‘sdoi9 Ayyys qypeq {sourAs ‘aiTqua -passaidde ‘reynp -jno sutjuedias ‘AIq] ¢-Z JO Jajysnpo ayeytded = ‘a}e[OVOURI-IeAUT'T =“ WI. 7-10 "Wd €-Z =-U[ da ‘ABIB-AIVATIS Il Dyipyidn9 yeuiqey VIUIISIIOYUL SIABI] JO aulneg [esegq aseqioy jo Jaquinu sa1dads pue ssuey jo pury SuUIqoO, pue sdeys SaAea] JO y)BuaT 9IN}SIA PUe IO[OD u a—€CoO‘_*wzwo?‘*tINNtweTw?00Ge0a«SOoooooo——o SSS SYANHONO,) ISHUVAN] HLIM VIVIIdVO VITAOVHG JO NOSIYVdWO, ‘T ATIv 1956] KRUCKEBERG: PHACELIA 213 Eastwood and Howell 1475 (CAS); Constance and Rollins 2966, and Cusick 4048 (both WS). Like the new species, the Roseburg-Myrtle Creek specimens have silvery-sericeous foliage and narrowly lanceolate cauline leaves. However, these five specimens do not have the capitate in- florescence or the thin wiry peduncles of P. capitata. A further difference is that the plant collected by Constance and Rollins proved to be tetra- ploid. Constance, at one time, annotated two of the four specimens in question as P. leucophylla Torr., thus placing them in a species, the usual distributional limits of which are east across the Cascade Range from Roseburg. An exploratory search for additional stations of P. capitata in the region of the type locality as well as along the south fork of the Coquille River was unrewarding. More intensive botanizing on the serpentine areas of southern Coos County and western Douglas County as well as on similar ecologic sites along the northern boundaries of Curry and Josephine counties may yield new locations. II. THE HEXAPLOID PHACELIA LEPTOSEPALA RYDBERG A widely accepted practice among field collectors and herbarium work- ers is to designate as either Phacelia leucophylia Torr. or P. heterophylla Pursh, those perennial phacelias of the magellanica complex that are collected from the crest of the Cascade Range eastward. This practice, at least, has the merit of correctly disposing of the majority of perennial phacelias collected in the Pacific Northwest, for these two taxa are cer- tainly the two most widespread of the complex in the region. Ranging from arid sagebrush plains to the mid-montane pine belt, the two species present a variety of altitudinal and habitat forms. Phacelia leucophylla is tetraploid (n = 22) in the several widely separated localities of the Pacific Northwest for which chromosomal numbers have been determined (Table 2 and Cave and Constance, loc. cit.). The equally distinctive and ubiquitous P. heterophylla is both diploid (n= 11) and tetraploid (n = 22) throughout the same general range. The more puzzling collec- tions from the Pacific Northwest have been, however, the several high- montane forms of the P. magellanica group. Some of these truly appear to be altitudinal races of the two aforementioned taxa, but others are definitely distinct. These montane forms have gone under a variety of names. One of the names in the P. magellanica complex recognized as possibly valid by Constance (in Abrams, 1951) is P. leptosepala Ryd- berg. Justification for including within this taxon many of the subalpine collections of Phacelia from the Pacific Northwest now will be considered. Because of the tawny-green herbage, plants considered by the author to be P. leptosepala often have been placed under P. nemoralis Greene (n = 11,22) a much taller, rank perennial of the Coast Ranges and bor- dering valleys having one to few stiffly erect stems clothed with strongly hispid-hirsute herbage. The montane plant in question, however, is ces- 214 MADRONO [Vol. 13 Fic. 2. Meiotic chromosomes in pollen mother cells of certain members of the Phacelia magellanica complex. a, P. capitata, n= 11 (Diak.), Kruckeberg 2803; b. P. corymbosa, n= 22 (Diak.), Kruckeberg S—110; c. P. frigida, n = 22 (Diak.), Kruckeberg s.n., July 1951; d, P. frigida, n = 33 (M,) Kruckeberg 3781; e, P.lep- tosepala, n = 33 (Diak.) Kruckeberg s.n., July 1951; f. P. mutabilis, n = 22 (Mj), Kruckeberg 3779. pitose, with several assurgent to erect flowering stems. This discrepancy led Jones (1936) in his treatment of the flora of the Olympic Peninsula to remark: “the lowland plants are erect, 60-180 cm. tall, strongly hispid- hirsute and probably represent true P. nemoralis; the subalpine plants are decumbent at the base, less harshly pubescent, with whiter corollas, and may represent an undescribed species or variety.” That Jones’ con- jecture about the “‘un-nemoraloid” character of these subalpine plants was a valid one may be judged from an examination of the following evidence. During the past four years, numerous collections of bud samples for chromosome counts, as well as of living plants and herbarium specimens have provided information which permits a better understanding of this puzzling Phacelia. Very soon it was discovered that all tawny-leaved, cespitose plants collected in the Cascade and Olympic ranges of Wash- 1956] KRUCKEBERG: PHACELIA 215 TABLE 2. HAPLOID CHROMOSOME NUMBERS OF NORTHWEST PHACELIAS Haploid N pects Chromosome o Number Collections P. capitata Kruckeberg 11 3 P. corymbosa Jepson (11) 7, 22 2 P. frigida Greene 22 n00" 143 P. heterophylla Pursh Tiler022) 1 P. leptosepala Rydberg 337 18 P. leucophylla Torrey 22 14 P. mutabilis Greene (T1)5,22 5 P. nemoralis Greene C18), 22 2 * Hitherto unreported chromosome numbers; other numbers confirm counts by Cave and Constance (loc. cit.). + Haploid numbers in parentheses are from Cave and Constance (loc. cit.). ington were hexaploid (n= 33, see fig. 2e and Table 2). Moreover, the chromosome number, n= 33, could be predicted with considerable ac- curacy in all subsequent collections that were “leptosepaloid” in charac- ter. This record of a hexaploid is the first for any Phacelia. The type of P. leptosepala from Vermilion Lake, British Columbia, as well as the three other specimens from British Columbia and Montana cited by Ryd- berg (1909) have been examined and prove to be good matches for the hexaploid material from Washington. Unfortunately, it has not yet been possible to obtain material for chromosome counts on the phacelias from the type locality. Despite the apparently extensive range of P. leptosepala throughout the Pacific Northwest, it has not been collected at all frequently. Only 102 specimens out of over 800 sheets of the P. magellanica complex from the Pacific Northwest on loan from ten herbaria were judged by the author to be P. leptosepala. This scarcity is perhaps easily understood in view of the likelihood that, in its superficial resemblance to montane forms of P. leucophylla and P. heterophylla, it might be regarded as too common to be collected. Phacelia leptosepala is perhaps more like P. corymbosa Jepson than any other of the possible congeners in the Northwest. Both are cespitose and have a ferrugineous cast to the herbage. However, P. corymbosa is distinguished from P. leptosepala in the possession of a decidedly glandu- lar pubescence. Moreover, it has been collected thus far as either a diploid or a tetraploid and may be separable from P. leptosepala on chromosome number. The two species are widely disjunct in distribution, P. corymbosa being confined rather closely to the serpentines of southwestern Oregon and northern California. Phacelia mutabilis Greene (n= 11 and n= 22) and P. leptosepala may in exceptional specimens bear close resemblance to each other. In typical 216 MADRONO [Vol. 13 material, however, P. mutabilis, with its weakly rosulate habit and few short and erect or assurgent stems is readily distinguishable from P. leptosepala. In mid-montane areas of the Cascade Range of southern Washington, P. nemoralis and P. leptosepala may intergrade. In the ex- cellent series of phacelias collected by Suksdorf around Mount Adams (Mount Paddo) are certain specimens which cannot be placed readily in either P. leptosepala or P. nemoralis. The mid-montane populations in the Mount Adams-Skamania County area may truly represent altitudinal variants of P. nemoralis. Data on recent collections from this area are note- worthy. One from Spirit Lake (Kruckeberg 3814) was tetraploid; the plants are like those collected by Suksdorf, and may thus be regarded as samples of a mid-montane ecotype of P. nemoralis. Two collections were made on the nearby slopes of Mt. St. Helens, possibly four air-line miles away and 1000 feet higher; both were hexaploid and “good” P. leptose- pala. When additional collections from this area have been examined cytologically, it should be possible to make a better evaluation of the phacelias in the territory of Suksdorf’s most intensive collecting. A number of characters distinguish P. leucophylla, typically an inhabi- tant of the sagebrush and yellow-pine belt, from P. leptosepala. Differ- ences between them have been compiled in Table 3. Equally separable from P. leptosepala is the lavender-flowered high-montane, P. leucophvlla var. alpina (Rydb.) Dundas. Although P. alpina Rydberg has been con- sidered an infra-specific element of the biennial, P. heterophylla Pursh, because of its greater similarity to P. leucophylla in habit, leaf shape and vesture, and flower color, the author agrees with both Dundas (1933) and Macbride (1917) that it had best be considered an altitudinal variant of the latter species. Like P. leucophylla, var. alpina has whitish, sericeous herbage and mostly simple leaves, it may be expected to be sympatric with P. lJeptosepala in eastern British Columbia, Montana, and Idaho. Less tangible are the morphological differences between P. leptosepala and P. frigida Greene, the latter from subalpine regions in the southern Cascades of Oregon and in the Sierra Nevada of California. Typically, the more diminutive P. frigida is not obviously ferrugineous-green, but merely sericeous or anthocyanous-gray in foliage. Though the range of P. frigida is mainly to the south of that for P. leptosepala, occasional specimens from more northerly areas of the Oregon Cascades cannot be clearly distin- guished from dwarfed specimens of P. leptosepala collected in the north- ern Cascades or Olympics of Washington. Of two such specimens seen, one came from Nash Crater lava flows, Linn County (A. W. Roach s.n., OSC) and the other from Three Fingered Jack, Jefferson County (Con- stance 3444, UC). The latter proved to be tetraploid (fide Constance, personal comm.). Thus in Oregon the ranges of the two may well over- lap. Even with more typical material, P. leptosepala and P. frigida can- not be clearly separated on the basis of chromosome number. Seven col- lections of “good” P. frigida were determined by Cave and Constance 1956] KRUCKEBERG: PHACELIA 2a TABLE 3. MORPHOLOGICAL COMPARISON OF P. LEPTOSEPALA AND P, LEUCOPHYLLA HapBiIt STEMS BASAL LEAVES CAULINE LEAVES INFLOR- ESCENCE CALYX COROLLA STAMENS PISTIL SEED DIsTRI- BUTION P. leptosepala Cespitose, flowering stems usually procumbent to assurgent, with many basal subrosettes; caudex with numerous branches either bearing stems or sterile rosulate shoots. Herbage tawny-ferrugine- ous throughout with mainly hir- sute-hispid pubescence. Several, decumbent to assurgent. 9-11 cm. long, 7-10 mm. wide, oc- casionally entire or usually with 3-5 leaflets, petiole 4 cm. long; lanceolate in outline; long-hispid mainly on veins and margin of blade, sparingly hispidulose in in- tercostal areas; pungent bristles usually reddish brown. Several, somewhat reduced, com- pound, with 3 leaflets, 5-8 cm. long, hispid. Flowers at anthesis in dense, tight- ly coiled cymes in corymbs, the cymes 4—5 cm. long, tawny-hispid throughout. Lobes at anthesis linear-lanceo- late, 9 by 1.0 mm.; long, stiff, pungent bristles mostly scattered on margin and on faint midrib; short soft hairs thinly scattered over lamina of lobe. Tubular-campanulate, 7 mm. long, exserted. Filaments pilose, 8-9 mm. long, exserted. Style 9 mm. long, bifid to % its length, hispidulose at base; cap- sule hispid. Lance-ovate, 2-3 mm. long, brown, reticulate-pitted on all faces. Montane and subalpine regions of Cascades and Olympic ranges of Washington; mountains of British Columbia and east along Canadian border to Rocky Mountains of British Columbia, Alberta, and Montana. P. leucophylla Flowering stems mainly erect from single or a few tufts of rosette leaves. Foliage silvery-canescent throughout, clothed with mixed strigose and sericeous pubes- cence. 1-3, assurgent to stiffly erect. 11-15 cm. long, 1-1.5 cm. wide, commonly entire or rarely with 1-3 leaflets, petiole 4-10 cm. long; strigose mainly on margins and veins, densely sericeous on inter- costal areas of lamina. Hairs of pubescence always glistening white. Numerous, scarcely reduced ex- cept just at base of cymes, mainly entire, 8-11 cm. long, strigose- sericeous. Flowers at anthesis in virgately disposed cymes, the cymes 6-8 cm. long, straight except for tight coil of unopened flowers at tip, silvery strigose-hispid throughout. Lobes at anthesis ovate-lanceolate, 6 by 1.5 mm.; long stiff, pungent bristles on margin and scattered on abaxial surface, short-pilose hairs intermixed on lamina of lobe. Rotate-campanulate, 5 mm. long, 5 mm. broad. Filaments pilose, 10-12 mm. long, exserted. Style 12 mm. long, bifid to 44 its length; hispidulose at base, cap- sule hispid. Ovate (to lance-ovate), 2.0-2.5 mm. long, brown to black, reticu- late-pitted on all faces. Open sagebrush or yellow-pine areas mainly east of the Cascade- Sierran axis from British Colum- bia south to California and east to Wyoming, Nebraska, and Colo- rado. 218 MADRONO [Vol. 13 (personal comm.) as tetraploid. Yet of the four cytological samples of typical P. frigida taken by the author, only one was tetraploid; the others were hexaploid (i.e., like P. leptosepala). A fifth hexaploid collection of P. frigida was made late in 1955, from Mt. Hood, Oregon (Kruckeberg 4000). One hexaploid was from Crater Lake in Oregon and the other two were from Mt. Lassen and the central Sierra Nevada of California. This is an instance where incomplete cytological evidence could have been used to emphasize taxonomic discontinuity, but where additional chro- mosome numbers make such use of cytological data unwarranted. Along the eastern flank of the Cascades of Washington, in areas of mid-montane altitudes, P. /eptosepala and P. leucophylla can be expected to occur sympatrically. One such station is on Blewett Pass in Chelan County, where in 1952 the sympatry was positively confirmed by chro- mosome counts of. both tetraploid P. leucophylla and hexaploid P. lepto- sepala. The following year a small population sample of ten plants was collected, together with buds, along a roadbank on the southeast-facing side of the pass. All plants appeared to be either good P. leptosepala or P. leptosepala with a suggestion of “leucophylloid” features. However, every one of the ten plants was hexaploid. It was hoped that the sampling would include both tetraploid and hexaploid plants and possible hybrids between the two different chromosome-numbered populations as well. Examinations of larger population samples would undoubtedly be more effective in judging whether or not these two taxa are interpollinating in this area of contact. SPECIMENS OF PHACELIA LEPTOSEPALA EXAMINED:2 OREGON. Wallowa County: Evergreen Forest Camp, upper Imnaha River, Kruckeberg 3315; upper Toomey Creek near mouth of Imnaha River, Kruckeberg 2303 (RM) ; on granitic ridges above Douglas Lake, Wallowa Mountains, Kruckeberg 2429. WaSHINGTON. Chelan County: Jolly Mountain, Easton sn. (OSC); Falls Creek near Bryan Butte, Ward 59 (WS); Ice Creek, Morill 369; Nelson Butte lookout, Kelly 28 (WS) ; Stuart Pass. St. John and Thayer 7308 (WS) ; Stevens Pass, Otis 789 (WS). Clallam County: Mount Storm King, Lawrence 351 (WS); Olympic Moun- tains, Elmer 2829 (WS); Mount Angeles, Kruckeberg 2782, J. W. Thompson 5519, Webster 1109; ridge trail between Hurricane Ridge and Mount Angeles, Kruckeberg 3807; Seven Lakes Basin, G. N. Jones 8249, 8267. Garfield County: Bluffs above Tucannon River, near Willow Spring road, Peters 388B (WS). Grays Harbor Coun- ty: Mount Colonel Bob, Thompson 7284. Jefferson County: south slope of Mount Constance, Rollins and Chambers 2649 (UC); Constance Ridge, G. N. Jones 5803, Meyer 707 (WS); Lake Constance, Thompson 7902. Kittitas County: Teanaway- Turnpike Basin trail, Kruckeberg 2614; summit of Iron Peak Trail, Teanaway River drainage, Kruckeberg 3287; Fish Lake, Kruckeberg 2954; divide between Jungle and Johnson Creeks, toward Malcom Mountain, Hitchcock 19131 (WS); near Naneum Meadows, Kruckeberg 3247; Blewett Pass region, Grant, August 1929; 1 mile south of Blewett Pass, Constance and Beetle 2758; head of Beverly Creek, Thompson 5851; Bald Mountain, Thompson 14796 (WS). Lewis County: mica schist peak north of White Pass, St.John 5631 (WS); Goat Rocks, Thompson 15219. Mason County: “Institutional abbreviations follow Lanjouw and Stafleu, 1952. Specimens for which no herbarium designation is given are from the University of Washington, Seattle. 1956] KRUCKEBERG: PHACELIA 219 Skokomish Valley, Kinkaid s.n. (WS); Mount Elinor, Freer 264; Heart Lake, Dick- inson 26. Okanogan County: Slate Peak trail, Kruckeberg 3269; Cash Creek near Harts Pass, Kruckeberg 3271; 2 miles below Mazama, Kruckeberg 3272. Pierce County: Chinook Pass Highway, 3 miles west of Ohanapecosh Junction, Kruckeberg 3803; Mount Rainier, C. V. Piper 486; Grant, August 1925; Cowlitz Glacier, Pope, August 1904; Mount Wow, Warren 1606; Nisqually River at Nisqually Glacier, Heller 14762; Carbon River, Thompson 5446; Nisqually Checking Station, Abrams 9213; below Carbon Glacier, Eide sn. (WS); White River, Warren 1456 (WS) ; north side Mount Rainier, Jones 77 (WS). Skamania County: Mount St. Helens, Thompson 12673; Dog Creek, Suksdorf 11483 (WS) ; Butterfly Lake, Suksdorf 3509 (WS); north slope Mount St. Helens, Gorman 5843 (WS); Meyer 762 (WS); St. John, et al. 7371 (WS). Snohomish County: Silverton, Bonck 148 (WS). Whatcom County: Mount Herman, Thompson 8049; Mount Baker, Turesson 22 (WS) ; Sour- dough Mountain, St. John 6836 (WS) ; Winchester Mountain, St. John 8972. Yakima County: Mount Aix, Thompson 15023; Mount Adams, Henderson s.n.; Suksdorf 3442; 3516; 4355; 6850; August 14, 1898 (all WS); Bald Mountain ridge, St. John 7788 (WS); Gold Hill, Pickett 1364 (WS). IpaHo. Kootenai County: Near Fernan Lake shore, Coeur d’ Alene, Rust s.n. (WS). MontTAna. Beaverhead County: Sheep Creek, near Wisdom, Berghuis 23. Flat- head County: Camp Misery, Mount Aeneas, Harvey 3325 (MONTU); Big Creek, Edie 7 (MONTU) ; 2 miles southeast of Essex, Brenckle and Shinners 41-093 (RM) ; top of Columbian Mountain, Rogers and Rogers 1155 (WS) ; China Wall, Kirkwood 2292 (MONTU); head of Moose Creek, China Wall, Kirkwood 2330 (MONTU). Gallatin County: Bridger Mountains, 3 miles north of Brackett Creek, Hitchcock and Muhlick 12446 (WS); Spanish Basin, Rydberg and Bessey 4850 (RM). Glacier National Park: Sperry Glacier, Umbach 826 (NY); Museum (Clements) Glacier, Elrod sn. (MONTU) ; trail from Logan Pass to Hidden Lake, Nelson and Nelson 3158 (RM); Clements-Reynolds saddle, Logan Pass, Ewan 18474 (UC); Glacier Park, Osterhout 8106 (RM). Lake County: 6 miles southeast and high above Mc- Donald Lake, Mission Range, Hitchcock 18327. Lincoln County: Mount Mcdougal, Kootenai Mountains, Umbach 139 (NY). Powell County: 6 miles south of Big Prairie Ranger Station, Hitchcock 18815. BriTisH CoLumMBiIA. Vermilion Lakes, Edith M. Farr 1013 (TYPE! deposited at NY); Lardo, Selkirk Mountains, Shaw 695 (NY); Sage Pass, 7. M.C. Taylor 8611 (UBC); Antimony Mountain, Lytton, Tisdale s.n. (UBC) ; Cowiche Lake, Cotton- wood Creek (Nanaimo District), Simpson and Simpson sn. (WS); Lake House, Skagit River, Macoun 76757 (WS); “probably near Lumby” (sic), Anderson s.n. (WS) ; Cameron River valley, Vancouver Island, Rosendahl 1996 (UC). ALBERTA. Upper Waterton Lake near Bungalow Camp, Nelson and Nelson 4727 (RM); Upper Carthew Lake, Waterton Lakes National Park, Hitchcock and Martin 7906; Prince of Wales Hotel, Waterton, Breitung 15897 (UC) ; Cameron Lake, Wat- erton Lakes National Park, Ewan 18497 (UC); Waterton Lakes Park, Moss 3352 (UC); VOUCHERS FOR CHROMOSOME COUNTS OF P. LEPTOSEPALA Ryps. The vouchers for chromosome counts of this species as well as for others cited in this paper are de- posited in the herbarium of the University of Washington, Seattle. All 18 collections were n = 33. WASHINGTON. Chelan County: Kruckeberg 3319 (population sample of ten plants at Blewett Pass). Clallam County: Mount Angeles, Kruckeberg 3809. Jefferson County: Anderson Pass, Olympic National Park, Kruckeberg 4010. Kittitas County: Upper Teanaway River, Kruckeberg s.n., July, 1951; Wenatchee Mountains, Kruckeberg 2963, 3050, 3219, 3238, 3253, 3257, 3287, 3945, 3953, 3965, 3968. Lewis County: Egg Butte, Goat Rocks Primitive Area, C. Leo Hitchcock s.n. Pierce County: Rainier National Park, Kruckeberg 3805. Skamania County: Mount St. Helens, Kruckeberg 3978, 3993. Snohomish County: Stillaguamish-Sauk rivers basin, Kruckeberg 3245, 3246; Twin Lakes above Monte Cristo, Kruckeberg 3948. What- 220 MADRONO [Vol. 13 com County: Mount Baker area, Kruckeberg 3797, 3801, 3802. Yakima County: Chinook Pass area, Kruckeberg 3260, 3804. VOUCHERS FOR CHROMOSOME COUNTS OF OTHER TAXA Phacelia corymbosa Jepson. (n= 11, 22). Catirorn1a. Lake County: 4 miles northeast of Middletown, Kruckeberg S—214 (n = 22). OrEGON. Josephine County: Illinois River road, Kruckeberg G-109 (n= 22). Phacelia frigida Greene. (n = 22, 23). Catirornia. Amador County: summit of Carson Spur, Constance 3090 (UC) (tetraploid). El] Dorado County: Meeks Bay, Lake Tahoe, Rollins 3029 (UC) (tetraploid). Fresno County: Seldon Pass, Krucke- berg 3491 (n= 33). Lassen County: slopes of Mount Lassen, Kruckeberg 3710 (n = 33). Mono County: east slope of Mono Crater, Constance and Bacigalupi 3435 (UC) (tetraploid) ; one-fourth mile east of Sonora Pass, Constance and Bacigalupi 3437 (UC) (tetraploid). NeEvapa. Washoe County: east side of Mount Rose Pass, Constance and Mason 2797 (UC) (n= 22). Orecon. Douglas County: North Junc- tion, Crater Lake National Park, Kruckeberg 3781 (n = 33). Deschutes County: three miles south of Lava Lake, McKenzie Pass area, Kruckeberg, July 1950 (n= 22). Hood River County: Mount Hood, above Cloud Cap, Kruckeberg 4000 (n= 33). Klamath County: northeast rim of Crater Lake near Palisade Point, Constance 3362 (UC) (tetraploid). Phacelia heterophylla Pursh. (n = 11, 22). OrEcon. Jackson County: Neil Creek, Kruckeberg G-116 (n~—11). Phacelia leucophylla Torr. (n==22). Orecon. Deschutes County: Between Tu- malo and Sisters, Kruckeberg 2071. Jackson County: Blair quarry near Ashland, Kruckeberg G-131. Wallowa County: Upper Imnaha River above Coverdale Guard Station, Kruckeberg G-229. Wasco County: 3 miles east of junction of Madras- Mt.Hood and Maupin-Mt. Hood highways, Hitchcock 19651. WASHINGTON. Kittitas County: 2 miles below junction of Hurley Creek and Liberty-Beehive roads, Krucke- berg 3240, 3254, 3255; lower Teanaway River valley, Kruckeberg (s.n.), July 1951; Mt. Lilian, Kruckeberg 3227; Swauk Creek, Kruckeberg 3317. Klickitat County (var. Suksdorfii Macbride): White Salmon, Phillips (s.n.) September 1952; between Blockhouse and Klickitat, Kruckeberg 3305; between Bingen and Lyle, Kruckeberg 3306. Yakima County: Satus Pass, Kruckeberg 3818. Phacelia mutabilis Greene. (n = 11, 22). CALIFORNIA. Sierra County: Between Webber and Meadow Lakes, Constance, Bacigalupi and Nasir 3472 (tetraploid). OrEGON. Deschutes County: 3 miles east of McKenzie Pass, Constance 3456 (tetra- ploid). Jackson County: Rogue River Natural Bridge, 2 miles southwest of Union Creek, Kruckeberg 2017 (n= 22). Wallowa County: Hat Point, Kruckeberg 2470 (n = 22). Crater Lake National Park: along Rim Drive, 3 miles above Park head- quarters, Kruckeberg 3779 (n= 22). Phacelia nemoralis Greene. (n= 11, 22). OrEGoN. Coos County: 2 miles east of Bridge, Kruckeberg 3314 (n = 22). WasuinctTon. Aberdeen area, Kruckeberg 3815, 3816, 3817. Skamania County: Spirit Lake, Kruckeberg 3814; Washougal River, Kruckeberg, July 1953 (greenhouse plant from seed collected by R. Ornduff (n=22). ACKNOWLEDGMENTS. The author wishes to express his gratitude to two students of Phacelia at the University of California, Dr. Lincoln Con- stance and Dr. L. R. Heckard who, through many cooperative efforts and stimulating exchanges of ideas have materially aided this study. Thanks are also due to Mr. Paul Pascal of the University of Washington for the Latin translation, to the several curators of herbaria who made their col- lections available to the writer, and to my wife for assistance with the drawings. Department of Botany, Univerity of Washington, Seattle 5, Washington. 1956] MULLER: QUERCUS BOYNTONI 2z1 LITERATURE CITED _ Branp, A. 1913. Hydrophyllaceae. Das Pflanzenreich 4251:95-102. Cave, Marton S., and L. Constance. 1942. Chromosome numbers in the Hydro- phyllaceae. Univ. Calif. Publ. Bot. 18:205-216. and —————. 1944. Chromosome numbers in the Hydrophyllaceae: II. Univ. Calif. Publ. Bot. 18:293-298. and —————. 1947. Chromosome numbers in the Hydrophyllaceae: III. Univ. Calif. Publ. Bot. 18:449-465. and —————. 1950. Chromosome numbers in the Hydrophyllaceae: IV. Univ. Calif. Publ. Bot. 23:363-382. Constance, L. 1951. Hydrophyllaceae, in Abrams, Illustrated Flora of the Pacific States: 3:476—532. Dunnpas, F. W. 1934. A revision of the Phacelia californica group (Hydrophyllaceae) for North America. Bull. So. Calif. Acad. Sci. 33:152-168. HeckArp, L. R. 1954. Cytotaxonomy of a polyploid complex. VIII® Cong. Int. Bot. Rapp. & Comm. Sect. 9:72, 73. Jones, G.N. 1936. A botanical survey of the Olympic Peninsula, Washington. Univ. Wash. Publ. Biol. 5:219. Macsriwe, J.F. 1917. III. Notes on the Hydrophyllaceae and a few other North American Spermatophytes. Contr. Gray Herb. (New Ser.) 49:23-47. RypbBeErc, P. A. 1909. Studies on the Rocky Mountain flora. XX. Bull. Torrey Club 36:767. THE DISTRIBUTION OF QUERCUS BOYNTONI CoRNELIuS H. MULLER Quercus boyntoni Beadle was discovered near the summit of Lookout Mountain in Etowah County, Alabama, in 1900. Beadle (1901) named the plant after its collector and the species was recognized subsequently by Small (1913, 1933) and by Trelease (1924). Sargent (1918) reduced it to the status of a variety of Q. stellata Wang., stating that “the dwarf habit of this little oak is due probably to the exposed position and high altitude where it grows.” In 1942 there first came to my attention a series of specimens of a dwarf oak collected in 1934 in Angelina County, eastern Texas, by Effie Boon and by B. C. Tharp. The distinction between this and Quercus stel- lata was obvious, but the disposition of the Angelina County plant posed a problem, especially in the light of shrubby forms of other tree species known to occur in the prairie regions of Texas. Its segregation appeared unwise without a study of the plant in the field. Consequently, in my treatment of the oaks of Texas (1951) these specimens were included in Q. stellata. It is to this inclusion that the description of Q. stellata owes such characters as ‘“‘small shrubs” and “leaves .. . obtriangular . . . bases cuneate ... blades undulately . . . 2— to 4-lobed.” The Boon collection cited from Angelina County under Q. stellata should have been identified as Q. boyntont. (ehp) MADRONO [Vol. 13 Recently I devoted an entire day in the field in Angelina County to a search for Quercus bovntoni and a study of its distribution compared with that of the related species. This was followed by a study of borrowed specimens of this species from the United States National Herbarium, including the Biltmore series and the Charles Mohr collection containing some Boynton specimens upon which Beadle based his original descrip- tion. For this loan I am deeply indebted to the curator. All the Alabama specimens cited are deposited in the United States National Herbarium, while the Texas specimens cited are in my private herbarium. Duplicates of my own collections have been sent to the United States National Her- barium and otherwise distributed. There follows an amplified description of Quercus boyntoni based upon both Alabama and Texas specimens. It should be emphasized that this description might as well have been drawn from one population as the other, so similar are the two. QUERCUS BOYNTONI Beadle, Biltmore Bot. Studies 1:47. 1901. Quercus stellata var. boyntonu Sarg., Bot. Gaz. 65:437. 1918. Rhizomatous shrubs 20 cm. to 3 m. tall (or 5 m., fide Beadle), trailing or sometimes semi-erect; twigs 1.5—3 mm. thick, densely fulvous-tomen- tulose with a mixture of simple appressed glandular hairs and moderately spreading stellate hairs, the pubescence darkening and persisting through the second season; buds 2-3 or even 4 mm. long, ovoid, acute or some- times rounded, russet, sparsely pubescent; stipules deciduous, 3-5 mm. long, subulate, sparsely hairy; leaves deciduous or subevergreen, thin and rather soft, 5-10 (12) cm. long, 2—6 or rarely 8 cm. broad, cuneate to oblanceolate, obovate or oblong, characteristically roundly 3-lobed at the broad apex or sometimes 5-lobed above the entire cuneate base, mar- gins minutely cartilaginous-revolute, upper surface glossy, in youth spar- ingly glandular-puberulent and with scattered stellate hairs, at length glabrate or the stellate pubescence persistent especially about the midrib, lower surface dull, persistently fulvous-glandular-puberulent and stellate- pubescent or the pubescence silvery, the veins about 6 to 8 on each side, very irregular and with some intermediates, those passing into the lateral lobes very prominent, slightly raised above and prominently so beneath, markedly and irregularly branching and anastomosing; petioles 5-10 mm. long, moderately slender, persistently pubescent like the twigs; staminate catkins 3-6 cm. long, fulvous-glandular-puberulent and stellate-pubes- cent, the puberulent anthers well-exserted from the ciliate perianth; pis- tillate catkins about 5 mm. long, about 3-flowered, and subsessile, densely fulvous-pubescent; fruit annual, solitary or paired on peduncles 2-10 or rarely even 35 mm. long; cups 10-13 mm. broad, 5-10 mm. high, deeply cup-shaped or more shallow, the scales densely fulvous- or silvery-tomen- tulose, the bases moderately or markedly thickened, the thin apices close- ly appressed, the acorns 10-17 mm. long, 7-10 mm. broad, broadly or narrowly ovoid, the ends broadly rounded, brown and minutely puberu- lent especially about the apex, about one-half or only one-third included. 1956] MULLER: QUERCUS BOYNTONI 223 Fic. 1 (top). Quercus boyntoni Beadle, the lectotype specimen from Etowah County, Alabama, C. L. Boynton s.n., sheet no. 780302, US. Fic. 2 (bottom). Quer- cus boyntoni Beadle, a typical specimen from Angelina County, Texas, C. H. Muller 9609. 224 MADRONO [Vol. 13 Specimens examined. ALABAMA: Etowah County, about summit of Lookout Mountain, near Gadsden, May 1900, C. L. Boynton s.n. (Herb. Chas. Mohr) ; July 29, 1900, C. L. Pollard and W. R. Mazon 339; September 1, 1900, Boynton s.n. (lectotype) ; April 1901, 7. G. Harbison s.n.; April 26, 1901, collector unknown 1905 (Biltmore Herb.) ; October 12, 1901, collector unknown 1905a (Biltmore Herb.) ; September 25, 1902, collector unknown 1905c (Biltmore Herb.). Jefferson County, glades along Lost Creek, Shades Mountain, near Birmingham, April 18, 1931, E. J. Palmer 38946. TexAs: Angelina County,! “Charlie Massengill’s Farm,” August 15, 1934, E. Boon 473; August 16, 1934, Boon 492; September 16, 1934, Boon s.n.; Shawnee Switch, September 16, 1934, B. C. Tharp s.n.; 1 mi. W. of Shawnee Creek, 2 mi. E. of Shawnee Store, Shawnee Prairie, 9.5 mi. S.W. of Huntington, August 26, 1953, C. H. Muller 9609, 9610, 9611. The Alabama collections taken in the first years of the century are sin- gularly lacking in certain data. The Biltmore specimens, of course, do not bear the collectors’ names. “Black Creek near Gadsden” is mentioned on a Charles Mohr label with a plant apparently collected by Boynton and appearing on the same sheet with another Boynton collection. Another Mohr sheet bears the only reference to habit—“trailing shrub not over 2 feet high’—written in pencil on a crude label and accompanying a duplicate of Boynton’s collection of May 1900. These Mohr herbarium specimens seem to be the only examples of Boynton’s collections extant. One of them (US no. 780303) bears the notation ‘‘Biltm. herb.” Another (US no. 780302) is here chosen as the type on the grounds that it is the only mature specimen of Boynton material available and was seen by Beadle prior to the publication of the species. Beadle refers to Boynton collections in April and October of 1900, but only May and September collections appear. The Biltmore herbarium contains no Boynton material at all. Small (l.c.) describes the range of Quercus boyntoni as extending along ridges in the Alleghany Valley into Georgia. I have not seen any specimens of this species from Georgia, but its occurrence there would be expected. The yellow pubescence on the lower leaf surface, referred to by Sar- gent, may be due to the age of the dried specimens. The 1934 collections from Texas match those of Alabama taken in 1900 and 1901. The 1953 Texas collections have (at the date of writing) a more silvery pubescence that may yellow with age. That Quercus boyntoni and Q. stellata are closely related is clearly indicated by the similarity of the fruit, puberulence of anthers, and the dense almost mealy puberulum or tomentulum of the twigs. That they deserve specific distinction, however, is equally plain. The shrubby habit of Q. boyntonz is not in any way related to the growing conditions. Neither the elevation of the Lookout Mountain site (300 m., fide Beadle) nor Sargent’s reference to “‘in the shelter of narrow glades” indicates the de- 1 The occurrence in Angelina County, Texas, of a community named Boynton located a few miles from the Shawnee localities is purely coincidental and in no way related to the name of the species, which was derived from that of its first collector in Alabama. 1956] MULLER: QUERCUS BOYNTONI 229 gree of altitude and exposure that constitutes a rigorous habitat. A nor- mally erect tree species such as Q. stellata (however small of stature in its xeric western range) is not changed by such moderate rigor to a sprawling rhizomatous shrub as little as 20 cm. tall at maturity. The even lesser stature of the lowland Texas population of Q. boyntoni, growing at about 60 m. elevation, further emphasizes the genetic nature of this character. The leaf shape and thin texture of the blades of Q. boyntoni are constant characters in which it differs further from Q. stellata. Although most of the collectors’ references to site conditions in the Alabama populations have stressed “‘shallow soil,” the Texas plant occurs on a deep sandy soil in the valley of a creek. The species was found by Boon and Tharp out on Shawnee prairie and in other prairie locations, but I failed to find it except in the shelter of forest. In the Texas population Q. boyntoni occurs under Pinus taeda and Liquidambar styraciflua mixed with arboreal Q. drummondu Liebm. Although there is some little evidence of hybridization of Q. boyntoni with QO. drummondu (Muller 9605, 9607), the shrub species occurs abun- dantly in pure form associated with the tree species equally free of hy- bridity. Furthermore, saplings of the tree species were noted as erect plants, showing neither the trailing habit nor the leaf characters of Q. boyntoni (Muller 9606). The surrounding clay and gravel hills are abun- dantly populated by typical Q. stellata which is not found on deep sand (Muller 9604). An inconspicuous dark green shrub trailing down banks and mixed with shrubs of taller stature is easily overlooked, especially if its fruit is hidden by the leaves and its leaf form is sufficiently similar to surrounding trees to cause it to be taken for a seedling of these. It is therefore unlikely that the full distribution of Quercus boyntoni is indicated by the available herbarium specimens. In fact, Mrs. Boon indicated (manuscript map in personal communication) four localities in Angelina County in which she had encountered Q. boyntoni, one in the Shawnee Creek drainage and the other three in the Biloxi (Balaxy) Creek drainage. One locality in each drainage is authenticated by specimens here cited. Between Texas and eastern Alabama there very possibly exists a number of sandy creek val- leys similar to that of Shawnee Creek in which QO. boyntoni may be found. University of California, Santa Barbara College and Santa Barbara Botanic Garden LITERATURE CITED Breapie, C. D. 1901. A shrubby oak of the southern Alleghanies. Biltmore Bot. Studies 1:47-48. Mutter, C. H. 1951. The oaks of Texas. Contr. Texas Research Found. 1:21-323. SARGENT, C.S. 1918. Notes on North American trees. I. Quercus. Bot. Gaz. 65:423-459. SMALL, J. K. 1913. Flora of the southeastern United States. 1395 pp. New York. . 1933. Manual of the southeastern flora. 1554 pp. New York. TRELEASE, W. 1924. The American oaks. Mem. Nat. Acad. Sci. 20:1-255. 226 MADRONO [Vol. 13 ON THE GENERIC LIMITS OF ERIOPHYLLUM (COMPOSITAE) AND RELATED GENERA SHERWIN CARLQUIST While recent authors largely agree in definitions of species referable to the genus Eviophyllum (Compositae, tribe Helenieae) or its neighbor- ing genera, the generic disposition of these synonymy-laden species has been less satisfactory. The genera studied here correspond to those grouped by Rydberg (1915) in the subtribe Eriophyllanae. Since species of all of these genera have been subject to repeated reassortment, all of the taxa required examination before any resolution could be attempted. The high degree of disagreement seems to stem from the small num- ber of characters by which these genera may be recognized. In the in- stance of these, and many other highly reduced members of the Com- positae, it may well be that relationships will never be properly under- stood, owing to the fact that only a limited number of characters has been left by evolution for the taxonomist to use. These characters have been taxed to the utmost by the systematist, and decisions have necessarily been arbitrary to a large extent. For this reason, the addition of a cyto- logical character, chromosome number, and a review of morphological characters, particularly in the anatomical details which underlie them, seemed highly desirable. HISTORICAL SURVEY A glance at the historical record will show that this subtribe is a par- ticularly poorly understood one. The genus Eriophyllum, created by La- gasca for a perennial species, EL. staechadifolium, was adopted by Gray (1884) to include all the perennial species, though he first referred these species to Bahia (1876). Subsequent authors have placed all perennial species in Eriophyillum. The chief problem, then, was whether or not to refer annuals similar in aspect to the same generic concept. The genus Actinolepis had been set up by de Candolle to receive annuals possessing pappus (including species of Baeria), while he had erected the genus Monolopia for the epappose species. From the time of Gray on, however, persuasive similarities in aspect of some of the annuals to some of the perennials led to their inclusion in Eriophyllum, rather than Actinolepis or Monolo pia, with the result of converting Eriophyllum into a capricious- ly heterogeneous assemblage poorly differentiated from incoherent groups of annual species left as outliers. Typical of the attitudes involved in this segregation is the description by Greene (1897) of Eriophyllum (now a Pseudobahia) Heermannii as “wholly an Eriophyllum, not only as to habit, but as to the character of the involucre and achenes,”’ while two species of Eriophvilum later, he finds E. ambiguum (considered here as close to the perennials) ‘A plant with more the habit of a Monolopia than any of the foregoing.” 1956] CARLQUIST: ERIOPHYLLUM Zi ERIOPHYLLUM. Gray (1884) added the species ambiguum to the peren- nials forming Eriophyllum, denoting all of these as “section Trichophyl- lum.’’ Once this species had been allowed in Eriophyllum, however, logic demanded the admission of others, and he appended a “‘section Actino- lepis” consisting of the entities Eriophyllum nubigenum, E. multicaule, E. Pringlei, E. lanosum, and E. Wallacei. Greene (1897) restored Actino- lepis as a genus, but to the perennials of Eriophyllum he added E. bahiae- folium and E. Heermannit, formerly considered by Gray under Monolo pia. Rydberg (1915) subtracted these last two species, but added Eriophyllum nubigenum (formerly in Actinolepis), E. Congdonu, a recently named species close to both E. ambiguum and E. nubigenum, and E. minus. This last species, known only from the type specimen, was formerly treated as Monolo pia minor, though it is actually a Baeria (see below). Jepson (1925) restored the full complement of species used by Gray, without sectional distinctions, adding, however, Eriophyllum Heermanni, and a species for which Johnston (1923) created the monotypic genus Ere- monanus, E. mohavense. Constance (1937) treated the genus similarly to Jepson, removing Eriophyllum Heermani to Monolopia and recognizing Eremonanus mohavense. ACTINOLEPIS. While Gray treated this group of pappose annuals as a genus in 1876, he considered it a section of Eriophyllum in 1884. Greene, reviving it as a genus, included A. nubigena, like Gray, though this species is close to Eriophyllum ambiguum, which both regarded as an Eriophyl- lum. Greene’s Actinolepis agrees with Gray’s. Rydberg performed a maxi- mum of segregation among the annuals, leaving the contents of Gray’s Actinole pis not included in Eriophyllum to be distributed among Actino- lepis and a new genus, Anthero peas. Actinolepis contained only A. multicaulis and A. Pringlei, while An- theropeas was created for a close pair of species, A. /anosum and A. Wal- lacei. Dubious of the segregation of numerous small genera, Jepson (1925) and Constance (1937) restored all the Actinolepis and Antheropeas spe- cies to Eriophyllum without sectional distinctions. Mono.opiA; PSEUDOBAHIA. Gray, subsequent to his original treatment of Monolopia (1876) soon recognized (1884) that the contents of this epappose genus could be split into two sections. The tall, virgate species with subentire leaves, M/. major and M. gracilens, became “‘section Mono- lopia,”’ while the low, Eriophyllum-like plants with pinnatifid leaves, WM. bahiaefolium and M. Heermannii, were “section Pseudo-Bahia.” Influ- enced by the Eriophyllum-like qualities of the latter, Greene placed them in Eriophyllum, while Rydberg created a new genus, Pseudobahia, for them; both authors retained the remaining species in Monolo pia. Jepson’s treatment agrees with Greene’s except for placing Eriophvllum bahiae fo- lium of Greene in Monolopia (without sections). Constance, though not dealing with this group directly, accepted Gray’s treatment provisionally. It remained for Crum (1940) to offer a careful and detailed revision of 228 MADRONO [Vol. 13 Monolopia (excluding Pseudobahia). The writer follows her treatment. Crum suggested that Monolo pia minor, a species the type and only collec- tion of which had apparently been seen by none of the authors mentioned, was actually a Baeria. Evidence accumulated by Mrs. Roxana S. Ferris confirms Miss Crum’s opinion and in a recent paper (Ferris, 1955) she has made the necessary nomenclatural changes. Fics. 1-4. Metaphase of somatic divisions in root tips. 1, Psewdobahia Heermannii, Carlquist 302; 2, Eriophyllum multicaule, Carlquist 293; 3, Eriophyllum ambiguum, Carlquist 312; 4, Eriophyllum lanatum var. arachnoideum, Carlquist 336. Fics. 5-8. Meiotic divisions of pollen mother cells. 5, first metaphase, Pseudobahia bahiaefolia, Bacigalupi & Carlquist 4014; 6, diakinesis, Eriophyllum confertiflorum, Carlquist 330; 7, first metaphase, Eriophyllum confertiflorum, Carlquist 325; 8, first meta- phase, Eriophyllum confertiflorum var. tanacetiflorum, Carlquist 327. All X* 1200. SYNTRICHOPAPPUS. The distinctive pappus of S. Fremontii prevented it from being confused with any of the other annuals, though Gray, who named it, realized it was closely related to them. Soon effacing his error of placing a new species, S. Lemmoni (thought epappose until now) in Actinolepis, Gray (1884), with remarkable intuition, designated it as the second species of Syntrichopappus. All subsequent authors have accepted this treatment. The writer’s disposition of these taxa is seen in the table of chromosome numbers, with the exception of Eriophyllum mohavense, which he re- gards as coordinate with E. Pringlei, and E. nubigenum, which is to be placed beside E. ambiguum and E. Congdonii. METHODS To obtain chromosome numbers, young heads, or root tips of plants grown from seed, were fixed in a Carnoy’s solution (3 parts absolute ethyl alcohol: 1 part glacial acetic acid). To assure quick penetration, it 1956] CARLQUIST: ERIOPHYLLUM 229 was necessary to break open the heads before placing them in the fluid, or aspirate them with a vacuum pump directly after immersion. The usual acetocarmine squash technique was employed for both pollen mother cells and root-tips. The designation ‘‘2n” indicates that the number was derived from a somatic division, while “‘n’”’ denotes a count made from meiotic material. Voucher specimens were made at the same time buds were fixed, or, if plants grown from seed were used, specimens of these were prepared. A set of these specimens has been deposited in the Her- barium of the University of California at Berkeley, and replicates of most of these have been distributed to other herbaria. Anatomical information was derived from both cleared and sectioned material. Mature flowers from the writer’s collections preserved in Car- noy’s fluid or from numerous herbarium specimens were cleared in 2.5 per cent aqueous NaOH, dehydrated, and stained in safranin. Heads of plants in the writer’s collections fixed in Carnoy’s fluid were also dehy- drated by means of Johansen’s tertiary butyl alcohol series, embedded in paraffin, sectioned, and stained with a safranin-fast green combination. The writer wishes to express especial appreciation to Dr. Lincoln Con- stance, at whose suggestion the problem was undertaken, for valuable comment and advice. Thanks are due to him and to Dr. Herbert L. Mason, Dr. G. Ledyard Stebbins, Jr., and Miss Annetta Carter for reading the manuscript and offering suggestions. Acknowledgment is also extended to those who provided fixed material and specimens, and whose names appear among the collections listed in Table 1. CYTOLOGY The accompanying table shows that distinctive chromosome numbers characterize the various taxa. Eriophyllum sect. Eriophyllum contains diploids having n = 8 as well as polyploid derivatives. While no diploids were found in E. Jepsonii, E. latilobum, E. Nevinti, or E. staechadifolium, both E. confertiflorum and E. lanatum contain diploids and tetraploids (figs. 4, 6, 7). No correlation was found in the relative size of parts of dip- loid versus polyploid plants. In fact, the polyploids showed more diminu- tive parts in most instances. Pollen-size differences likewise were negli- gible. A strong exception is the extremely robust octoploid E. conferti- florum var. tanacetiflorum (fig. 8). The meagre coverage of the large E. /anatum complex suggests that both diploids and tetraploids may be found independently in at least two of the varieties. Since a better coverage of this group did not seem neces- sary for the purposes of generic definition, the nature and distribution of diploid and polyploid plants presents an interesting subject for further investigation. Likewise, the relative frequence of diploids and polyploids in E. confertiflorum remains to be studied. The annuals of Eriophyllum, considered here as a section, Actinole pis, are characterized by a haploid chromosome number of 7 (figs. 2, 3). The 230 SPECIES Eriophyllum Lagasca section Eriophyllum, sect. nov. (perennials) E. confertiflorum (DC.) Gray E. confertiflorum (DC.) Gray var. tanacetiflorum (Greene) Jepson E. Jepsonii Greene E. lanatum (Pursh) Forbes var. achillaeoides (DC.) Jepson E. lanatum (Pursh) Forbes var. arachnoideum (F.& L.) Jepson E. lanatum (Pursh) Forbes var. grandiflorum (Gray) Jepson E. latilobum Rydberg E. Nevinii Gray E. staechadifolium Lagasca MADRONO COLLECTION Carlquist 330 Carlquist 338 Carlquist 316 Carlquist 318 Carlquist 325 Carlquist 334 Carlquist 327 Carlquist 317 Chisaki 565 Post 105 Balls & Everett 11 October 1952 Gillett 266 Sweeney 8 May 1953 Carlquist 336 Carlquist 324 Heckard & Sweeney 341a Carlquist 305 Carlquist 315 Carlquist 307 Carlquist 309 Carlquist 335 TABLE 1. CHROMOSOME NUMBERS APPROXIMATE LOCALITY Fort Tejon, Kern County* Pasadena, Los Angeles County Avalon, Los Angeles County Arroyo del Puerto, Stanislaus County Coulterville, Mariposa County Midpines, Mariposa County Coulterville, Mariposa County Arroyo del Puerto, Stanislaus County Black Oak Villa, Lake County Alder Point, Mendocino County Lava Beds, Modoc County Lassen National Park, Shasta County Black Oak Villa, Lake County Little River, Mendocino County Coulterville, Mariposa County Oroville, Butte County San Mateo, San Mateo County Avalon, Los Angeles County Castroville, Monterey County Piedras Blancas, San Luis Obispo County Asilomar, Monterey County [Vol. 13 NUMBER n= 8 n= 8 n= 16 nN ==16 hi— 16 m= 16 Ne=—"35i2 n= 16 m=-8 n=8 2n = 32 n==16 n= 16 2n = 116 n=8 n= 16 n= 16 ne— 16 n= 16 n= 16 n= 16 1956] CARLQUIST: ERIOPHYLLUM TABLE 1. CHROMOSOME NumMBERs (continued) Eriophyllum Lagasca section Actinole pis Gray (annuals) E.ambiguum Gray E. Congdonii Brandegee E. multicaue Gray E. Pringlei Gray Antheropeas Rydberg A. lanosum (Gray) Rydberg A. Wallacei (Gray) Rydberg Monolopia DC. M. gracilens Gray M. lanceolata Nutt. M.major DC. M. stricta Crum Pseudobahia Rydberg P. bahiaefolia (Benth.) Rydberg P. Heermannii (Dur.) Rydberg P. Peirsonit Munz Syntrichopappus Gray S. Fremontii Gray S. Lemmonii Gray Carlquist 312 Carlquist 313 Carlquist 329 Gillett 411 Gillett 417 Carlquist 333 Bacigalupi 4117 Carlquist 293 Carlquist 296 Carlquist 337 Carlquist 295 Carlquist 306 Carlquist 289 Carlquist 285 Carlquist 288 Bacigalupi & Carlquist 4014 Carlquist 302 Carlquist 304 Carlquist 287 Carlquist 301 Carlquist 314 Hobo Hot Springs, Kern County Red Rock Canyon, Kern County Fort Tejon, Kern County Edison, Kern County Mt. Breckenridge, Kern County El Portal, Mariposa County Gonzales, San Benito County Oceano, San Luis Obispo County Boron, Kern County Las Vegas, Clark County, Nevada Boron, Kern County Hecker Pass, Santa Cruz County Lost Hills, Kern County Tracy, San Joaquin County Lost Hills, Kern County North of Friant, Madera County Coulterville, Mariposa County Bagby, Mariposa County Ducor, Kern County Stoddard’s Well, Kern County Cajon Pass, San Bernardino County * Unless otherwise noted, all localities are in California. ie, / Zn 14 n= / Dies ne / 1A ef) hi, N=, T=e7, He <7 n= 4 Te=sr0 n= 12 i110 n= {2 == 10 hea 4 h=3 20 = 6 3 n= 18 i=: n= 7 Zot 232 MADRONO [Vol. 13 i) Fics. 9-14. Stamen tips. 9, Antheropeas lanosum, Carlquist 337; 10, Antheropeas Wallacei, Carlquist 295; 11, Eriophyllum Pringlet, Carlquist 296; 12, Pseudobahia Heermannii, Carlquist 304; 13, Syntrichopappus Fremonti, Carlquist 301; 14, Syn- trichopappus Lemmonii, Carlquist 314. All X< 90. two species of Antheropeas show numbers of 5 and 4 respectively. Though similar in habit to the seven-paired Eriophyllum ambiguum, both Pseudo- bahia bahiaefolia and P. Heermannii have the distinctive numbers n = 4 and n =3 respectively (figs. 1, 5). Pseudobahia Peirsoniu, a robust weedy plant, is seen to have n = 8, which may be tetraploid in relation to the n = 4 of this genus. The species of Monolopia, similar to Pseudobahia Peirsonu in habit, may also be polyploid, since their high chromosome numbers of ten and twelve suggest a derivative rather than a basic set. Further evidence would be needed to affirm this suggestion. The two species of the well-differentiated genus Syntrichopappus show n = 7 and n = 6 respectively. 1956] CARLQUIST: ERIOPHYLLUM 233 od mie ey he i" RA Say Badd aE eSae: Red Stee Saas Bs sete rn rhal nts) Sain Coe & oe se, g CPR eee SER ee BES ; pare Aree: on, Sens y Q Fics. 15-18. Upper portions of styles. 15, Eriophyllum staechadifolium, Carlquist 309; 16, Eriophyllum multicaule, Carlquist 293; 17, Antheropeas Wallacei, Carl- quist 295; 18, Syntrichopappus Fremontii, Carlquist 301. All X 85. 234 MADRONO [Vol. 13 MoRPHOLOGICAL AND ANATOMICAL CHARACTERS STAMEN TIPS. The terminal projection of the anther connective takes markedly different forms in the taxa considered here. The deltoid shape, widened above the base (fig. 12) is characteristic of many Helenieae, and is found throughout Eriophyllum sect. Eriophyllum, Pseudobahia, and Monolopia. Occasional specimens of Eriophyllum sect. Eriophyllum may exhibit deltoid tips tapered from the base. In Eriophyllum sect. Actino- lepis, however, deltoid tips widened above the base may be found only in E. ambiguum, E. Congdonii, and E. nubigenum. In these three species, tips tapered from the base may also be found. In the remaining species of sect. Actinolepis, E. multicaule, E. Pringlei, and E. mohavense, tips ta- pered from the base are characteristic (fig. 11). These tips are relatively narrow and cuneate in some specimens of EF. multicaule. Antheropeas, on the contrary, shows conspicuously narrower subulate tips (figs. 9, 10), extremely long in A. lanosum. In a similar way, the species of Syntrichopappus exhibit quite attenuate stamen tips, varying from cuneate in S. Lemmoni (fig. 14) to lanceolate in S. Fremontu (fig. oye STYLE TPs. The tips of the style-halves, frequently termed “style ap- pendages,” have long been recognized of diagnostic value. The predomi- nant shape in the Helenieae is deltoid, tapering abruptly above a fringe of elongate hairs (fig. 15). Within both sections of Eriophyllum this shape is found to be characteristic, though some species show a very much blunt- ed form. E. multicaule has nearly flattened tips (fig. 16), and is the most extreme expression of a blunted tip. Pseudobahia and Monolopia are found to have a deltoid tip such as shown in fig. 15. Antheropeas, on the other hand, has an appreciably more elongate tip, varying from narrowly deltoid to cuneate. Many style-tips of A. lanosum and A. Wallacei are much more attenuate than the one shown in fig. 17. In addition, in both species. of Antheropeas, prominent elongate hairs terminate the style tips. The genus Syntrichopappus shows the most striking difference from the type found in Eriophyllum, the style tips being very much prolonged into a lanceolate form (fig. 18). FLORAL VENATION. Though the floral vasculation of composites is highly stereotyped, certain differences may be found. The disk flowers alone serve for comparison here, since in the taxa studied, wide variation within a single population occurs in ray-flower venation. In the ray corolla, a series of veins runs the length of the corolla, forming interconnections at the tip. While characteristic patterns may be found, these are profoundly altered in depauperate or extremely robust plants. Some species, however, such as Eriophyllum multicaule, E. nubigenum, E. Nevinu, Antheropeas lanosum and A. Wallacei, show few if any interconnections between the veins at the tip of the ray. Disk flowers in the genera studied show an identical venation in the corolla, and either four, five, or six veins in the achene (figs. 19-21). An 1956] CARLQUIST: ERIOPHYLLUM 235 Fics. 19-21. Disk flowers, showing venation. 19, Eriophyllum lanatum var. grandi- florum, Carlquist 324; 20, Syntrichopappus Lemmonii, Carlquist 314; 21, Eriophyl- lum mohavense, Jepsen 17180a. All K 14. ovule trace and two style traces are always present; stamen traces, slightly better developed in the perennials of Eriophyllum, are usually rudimen- tary or absent. The species of Eriophyllum sect. Eriophyllum are charac- terized by four, sometimes five, achene-wall bundles (fig. 19), as are the members of the genus Pseudobahia. Monolopia and Antheropeas have four achene-wall bundles, while Syntrichopappus has five (fig. 20). In Eriophyllum sect. Actinolepis, E. ambiguum, E. Congdon, E. nubigenum, and E. multicaule have four achene-wall bundles, while E. Pringle: and mohavense have five; occasionally six bundles appear in achenes of E. mohavense (fig. 21). The patterns of bundle interconnections are always the same in flowers having four, five, or six achene bundles respectively as shown (figs 19-21). Pappus. Particular interest was focussed on finding if any constant characters besides annual versus perennial habit and 7 versus 8 (16, 32) chromosomes distinguish the annuals, sect. Actinolepis, of Eriophyllum, from the perennials, sect. Erviophyllum. Though differences largely of degree rather than of character may be found, such as the generally more indurate involucral bracts of the perennials, pappus structure offers a 236 MADRONO [Vol. 13 NSE | alle ae eee \ CH A , hy \) (\) D Rh RW : haces Sizes ean Rey ge vaacic oi e enn apele\i| vee = % A ay; YX ) BNisila \\ ait Ny Woy Bete URS S { Lh . we we if 7 VL Nil RURSS WSS CANN SeSU eee KBR SST RS ANS 0 i O47 f \ NSSNE ai UE \ Soy we Y i \ ; Fics. 22-23. Portions of longitudinal sections of achenes, showing pappus and basal portions of style and corolla in section; 22, Eriophyllum Jepsonii, Carlquist 317, palea at left seen in median section; 23, Eriophyllum ambiguum, Carlquist 313. Both x 80. more nearly discrete distinction. In longitudinal section, the pappus base is wider in the perennials (fig. 22), the outer epidermis of the pappus forming a nearly continuous line with that of the achene wall. In the annuals, the achene top is more constricted at the pappus base (fig. 23). This is probably associated with the tendency of the pappus to reflex more widely at maturity in the annuals. Difference may also be seen in the structure of pappus as seen in transection. The paleae of the pappus of annuals vary from simple (fig. 24) to moderately thick (fig. 25), the extremes being represented in these two figures. While pappus paleae such as shown in fig. 25 may also occur in the perennials, paleae with more numerous layers of cells, some of the internal cells smaller, can be found only in the perennial species (fig. 26). Fics. 24-26. Transections of pappus paleae. 24, Eriophyllum multicaule, Carl- quist 293; 25, Eriophyllum Congdonii, Carlquist 333; 26, Eriophyllum Jepsonii, Carlquist 317. All X 100. The traditional differences in external structure of pappus are quite useful in differentiating the smaller genera. While at least a short crown of pappus or pappus-vestige is present in Eriophyllum, the genera Mono- lopia and Pseudobahia lack any such structure. Syntrichopappus, as the name implies, is characterized by a circle of numerous setae united at the CARLQUIST: ERIOPHYLLUM 237 1956] ye01y} pue yeo1y} jo sd1}-9qo] 3 aqn} suo[e aseq }e 3UlI sirey Ie[npurys yeo1y} pue yeoiy} pure yeo1y} pue pa.19}}¥89s e ul podnois -uou snjd ‘paia} aqn} suo[e pala} aqn} suo[e pei} 9qn} Suole pais} saley ‘re[npuerys ‘re[npueyls -yeos ‘1epnpurys -yeos ‘1enpueys -yeos ‘repnpurys -yeos ‘1e[npurys B][O107 Juasqe 10 ae}aS quasqe juasqe UMOII B 0} aevaled Aueul jo Sul Ajeatue Ajaatjue aeayed aeayed 310Ys q1oys snddeg (satseds gudyIV uo sutpusdap) stp ut 4 S‘b v v 9‘S ‘Pp Sy so[pung 9}e[OIIUP| proj}[eap projep ayeaund yuny{q 0} plozep qun[q 0} plojep sdiq 3JA3S aseq WlO1} Sul 9}e[OIIUL] aseq dAoqe aseq aAoqe -19d¥} 10 aAoqe aseq 9A0qe sdi} 0} 9}e9uUNd pousprM ‘piojpfap pauepra ‘ptojfap a}ye[nqns pouspt ‘projep pauspIM ‘projjap UIUIE}S OT AS ‘ou awios Co areas g‘¢ymu ‘77 =U tS 41 ,=.u ‘ol ‘gs =u -owlo1y7s o11}Ua pyieuuridiq p2y}00} pey}00} (sjuvid Maj ® oitud ATIeou 10 paqo[ ‘pyrjeuurd 0} o1IQUa 0} 941}Ua UI 911]U9) posqoT 0} pyteuuid SIARIT [enuue jenuue jenuue jenuue [enuue [etuuased yqeH sndqdog DIYDQopNasd Dig 010u0 Jy spagosay Up s1gajourjop wmnyc yd O1agq -OYDAJUKS UO0T}99S U01}9aS mn yKY F OAT mny syd Oy VUANdAY) GALVTIAY ATASOTD) GNV WOTIAHAGOIY JO NOSTYVAIWO!) °7 ATEV 238 MADRONO [Vol. 13 base. While all authors have considered S. Lemmont epappose, popula- tions of this plant contain a few individuals bearing the pappus typical of S. Fremontti, though the bristles are shorter and less numerous than in S. Fremontii. Despite the prolongation of four or five of the pappus paleae of Antheropeas lanosum into setae, there is no fundamental difference be- tween the genus Antheropeas and the annuals of Eriophyllum in pappus structure. DISCUSSION The various features already discussed as well as some additional points are summarized in Table 2. It will be seen that in addition to the chromo- some number, other differences may be found to support the division of genera which is proposed here. Though the chromosomes of Eriophyllum mohavense and E. nubigenum have not been seen, other features of these rarities, which may now be extinct, make certain their placement in Erio- phyllum sect. Actinolepis. Eriophyllum nubigenum appears to be an alpine extreme closely connected with EL. ambiguum and E. Congdonit. Erio- phyllum mohavense, despite its distinctive heads consisting of four (or three) flowers grouped around a short central projection of the receptable, is closely related to E. Pringlet. These two species agree in their discoid heads, the numerous short pappus paleae, the short, broadly funnelform corolla, morphology of stamen and style tips, etc. The writer does not agree with Johnston (1923) that Erviophyllum (Eremonanus) mohavense must be considered in conjunction with the monotypic genus Dimeresia Gray. Dimeresia seems only superficially similar by virtue of its two-flow- ered heads, while the larger flowers (larger than any of the above, with the exception of a few perennials of Eviophyllum), the prominent setose pappus, curiously involute at the base, the essentially opposite leaves, lacking any indications of lobes or teeth, all seem to remove it from consid- eration with any of the Eriophyllanae, including Syntrichopappus. Al- though Gray originally placed Dimeresia in the Inuleae, the treatment of Cronquist (1955), who includes this genus in the Senecioneae, seems the most acceptable. The annuals of Eriophyllum, despite their differences from the peren- nials in chromosome number, are structurally close to them in such species as E. Congdonit, so that it has not seemed feasible to segregate the diverse contents of this group into still another genus. Consequently, recognition of them as a section of Eriophyllum seems most logical. Antheropeas, however, which is often accepted as part of Eriophyllum, seems to merit separation from the annuals by virtue of its lower chromosome numbers and notably different stamen-apex and style-tip structures. Likewise, the anomalous chromosome numbers of Pseudobahia reinforce the lack of pappus and aggregation of corolla hairs in distinguishing it from Evio- phyllum. Chromosome numbers, habit, leaf characters, and corolla hairs in turn separate Pseudobahia from Monolo pia. 1956] MASON: ELATINE 239 LITERATURE CITED ConsTANCE, L. 1937. A systematic study of the genus Eriophyllum Lag. Univ. Calif. Publ. Bot. 18:69-136. Cronouist, A. 1955. Phylogeny and taxonomy of the Compositae. Am. Midl. Nat. 53:478-511. Crum, ETHEL. 1940. A revision of the genus Monolopia. Madrono 5:250-270. FERRIS, RoXANA. 1955. The identity of Monolopia minor DC. Contr. Dudley Herb. 4:331-334. Gray, A. 1884. Synoptical Flora of North America. Vol. I, Part II. New York. GREENE, E. L. 1897. Flora Franciscana, Vol. IV. San Francisco. Jepson, W.L. 1925. A manual of the flowering plants of California. Berkeley. Jounston, I. M. 1923. Diagnoses and notes relating to the spermatophytes, chiefly of North America. Contr. Gray Herb. 68:80—-104. Muwnz, P. A. 1949. California miscellany—I. El Aliso 2:77-86. RypsBerc, P. A. 1915. Eriophyllanae, in North American Flora 34:81-100 (Monolopia by H. M. Hall). NEW SPECIES OF ELATINE IN CALIFORNIA HERBERT L. MASON In preparing the treatment of the Elatinaceae for a work on the flora of the marshes and ponds of California, it became clear that Elatine pre- sents a perplexing problem in speciation. The differences between aquatic and terrestrial forms of the same species often seem greater than the differences between species. The genus is in need of a thorough cultural study designed to test the nature of characters and their validity as cri- teria of species. In the meantime one is faced with the problem of ‘“‘lump- ing” the various entities in a few long-recognized species and thereby concealing the problem, or of recognizing more taxa in an attempt to at least pose some of the problems in the group. I shall follow the latter course. It seems clear that there is a New World and an Old World facies in the genus as evidenced in the tendency towards reduction of the calyx in the 3-merous species of the New World. Furthermore the fact that in some 3-stamen species the stamens are opposite the carpels and in others alternate, but never are they opposite the petals, suggests marked in- stability in the number of whorls of stamens in the genus. Because of the above problems we find no evidence to support the reference of our 3- merous species to the European E. triandra as has long been the practice. This latter species has a regular 3-merous calyx with all sepals very small in proportion to the corolla. Our plants which have in the past been re- ferred to E. triandra all have 2 large oblong sepals often equalling the corolla in length and the third much reduced or absent. On the other hand the introduced rice field weed, E. ambigua, has three regular sepals. Field study of the group makes one suspicious that both apomixis and cleistog- amy have operated to complicate the pattern of variation. This needs in- vestigation. I have found nothing referrable to E. americana in California. 240. MADRONO [Vol. 13 Elatine gracilis sp. nov. Plantae graciles erectae 2—4 altae foliis op- positis subaequalibus vel internodia paulo excedentibus ad_ bases petiolis similes angustatis stipulis laceratis attenuatis instructis flore nodo quoque unico sessili sepalis duo tertio reducto vel deficiente petalis tribus mem- branaceis fere orbicularibus staminibus tribus vel uno vel nonnumquam deficientibus carpellis alternantibus seminibus rectis vel paulo curvatis in loculo quoque 7—8 areolarum ordinibus 9-10 areolis in ordine uno- quisque 20-30 partibus testae elevatis horizontalibus infra areolarum ordines manifestioribus quam eis longitudinalibus habitu plantarum aqua- tilum terrestriumque simili. Plants slender, erect, 2—4 cm. high; leaves opposite, subequal to slightly longer than internodes, narrowed to a petiole-like base with attenuate lacerate stipules; flowers 1 to a node, sessile; sepals 2, a third sepal re- duced or wanting; petals 3, thin membranous, almost orbicular, stamens 3-1 or sometimes none, alternate the carpels; seed straight to slightly curved, 7-8 per locule, the areolae in 9-10 rows, 20-30 areolae in each row; the horizontal ridges more conspicuous than longitudinal; aquatic and terrestrial plants similar in habit. Type. Terrestrial; 0.4 mile south of Little Truckee River Bridge on Hiway 89, altitude ca. 6300 feet, Sierra County, California, 17 August 1952, Herbert L. Mason 14,494 (UC 985951). Elatine heterandra sp. nov. Plantae procumbentes foliis oppositis 2—4 mm. longis obovatis vel oblongo-ellipticis stipulis hyalinis lanceolatis ca. 1 mm. longis floribus nodalibus solitariis inter se alternantibus trimeri- bus staminibus 3-6 ac quando 3-carpellis oppositis seminibus in loculo quoque 8-12 rectis vel curvatis ad apicem versus rotundatis hilo truncatis apiculatisque areolis in ordinibus 9-10 positis in ordine quoque 12-15. Prostrate plants, leaves opposite, 2-4 mm. long, obovate to elliptic- oblong; stipules hyaline, lanceolate about 1 mm. long; flowers solitary at nodes and alternate one another, 3-merous, stamens 3—6 when 3 oppo- site carpels; seed 8-12 per locule, straight or curved, rounded above, truncate and apiculate at hilum, areolae in 9-10 rows, 12—15 in each row. Type. Pond, 1.5 miles east of Calpine at junction of Calpine with Sierraville-Beckwourth road, altitude ca. 5000 feet, Sierra Valley, Sierra County, California, 19 July 1952, Herbert L. Mason 14,450 (UC 985952), topotype, Mason 14,448 ; Snow’s Lake, Lake County, California, Mason 14547. Elatine obovata (Fassett) comb. nov. Elatine triandra var. obovata Fassett, Rhodora 41: 375. 1939. Elatine obovata has been treated by Fassett as a variety of the Euro- pean E. triandra. It seems more closely rleated to E. brachysperma in seed character while the size and shape of the leaves and the pattern of distribution clearly set it apart from that species. Department of Botany, University of California, Berkeley INFORMATION FOR CONTRIBUTORS Manuscripts submitted for publication should not exceed an estimated 20 pages when printed unless the author agree to bear the cost of the ad- ditional pages at the rate of $15 per page. Illustrative materials (includ- ing “typographically difficult” matter) in excess of 30 per cent for papers up to 10 pages and 20 per cent for longer papers are chargeable to the author. Subject to the approval of the Editorial Board, manuscripts may be published ahead of schedule, as additional pages to an issue, provided the author assume the complete cost of publication. Shorter items, such as range extensions and other biological notes, will be published in condensed form with a suitable title under the general heading, ‘“Notes and News.” Institutional abbreviations in specimen citations should follow Lanjouw and Stafleu’s list (Index Herbariorum. Part 1. The Herbaria of the World. Utrecht. Second Edition, 1954). Articles may be submitted to any member of the Editorial Board. MADRONO A WEST AMERICAN JOURNAL OF BOTANY A quarterly journal devoted to the publication of botanical re- search, observation, and history. Back volumes may be obtained from the Secretary at the following rates: Volume E{1916-1929) ica ae Oo Volume il, FOSO=89S40 0.2 1) te es 5.00 Volume IIT, 1935-1936 Nan eee eae 5.00 Volume IV, 1937-1938 . . .) . 5.00 Volume V 1039-2940... “ey a 5.00 Volume VI, 1941-1942 PRO Sd 5.00 Volume VII, 1943-1944 . . .) , 5.00 Volume VIII. 1945-1946 . . . °°. 5.00 Volume IX, 1947-1948 Sa ee 6.00 Volume X, 1949-1950. . . . . 7.00 Volume XI, 1951-1952 EMO 2 Mi: 8.00 Volume XII, 1953-1954 . . . 8.00 Single numbers, Volumes: land’ 2-U%' 20 Waa 50 Succeeding numbers, (except Volume XL, No: 2) 3" fae 1.00 Volume: XT Nov2 2) este 1.50 The subscription price of MADRONo is $4.00 per year. If your institution does not now subscribe to MADRONO, we would be grateful if you would make the necessary request. Since there is no extra charge for institutional subscriptions, an individual may hold membership in the California Botanical Society on the basis of his institution’s subscription. Address all orders to: G. THomMAS Rossins, Corresponding Secretary Department of Botany University of California, Berkeley 4, California bi \J0 *¥a “4 i ADRONO VOLUME 13, NUMBER 8 OCTOBER, 1956 Contents PAGE Notes ON MAtvaceEae. VIII. EREMALCHE, Thomas H. Kearney 241 “~~ MUHLENBERGIA BRANDEGEI, A NEW SPECIES FROM BAJA CALIFORNIA, MEXICO, AND ITS RELATION- SHIP TO MUHLENBERGIA BILOBA, Charlotte G. Reeder 244 PHYTOSEROLOGY VERSUS GENEALOGY IN ZEA Mays, J. F. Davidson and T. L. Thompson 252 JosEPH BuRKE IN 1853, R. Kent Beattie 259 A NEw GossypiuM FROM MicHoAcaNn, MExIco, Howard Scott Geniry 261 A SUMMARY OF THE NOMENCLATURE OF DOUGLAS-FIR, PsEupDOTSUGA MEnzissiI, Vladimir J. Krajina 265 NoTEs AND NEws: AN EXTENSION OF RANGE FOR TRIFOLIUM OWYHEENSE GILKEY, Helen M. Gilkey; MEMORIAL ENDOWMENT FuND; NEws 268 INDEX ee ee 269 A WEST AMERICAN JOURNAL OF BOTANY PUBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY MADRONO A WEST AMERICAN JOURNAL OF BOTANY Entered as second-class matter at the post office at Berkeley, California, January 29, 1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price $4.00 per year. Published quarterly and issued from the office of Madrofio, Herbarium, Life Sciences Building, University of California, Berkeley 4, California. BOARD OF EDITORS HERBERT L. MAson, University of California, Berkeley, Chairman Epcar ANDERSON, Missouri Botanical Garden, St. Louis. LymMAN BENSON, Pomona College, Claremont, California. HERBERT F’, COPELAND, Sacramento College, Sacramento, California. Joun F. Davipson, University of Nebraska, Lincoln. IvAN M. JoHNsTON, Arnold Arboretum, Jamaica Plain, Massachusetts. MitprepD E. Martuias, University of California, Los Angeles 24. MARION OWNBEY, State College of Washington, Pullman. Ira L. Wiccrns, Stanford University, Stanford, California. Secretary, Editorial Board — ANNETTA CARTER Department of Botany, University of California, Berkeley. Business Manager and Treasurer—Matcotm A. Nogs Carnegie Institution of Washington, Stanford, California CALIFORNIA BOTANICAL SOCIETY, INC. President: Wm. M. Hiesey, Carnegie Institution of Washington, Stanford, Cali- fornia. First Vice-president: Roger Reeve, Western Regional Research Laboratory, Albany, California. Second Vice-president: Lewis Rose, California Academy of Sciences, San Francisco, California. Recording Secretary: Mary L. Bowerman, De- partment of Botany, University of California, Berkeley, California. Corresponding Secretary: G. Thomas Robbins, Department of Botany, University of California, Berkeley, California. Treasurer: Malcolm A. Nobs, Carnegie Institution of Wash- ington, Stanford, California. 1956] KEARNEY: EREMALCHE 241 NOTES ON MALVACEAE. VIII. EREMALCHE Tuomas H. KEARNEY The genus Eremalche was published by Greene (1906, p. 208) with a few words of description and citation of three species previously re- ferred to Malvastrum — M. rotundifolium A. Gray, M. exile A. Gray, and M. Parryi Greene. A fourth species, Eremalche kernensis C. B. Wolf (1938, pp. 66, 67), was published subsequently. Jepson (1925, p. 633 and 1936, pp. 496, 497) referred the three species cited by Greene to Sphaeralcea, but they have been retained in Malvastrum in other recent floras (Munz, 1935, p. 308; Kearney et al., 1951, p. 548). It remained for Wiggins (1951, pp. 96, 97) to recognize Eremalche as a valid genus and to describe it formally, with a key to and descriptions of the four species. In my synopsis of the American genera of Malvaceae (Kearney, 1951, pp. 119, 120), Evemalche was treated as a section of Malvastrum. Now, however, I find myself in agreement with Wiggins as to its validity as a genus. Krapovickas (1954, pp. 609, 610) had arrived at the same conclu- sion, pointing out that in Erematlche exilis the diploid number of chromo- somes is 20, hence a multiple of 5, as was found to be the case in Sphaeral- cea (Eusphaeralcea) by Webber (1936, pp. 314, 315) in fifteen North American species and by Krapovickas (1949, p. 185) in thirteen South American species. On the other hand, in Malvastrum, as I have sought to re-define the genus (Kearney, 1955), Krapovickas (1954, p. 610) found the basic chromosome number to be 6 (2n = 12, 24, 36). He considered Eremalche to be nearer Sphaeralcea than Malvastrum, but mentioned a character stated to be unique in Tribe Malveae — adherence of the carpel- wall to the seed. As treated herein, Sphaeralcea is the “Eusphaeralcea”’ of an earlier publication (Kearney, 1935), Malvastrum the genus as restricted in a recent paper (Kearney, 1955), and Eremaiche as defined by Wiggins (1951). The more important distinguishing characters of these three genera may be stated as follows: SPHAERALCEA. Carpels (Kearney, 1935, pls. 9-12) 1 to 3-ovulate, apical section smooth, dehiscent, with ventral notch at base, the basal section indehiscent, reticulate. Annual or perennial herbs, the flowers commonly thyrsoid-paniculate, the corolla grenadine-red, varying to mauve-pink (drying violet) or nearly white, also yellow or orange in a few species. Basic chromosone number 5. MALVASTRUM. Carpels (Kearney, 1935, pl. 2, figs. E, F, H) 1-ovulate, little differentiated apically and basally but with an (often deep) ventral notch, reniform-suborbicular to almost triangular in outline, incurved- Madrono, Vol. 13, No. 8, pp. 241-272, October 31, 1956. NOV 1 2 1956 242 MADRONO [Vol. 13 rostrate, radially rugose to nearly smooth on the sides, usually pubescent above, with or without dorsal and subapical cusps or tubercles, slightly and tardily dehiscent apically. Perennial (sometimes annual?) herbs or undershrubs, the inflorescences various, the corolla yellow or orange. Basic chromosome number 6. EREMALCHE. Carpels (Kearney, 1935, pl. 2, fig. G) 1-ovulate, undiffer- entiated apically and basally, obscurely or shallowly notched ventrally, orbicular or nearly so, muticous, bordered by a radially rugose or reticu- late-rugose cushion-like thickening, with smooth or faintly reticulate lateral walls, the carpels indehiscent or tardily and irregularly dehiscent by rupture of the thin lateral walls. Small annual herbs, the flowers soli- tary in the axils, often long-pedunculate, the corolla whitish to mauve- pink (drying violet). Basic chromosome number 5, at least in E. exzilrs. Sphaeralcea is widely distributed in western North America, from southern Canada to Mexico, and is also well represented in southern South America. Malvastrum ranges from southern Arizona to Florida and the West Indies, south to Bolivia and Argentina. EHremaiche is confined to the southwestern United States and northern Baja California, perhaps also northern Sonora. On the basis of my observations on herbarium specimens, a key to the four taxa in Eremalche recognized by Wiggins is presented here. The term “taxa” 1s used because I am uncertain that E. kernensis can be maintained as specifically distinct from EH. Parry, but in the present state of knowledge I am not prepared to reduce it to subspecific status. KEY TO THE TAXA 1. Leaves merely coarsely crenate, reniform-orbicular; herbage, involucel, and calyx hispid with long, simple or 2-3 furcate hairs; petals with conspicuous basal spots; carpels 3—3.5 mm. in greatest diameter, thin, flat, rugose-reticulate dor- sally and subdorsally and often more finely so laterally, black at maturity. Petals (15?) 20-25 mm. long, lilac or mauve-pink, drying violet. Southern California and Nevada, western Arizona.......... E. rotundifolia (A. Gray) Greene 1. Leaves palmately 3—5-lobed or -parted; herbage, etc., shortly stellate-pubescent and often with longer simple or sub-simple hairs but all of the pubescence rela- tively soft and fine; petals without basal spots; carpels not more than 2 mm. in greatest diameter, somewhat turgid, radially rugose dorsally and subdorsally, smooth laterally, pale— to dark-brown at-maturity.. 2). eee (2) 2. Petals not more than 6 mm. long, little surpassing the calyx; peduncles in fruit up to 4 cm. long but usually much shorter; leaves mostly cleft to about the middle. the lobes shallowly few-toothed or nearly entire. Herbage, etc., shortly stellate- pubescent but occasionally with a few longer, simple or subsimple hairs; calyx- lobes lanceolate or ovate-lanceolate, attenuate-acuminate; corolla whitish or pale lavender. Southern California to southwestern Utah, Arizona, and northern Baja Cai ORT A ste cscs, oeFs, Gr as eR A he Pe EE en eerie E. exilis (A. Gray) Greene 2. Petals 8 mm. or longer, much surpassing the calyx; peduncles in fruit often much more than 4 cm. long; leaves usually more deeply dissected and the lobes with more numerous and longer teeth:or lobules... ee ee (3) 1956] KEARNEY: EREMALCHE 243 3. Herbage normally with few or many, relatively long, simple or 2—3-furcate hairs in addition to the short stellate pubescence; calyx-lobes normally ovate and abruptly long-acuminate, but exceptionally lanceolate and attenuate-acuminate ; petals mauve-pink, drying violet, 13-26 mm. long. Inner Coast Ranges of south- ern California, principally in western Kern County....£. Parryi (Greene) Greene 3. Herbage shortly stellate-pubescent, occasionally with a few longer hairs; calyx- lobes lanceolate or ovate-lanceolate, attenuate-amuminate; petals whitish or pale lavender, 8-13 mm. long. Apparently restricted to western Kern and San Luis Obispo counties,;Califormia.. 2... en eee tee E. kernensis C. B. Wolf Eremalche rotundifolia is a remarkably uniform species and no inter- gradation with the other three taxa has been observed. The latter, how- ever, as indicated in the key, seem to intergrade. The possibility that EZ. kernensis originated as a hybrid between £. Parryi and E. exilis is sug- gested. It approaches E. exzlis in character of the pubescence, shape of the calyx-lobes, and corolla-color, but is nearer . Parryi in leaf-shape and size of the corolla. Specimens collected recently by Ernest C. Twis- selmann in the Temblor Valley region, western Kern County, California, where the type of E. kernensis was found, show various combinations of the characters of E. Parryi, E. exilis, and E. kernensis. Some of these are probably edaphic variants, as the habitat varied from dry sandy situa- tions to clay-loam and more or less saline flats. It is probable that biosystematic investigations are needed to solve the problems of this interesting little genus. California Academy of Sciences, San Francisco, California. PUBLICATIONS CITED GREENE, EpwArp L. 1906. Certain malvaceous types. Leafl. Bot. Obs. and Crit. 1: 205-209. Jepson, Wiriis Linn. 1925. A Manual of the Flowering Plants of California. Assoc. Students Store, Univ. Calif., Berkeley. (Malvaceae pp. 626-636.) . 1936. A Flora of California. Vol. 2. California School Book Depository, San Francisco. (Malvaceae pp. 482-506.) KEARNEY, THomMAS H. 1935. The North American species of Sphaeralcea subgenus Eusphaeralcea. Univ. Calif. Publ. Bot. 19:1-128, pl. 1-12. . 1951. The American genera of Malvaceae. Am. Midl. Nat. 46:93-131. . 1955. Malvastrum A. Gray.—a re-definition of the genus. Leafl. West. Bot. 7:238-241. KEARNEY, THOMAS H., Ropert H. PEEBLES, and collaborators. 1951. Arizona Flora. Univ. Calif. Press, Berkeley. (Malvaceae pp. 536-553.) KRAPOVICKAS, ANTONIO. 1949. Las especies de “Sphaeralcea” de Argentina y Uru- guay. Lilloa 17:179-222. . 1954. Estudio de las especies de “Anurum,” nueva seccion del género “Urocarpidium” Ulbr. (Malvaceae). Darwiniana 10:606-—636. Mouwnz, Puitrip A. 1935. A manual of southern California botany. J. W. Stacey, Inc., San Francisco. (Malvaceae pp. 303-311.) WEBBER, J. Mitton. 1936. Chromosomes in Sphaeralcea and related genera. Cyto- logia 7:313-323. Wiccins. Ira L.. 1951. Zn Abrams, Ill. Flora Pacific States, vol. 3. Stanford Univ. Press, Stanford. (Malvaceae pp. 82-112). Wo tr, Cart B. 1938. California plant notes. II. Occasional Papers Rancho Santa Ana Bot. Garden, ser. 1 no. 2:44-90. 244 MADRONO [Vol. 13 MUHLENBERGIA BRANDEGEI, A NEW SPECIES FROM BAJA CALIFORNIA, MEXICO, AND ITS RELATIONSHIP TO MUHLENBERGIA BILOBA CHARLOTTE G. REEDER Early in the year 1889, Townshend Stith Brandegee, botanist and plant collector, set out on the first of several expeditions into Baja Cali- fornia, Mexico. Leaving California by boat, he entered the peninsula at Magdalena Bay in mid-January and apparently spent about six weeks in this general area collecting the plants which he found there. In late Feb- ruary he was joined by two other naturalists from the California Academy of Sciences, Walter E. Bryant and Charles D. Haines. Shortly after the arrival of these men, the three colleagues visited Santa Margarita Island where they spent a week studying and collecting the flora and fauna. Santa Margarita, together with a sister island, Magdalena, lies just off the western coast of southern Baja California and these two islands form the seaward protection to Magdalena Bay. This was not only Brandegee’s first, but apparently his only visit to Santa Margarita, an island which remains relatively little known botanically even today. General remarks concerning the trip and the collecting were published by Bryant (1889), his report being concerned principally with the bird life of the area. (For further details of the Brandegee itinerary see: Setchell, 1926; Ewan, 1942; Moran, 1952.) Brandegee, in 1889, published a list of the plants which he had collected early in that year, and in 1891 he decribed, in a general way, both Magdalena and Santa Margarita islands. One of the plants from Santa Margarita Island, a small annual grass listed by Bran- degee (1889) as Bealia mexicana, is the subject of the present paper. The geography and geology of these islands was discussed by Lindgren (1890) and Nelson (1921). Of chief concern to us here is a statement by Lindgren that the highest point on Santa Margarita Island is 1900 feet. Nelson (1921) devoted a small section of his account of “Lower Califor- nia and Its Natural Resources” to the various coastal islands, and there is a short discussion of Santa Margarita. He indicates that the island is mostly barren and desert-like with little plant life. Such vegetation as does exist is primarily in the low middle area and is, for the most part, similar to that of Magdalena Island. He also notes that although these two islands lie close together and are relatively close to the mainland, nevertheless there are several forms of small mammals peculiar to Santa Margarita. Although Brandegee (1891) entitled his second article, “The Plants Peculiar to Magdalena and Santa Margarita Islands,” he included no list of species, but dealt with the geology and some phytogeographic rela- tions of the plants of the area. Since, as he states, the rocks of the high- lands of both islands differ in geological composition from those of the nearest mountains of the peninsula, he expected that there might be a 1956] REEDER: MUHLENBERGIA 245 considerable number of local endemics even though the two islands are relatively close to each other and to the mainland. This, however, did not prove to be the case, as most of the plants appeared to be the same as those he had encountered in various parts of the peninsula. That there is some degree of endemism, however, is indicated by the fact that Bran- degee found a few species, largely inhabitants of the rocks, that he had not collected elsewhere, and most of them proved to be undescribed. Although he identified the major part of his collections himself, Bran- degee sent some groups of plants to authorities for determination. George Vasey of the United States Department of Agriculture identified the Gramineae of the 1889 trip and his report was included by Brandegee (1889, pp. 210-214). Here Vasey placed “Bealia mexicana Scribner — Santa Margarita Island” between the species of Muhlenbergia and Sporo- bolus. This entry is made more conspicuous than the others by the foot- note accompanying it: (p. 212). “This grass was first collected by Mr. C. G. Pringle in Chihuahua, Mex., in October, 1886. It was named by Prof. F. L. Scribner, as a new genus, in honor of Prof. W. J. Beal, of the Michigan Agricultural College. I am not aware that a description has been published. It is considered by Prof. Hackel as forming, with Clomena Beauv., a section of Muhlenbergia. It appears to have pretty strong claims to specific [generic] distinction.” The inclusion with Clomena Beauv. was apparently based upon the fact that both possess a bidentate lemma which is awned from between the teeth. This character alone, however, is misleading, and the two com- plexes are not closely related. In Bealza both glumes are 1-nerved, while in Clomena the second is 3-nerved and 3-toothed. Beauvois (1812) re- alized this and stressed it in both his description! (p. 28) and in his figure (pl. 7, fig. 10). Species belonging to the Clomena complex include Muhlenbergia peruviana (Beauv.) Steud. and its close relatives. The publication of the binomial Bealia mexicana apparently dates from Brandegee’s paper (1889, p. 212) in which Vasey listed the grasses from Lower California. It should be noted, however, that it is here only a nomen nudum. Hackel (1887, p. 47) makes no mention of the genus Beailia in the text, although Clomena is included as a synonym of Muhlen- bergia. In his “Nachtrag” (op. cit. p. 97), however, we find the following note: “S. 47 zu Miuhlenbergia: Clomena Beauv.. wozu auch Bealia Scribn., bildet eine Untergattung mit 2zahniger, aus dem Ejinschnitte begrannter Decksp., in Mexiko und Peru.” 1 Beauvois states that the lower glume is tridentate, but this is obviously an error. His figure correctly shows that the first glume is entire and 1-nerved, while it is the upper (second) which is 3-nerved. Niles (1925, pp. 162, 163) translates the Beauvois manuscript, but makes no mention of this discrepancy. 246 MADRONO [Vol. 13 This addendum accounts for the comment inserted under Muhlenber- gia in Scribner’s translation of the Hackel manuscript (1890) and the subsequent consideration of Bealza as a synonym of Muhlenbergia. That Scribner did not agree with Hackel’s disposition of Bealza, however, is evi- dent from his comment in a footnote under Muhlenbergia on p. 104 of the translation. In this footnote Scribner discusses at some length the dis- tinctions between that genus and Bealza, and ends with the statement: ‘“,. Tcan but consider it | Bealia| generically distinct.” In this footnote the genus is described only superficially and no specimens are cited, but an original figure (45a) by Scribner shows the spikelet and its various parts in detail. The legend under the figure bears the name “‘Bealia Mexi- wcana, Scribner (provisional.)”’ It was not until six years later that an adequate description of Bealia mexicana was published, and this by Beal (1896, pp. 267, 268) himself, although he credits the name to Scribner. In the same publication Beal makes the new combination Bealia speciosa which is based on Muhlen- bergia speciosa Vasey. These two specific epithets are the only ones which have ever been associated with the generic name Bealia. Our present con- cern is only with Bealia mexicana, since B. speciosa appears to be a valid Mexican perennial species of Muhlenbergia (see Hitchcock, 1935, p. 457). In the amplified diagnosis of Bealia mexicana, Professor Beal states that the plant is ‘“‘an elegant tufted perennial, 20-40 cm. high.” A Pringle specimen (no. S19) from the State of Chihuahua, Mexico, is the only col- lection cited. Actually specimens of this collection, which are still extant, show that the plants were certainly annual in growth habit. The chief characters which served to separate Bealia from Muhlenber- gia were the deeply bifid lemma, between the lobes of which an awn arises, as well as the long, faintly-nerved glumes, which commonly exceed the lemma in length. These characters are present also to a greater or less degree in those Muhlenbergia species which are sometimes rcognized as the distinct genus Epicampes Pres}, all of whose species are stout peren- nials. Despite its annual habit, if Bealia mexicana were not to be recog- nized as a member of the genus Muhlenbergia, it would appear to belong to the genus Epicam pes, a disposition which was made of it by M. E. Jones (1912). At present, however, Bealza is not recognized, and Epicampes usually is considered as little more than a section of Muhlenbergia. Hitchcock (1913), reporting on the Mexican grasses in the United States National Herbarium, attempted to clarify the situation by (1) transferring Bealia mexicana to Muhlenbergia; (2) pointing out that the combination Muhlenbergia mexicana was not tenable since the epi- thet mexicana had been preempted in the genus by M. mexicana (L.) Trin. in 1824; (3) proposing a substitute name, M. biloba; and (4) citing the collections then presumed to represent the species, which in- cluded the type (Pringle 819) from Chihuahua; Pringle 10147 from Dur- ango; and Brandegee in 1889 from Santa Margarita Island. 1956] REEDER: MUHLENBERGIA 247 In a later work, Hitchcock (1935, pp. 436, 437) gave the type locality as “Chihuahua City, (Pringle 819)” and the distribution as “Gravelly hills, Baja California to Durango.”’ I. M. Johnston (1943), apparently following Hitchcock’s lead, said of Muhlenbergia biloba: “‘Known from a few stations in Baja California, Chihuahua, and Durango.” Conzatti (1946) in his “Flora Taxonomica Mexicana,” evidently fol- lowing Hitchcock, includes Muhlenbergia biloba and cites the same three localities. It is interesting to note that Conzatti, in the synonymy, gives the impression that Bealia mexicana Scribn. ex Beal (1896) is based on Muhlenbergia mexicana (L.) Trin. This, as we have seen, is not true: rather, it is the reason for the new binomial. Pringle collected many replicates and consequently his specimens are distributed rather widely. The Brandegee collection, on the other hand, apparently was more restricted, and to my knowledge specimens are to be seen in only two herbaria. In addition to the three collections cited by Hitchcock as Muhlenber- gia biloba, there is yet a fourth, overlooked by most authors, but brought to my attention by an entry in Pringle’s journal (Davis, 1936, p. 36): “Oct. 7, (1886). A great day for us. Rising at daybreak at 6:40 A.M. we were on our way to the northwestern hills. There we secured two species of Cyperus, 811 C. amabilis Vahl and 810 C. Hochstetteri Nees, and a large supply of a strange grass of which last year I found but two specimens in a wash — 819 Bealia mexicana Scribner.” In the itemized list of Pringle’s specimens (Davis, 1936, p. 340), there are three separate numbers given under Bealia mexicana. In addition to 819 (type) and 10147 (cited by Hitchcock, 1913), there is a third, 507. A sheet in the Pringle Herbarium bearing this number (507), which I have examined, contains two small plants. This collection, which is from the same general locality as the type, is apparently that referred to by Pringle in the quotation above and thus represents all of the material of this species which he found in 1885. While all of the Pringle collections appear to be conspecific, the one from Durango (10147) differs slightly from the other two. It has longer hairs on the glumes, shorter hairs on the palea, and the glumes and floret are about equal in length. These slight differences, how- ever, seem to be well within the range of variability of the species. While the specimens from Chihuahua and Durango are certainly con- specific, the Brandegee collection from Baja California, even though ap- parently related to Muhlenbergia biloba, differs in morphological details which appear to be of such a magnitude as to represent a distinct species. The geographical ranges of the two taxa moreover appear to be quite dis- tinct. Since a new specific epithet is required, I should like to name this new taxon in honor of the collector, T. S. Brandegee, a Yale alumnus, who did so much to further the botanical knowledge of Mexico and Cali- fornia. 248 MADRONO [Vol. 13 Fic. 1. Muhlenbergia Brandegei C. G. Reeder. A, habit, X 14; B, panicle branch with spikelets; C and D, two views of ligule (B—D, x 5); E. glumes; F, floret; G, lemma flattened out, adaxial view (E-G, X 10). From type specimen (Brandegee 16 in 1889). Muhlenbergia Brandegei sp. nov. Annua, 15-25 cm. alta; culmis teretibus purpureis striatis plus minusve glabris erectis vel adscendenti- bus, nodis glabris fuscis inferioribus multiramulosis; vaginis glabris vel minute scabris quam internodiis plerumque brevioribus, marginibus scari- osis; ligule membranacea erosa circiter 0.7—0.8 mm. longa sed marginibus in dentes subulatos ad 1.5 mm. longis extendentibus; laminis planis vel apicem versus plus minusve subinvolutis 3—4 (raro ad 7) cm. longis 1—-1.5 (raro ad 2) mm. latis, subtus scaberulis, supra sparsim pnubescentibus; panicula contracta pallido-viride 3-10 cm. longa, rhac:. ‘a, ramis 1956] REEDER: MUHLENBERGIA 249 = SIE Sos Fic. 2. Muhlenbergia biloba Hitchc. A, habit, <%; B, panicle branch with spike- lets; C and D, two views of ligule (B—D, 5); E. spikelet; F, floret (E-F, x 10). From type collection (Pringle 819). adscendentibus e base plerumque densifloris; pedicellis angularibus ad- pressis 1-3 mm. longis; spiculis circiter 3 mm. longis; glumis subaequali- bus 1-nerviis plus minusve glabris sed sub lente minute papilloso-asperis; lemmate circiter 2.5-2.8 mm. longo quam glumis paullo breviore dense pubescente (pilis albis rigidis basim versus brevibus apicem versus gra- datim longioribus), apice acuminato bifido, dentibus ad 1 mm. longis; arista flexuosa scaberula 7—8 mm. longa e apice lemmatas bifida exserta; antheris circiter 1.mm. longis; caryopsis 1.4 mm. longa. Lower CavirorniA, Mexico: Santa Margarita Island, T. S. Brande- gee 16 (UG yoe; US). March 6, 1889. 250 MADRONO [Vol. 13 TABLE 1. SUMMARY OF CHARACTERS DISTINGUISHING MUHLENBERGIA BILOBA Glume length Glume surface Glume nerve Lemma length (including teeth) Lemma surface Lemma teeth Lemma nerves Anther length Pedicels Panicle Ligule length Flowering time Distribution FROM M. BRANDEGEI M. Brandegei (see fig. 1) 3 to 3.5 mm. glabrous (minutely papillose- roughened under a lens) prominent, green about 2.5 to 2.8 mm. densely stiff pubescent on lower 14 to %’s;_ hairs white, short below, longer above acuminate to aristate, erect 3 prominent, the nerves evi- dent even in the teeth when opened out 1mm. angular contracted, branchlets and pedicels stiffly appressed along the axis; the main branches ascending or ap- pressed, mostly spikelet- bearing for entire length 1 to 1.5 mm. March (plants mature) Baja California: Santa Mar- garita Island. Not over 1900 feet elevation M. biloba (see fig. 2) 4to 5 mm. pubescent, especially on the lower 34 indistinct 3 to 355 mm: sparsely appressed tawny pu- bescent on about lower 34 = rounded, broadly toothed, spreading 1 prominent, the lateral nerves indistinct above, more evident on lower 1 2mm. terete somewhat open, with branch- lets and pedicels ascending to spreading, = flexuous; spikelet-bearing on termi- nal half 2.5 to 3 mm. September and October Chihuahua and Durango, in the mountains, elevation from 4700 to 6800 feet As pointed out above, Muhlenbergia Brandegei, which appears to be endemic to Santa Margarita Island, has been confused with VM. biloba Hitchc. from the time of its discovery. For this reason, the complete synonomy of M. biloba is given below, along with a list of the specimens examined. I believe that this short list represents all the collections that have been made of this species. MUHLENBERGIA BILOBA Hitchcock, Contr. U. S. Nat. Herb. 17: 294. 1913; N. Am. Fl. 17 (6): 436. 1935. Bealia mexicana Scribner ex Vasey, in Brandegee, Proc. Calif. Acad. Sci. II. 2: 212. 1889, nomen, in Scribner & Southworth, The True Grasses p. 103, fig. 45a. 1890, nomen, ex Beal, Grasses N. Am. 2: 267. 1896, descr. [basis of Muhlenbergia biloba Hitchc.]. Epicampes mexicana (Scribner ex Beal) M. E. Jones, Contr. West. Bot. 14: 7. 1912. Not Muhlenbergia mexicana (L.) Trin. 1824. 1956] REEDER: MUHLENBERGIA 251 Specimens examined: MEXICO. Curuanva: [vicinity of Chihuahua City], thin soil of porphyry mountains, October, 1886, Pringle 819, US (type), F, GH, MO, NY, VT, UC (not seen), YU; hills west of Chihuahua, October 21, 1885, Pringle 501, VT; Majalca [north of Chihuahua City], September 16, 1935, Le Sueur Mex-026, CAS (not seen), GH, MO, UC, US. Duranco: Barranca below Sandia Station, 6800 feet elevation, October 12, 1905, Pringle 10147, CAS (not seen), F, MO, NY, UC (not seen), US, VT. [According to Davis (1936, p. 228) on October 12, 1905, Pringle was on the Mesa de la Sandia at the base of the Sierra Madre Range. | Although Muhlenbergia biloba and M. Brandegei resemble each other in certain characters, they are apparently not as closely related as was once supposed. The important differences are summarized in Table 1. ACKNOWLEDGMENTS I am indebted to the curators of the following herbaria for the loan of specimens: the University of California at Berkeley (UC), Chicago Natural History Museum (F), Gray Herbarium (GH), Missouri Botani- cal Garden (MO), New York Botanical Garden (NY), U. S. National Herbarium (US), and Pringle Herbarium of the University of Vermont (VT). Especial thanks are due the head curators of the United States National Herbarium and of the New York Botanical Garden for making the facilities of their respective institutions available on numerous occa- sions, and my husband, John R. Reeder, for his endless help throughout the course of the study. Osborn Botanical Laboratory, Yale University, New Haven, Conn. LITERATURE CITED BEAL, W. J. 1896. Grasses of North America. New York: Henry Holt & Co. Vol. II. The grasses classified, described, and each genus illustrated, with chapters on their geographical distribution and a bibliography. pp. viii + 706. illus. Beauvois, A. M. F. J. Partsor pr. 1812. Essai d’une Nouvelle Agrostographie; ou Nouveaux Genres des Graminées; avec figures représentant les Caractéres de tous les Genres. Paris. pp. Ixxiv + 145. illus. BRANDEGEE, TOWNSEND S. 1889(1890). A collection of plants from Baja California, 1889. Proc. Calif. Acad. Sci. II. 2: 117-216. . 1891. The plants peculiar to Magdalena and Santa Margarita Islands. Zoe 2: 11-12. BRYANT, WALTER E. 1889(1890). A catalogue of the birds of Lower California, Mex- ico. Proc. Calif. Acad. Sci. II. 2: 237-320. ConzaTTI, CASIANO. 1946. Flora Taxonomica Mexicana (Plantas Vasculares). Tomo I.: Clave analitica de familias Pteridofitas 0 Helechos, Monocotiledoneas mona- periantadas. Mexico, D. F.: Sociedad Mexicana de Historia Natural. xxxv + 378. (Grasses pp. 148-327.) Davis, HELEN Burns. 1936. Life and work of Cyrus Guernsey Pringle. Burlington, Vermont: University of Vermont. 756 pp. illus. Ewan, J. A. 1942. Bibliographical miscellany—IV. A bibliographical guide to the Brandegee botanical collections. Am. Midl. Nat. 27: 772-789. HackEL, E. 1887. Gramineae (echte Graser). In Engler, A. & K. Prantl, Die natiir- lichen Pflanzenfamilien. Leipzig. 7(2): 1-97. illus. Hircucock, A. S. 1913. Mexican grasses in the United States National Herbarium. Contr. U. S. Nat. Herb. 17(3): 181-389. + index. (See pp. 294-295.) 22 MADRONO [Vol. 13 . 1935. (Poales) Poaceae (pars). N. A. Fl. 17(6) : 419-482. (See pp. 431-432.) JounstTon, I. M. 1943. Plants of Coahuila, Eastern Chihuahua, and adjoining Zaca- tecas and Durango, II. Jour. Arnold Arb. 24: 375-421. Jones, M. E. 1912. New species and notes. Grass notes. Contr. West. Bot. 14: 1-21. LINDGREN, WALDEMAR. 1890. Notes on the geology and petrography of Baja Cali- fornia, Mexico> Proc. ‘Cahf. Acad. Sci. IT. 33 25-33. Moran, Rein. 1952. The Mexican itineraries of T. S. Brandegee. Madrono 11(7): 253-262. NELSON, Epwarp W. 1921. Lower California and its natural resources. Mem. Nat. Acad. Sci. 16(first Memoir): 1-171. pl. 1-35. Nizes, Corneria D. 1925. A bibliographic study of Beauvois’ Agrostographie. Contr. U.S. Nat. Herb. 24(6): 135-214. + index. SETCHELL, WiiittamM A. 1926. Townshend Stith Brandegee and Mary Katherine (Layne) (Curran) Brandegee. Univ. Calif. Publ. Bot. 13(9): 155-178. pl. 13, 14. SCRIBNER, F. LAMson & Erie A. SoUTHWoORTH. 1890. The true grasses. Translated from Hackel, Eduard. Die natiirlichen Pflanzenfamilien. New York: Henry Holt & Co., viii + 228. illus. PHYTOSEROLOGY VERSUS GENEALOGY IN ZEA MAYS? J. F. DAVIDSON AND T. L. THOMPSON For many years the ideas of phytoserology, as expounded by Mez (1936) and by Chester (1937), have intrigued the senior author as being potentially highly significant in the field of plant taxonomy. Discussions with the junior author, and with his colleague, Dr. Warren Engelhard, stimulated a desire to attempt some phytoserological tests. While it is obvious from the literature that phytoserology has yielded taxonomic data, (1.e., data which give evidence of proximity of relationship), there is no assurance that these data are valid. In short, do the data obtained from serological studies accurately reflect the genealogies of the individ- uals in question? This was our problem. In order to set up a test for the validity of phytoserology in taxonomy, it was necessary to find plants of known genetic (genealogical) relation- ships. It was felt that checking members of the same genus against mem- bers of other genera in the same plant family, or against members from other families, would not be accurate enough, since such relationships are assumed, not definitely known. It is true that such assumptions are based upon many data, and are probably valid, but the authors wished to test definite, not probable, genealogies. With the thought that Zea Mays had been bred for many years, the authors approached the corn breeders to see if they could obtain some strains of known genealogy. Many of their stocks were of presumed closer or more distant relation- ship, but the genealogies of some were known. Through the gratefully appreciated cooperation of the Department of 1 This work was financed by a grant from the University of Nebraska Research Council, whose cooperation is hereby gratefully acknowledged. 1956] DAVIDSON AND THOMPSON: PHYTOSEROLOGY 205 Agronomy of the University of Nebraska, we obtained six strains of corn in which the genealogies were known. In the interest of simplicity, the agronomists selected sample No. 1 as the reference sample, so that the others were more or less distantly related to No. 1. We arranged with the agronomists that after the serological relationships of the six samples had been determined, a check would be made against the recorded geneal- ogies of the samples involved. While the principles of serology are available in any standard textbook, such as Boyd (1943) or Rabat and Mayer (1948), the senior author has considered it advisable, in the interest of completeness, to include the following resumé of the serological principles as they apply to plant tax- onomy. Furthermore, since the great majority of taxonomists are not fa- miliar with the methodology, it'is presented here in considerable detail. The basis for phytoserology rests in the thesis that although the total protein constitution of a plant is antigenically different from that of any other plant, certain proteins are found in common in closely related in- dividuals. Although each plant specimen is considered to possess some proteins that are peculiarly its own as an individual, still, genetic processes pass common proteins along to their offspring. Thus, those plants which are more closely related genetically are presumed to have a greater num- ber of proteins in common. A protein, when injected into the blood stream of a laboratory animal, may stimulate the formation of antibodies in the serum. The injected protein is termed the antigen. Injections of the same protein at frequent intervals may increase the concentration of the antibodies in the blood serum. If, after a “‘rest from injection” period, the serum is withdrawn from the animal and the original protein (antigen) is added to the serum, the anti-bodies present in the serum react with the antigen. In some instances, this reaction may express itself visually as agglutination or precipitation. In other instances it is necessary to employ “indicator” systems to detect the reaction. If the proteins are efficient as antigens, and if they are present in sufficient amounts, either the serum or the antigen may be greatly diluted (1:10,000) and still give a discernible reaction. Thus, unknown antigens may be checked against the antibodies present in the serum by making a series of dilutions for each unknown antigen, adding a standard amount of serum, and seeing at what dilutions the re- action is apparent. The greater the dilution showing a reaction, presum- ably the greater the number or amount of common proteins. For antigen injection, it is desirable to employ an isotonic dilutent to avoid possible shock to the experimental animals. Thus, antigens which are insoluble in the dilutent used are not available for testing. The most common dilutent employed is physiological saline solution. However, when plant materials are extracted in a dilutent they frequently contain sub- stances which are toxic to the test animals. Such toxins are commonly re- moved by one or several pre-extractions with some other solvent, such as ethyl alcohol or sulphuric ether. The solutions resulting from such pre- 254 MADRONO [Vol. 13 extractions are discarded, and the plant material is then extracted in the physiological dilutent. In certain tests in immunology it is desirable to inject either a single protein or a small number of proteins. Hence, the original extract is nor- mally purified by successive pre-extractions with various organic solvents, thus eliminating the undesirable fractions. However, in taxonomic work, there is no method of predetermining which proteins, if any, characterize the various phyletic lines. It is prefer- able, therefore, to have both the maximum amount and maximum number of proteins present. For this reason, in taxonomic work it is desirable that pre-extraction procedures, as shown in Chart 1, should be kept toa minimum. CuarT 1. SEQUENCE OF PRE-EXTRACTION PROCEDURES Proteins in A* Oneida a ———— plant Proteins Proteins Soluble lost in — Canine alcohol Proteins lost in ether Proteins remaining after pre-extraction in alcohol and ether *A = Proteins insoluble in saline. From this chart it can be seen that not all the proteins in the plant are necessarily soluble in saline. We have no method of determining or testing fraction A. The maximum we can test is the amounts or kinds present in the saline solution. Each succeeding pre-extraction reduces the number and/or kinds of protein tested. If the characteristic proteins of a gene- alogical lineage are not present in the antigen, the serological results can- not be indicative of the lineage. PROCEDURE PREPARATION OF ANTIGENS. Corn was selected for the experimental plants because of the known genealogies. Since the corn fruit is a genetic mixture, as far as tissues are concerned, with fruit coat which is strictly maternal, endosperm which is largely maternal, and embryo which is half maternal and half paternal, only the young plant, derived from the em- bryo was used. It was felt that the use of the grain itself would introduce probable errors. 1956] DAVIDSON AND THOMPSON: PHYTOSEROLOGY 255 The six samples of corn were planted in vermiculite, 50 plants per flat, one flat per sample. They were harvested after 38 days and each plant was uprooted and the whole plant was cut up with shears into one-fourth to one-half inch pieces. Any remnants of the original grain were dis- carded. From each sample, 400 grams of plant material were used. This material was pulverized, 25 grams at a time, in 200 ml. of physiological saline in a Waring Blendor for 45 seconds. The resultant mixture was squeezed through cotton toweling and the debris was discarded. (The authors now feel that it might have been wise to have allowed the mixture to extract overnight in a refrigerator before the above rough filtration, as a possible method of increasing the protein concentration. ) The extractions were held overnight at 5°C. and were then centrifuged 30 minutes at 2,000 r.p.m. After centrifuging, the residue was discarded and the supernatant was passed through a sterile Seitz filter. The filtered extract was transferred to 200 ml. rubber-capped serum bottles, and 1:10,000 Merthiolate was added as a preservative. Extracts were stored at O°C. until needed. Nitrogen determinations were made by the colorimetric method of Miller and Miller (1948). Prior to the serological experiments, each sample was adjusted to contain 0.16 mg. of nitrogen per ml. The six sam- ples varied as shown: TABLE 1. VARIATION IN NITROGEN CONTENT OF EXTRACT Sample No. Ninmg./ml. Diluted to 1 0.20 8 in 10 Z 0.40 8 in 20 3 0.50 8 in 25 4 0.16 undiluted 5 0.24 8 in 12 6 0.18 8 in 9 TrEst ANIMALS. Preliminary experiments indicated that mature rabbits did not produce a good antibody titer in response to the corn proteins. Good titers were, however, obtained when rabbits two months of age were employed. INJECTION OF ANTIGEN. The animals were injected every fourth day with the amounts of plant extract No. 1 shown in Table 2. Four injection routes were employed to determine the most efficient in building up the antibody titer (see Table 4). These were: intravenous, intraperitoneal, intramuscular, and subcutaneous. The intravenous and intraperitoneal routes proved to give the highest titers, and since these titers were both approximately equivalent, the in- traperitoneal route was employed as being the easier to administer. TuBE AGGLUTINATION TEsTS. For preliminary titrations, 5 ml. of blood were withdrawn from the terminal ear vein, and allowed to clot at room 256 MADRONO [Vol. 13 TABLE 2. INJECTION ROUTES AND QUANTITIES Millimeters injected Intra- Intra- Day Intravenous peritoneal* muscular* Subcutaneous 0.2 ml. 0.5 ml. 0.5 ml. 0.5 ml. 4 0.4 1.0 LO 1.0 8 0.8 2:0 2.0 2.0 12 1.0 2.0 2.0 2.0), 16 2.0 5.0 5.0 5.0 25 Animals bled, preliminary titer determined. Sf 440) 5.0 51 Final bleeding. *Addition of an adjuvant (mucin) did not enhance the antibody titer. temperature. The clots were broken, and the serum was separated by centrifugation at 3,000 r.p.m. for 30 minutes. The serum was serially diluted with physiological saline as follows: 0, 1:2, 1:4, 1:8 up to 1:256. One ml. of each dilution was pipetted to a serological tube, one ml. of antigen diluted 1:10 was added, and the tubes were shaken vigorously. The tubes were then incubated at 37°C., and observed at intervals over an 8-hour period. A reciprocal test was also set up using antigen dilutions and a constant amount of a 1:2 dilution of serum. RING PRECIPITATION TEsTS. Small amounts of undiluted serum were introduced into 5 cm. micro tubes with a bore of 2 mm. By means of a capillary pipette, each antigen dilution was carefully overlaid on the sur- face of the serum. The tubes were incubated at 37°C., and checked at in- tervals for the characteristic precipitin ring at the interface. Titers obtained from these two methods were disappointingly low. In no instance did the titer exceed 1:32. In order to determine the effect of incubation temperature on aggluti- nation, tubes were incubated at 5°C., 25°C., and 37°C. Incubation tem- perature was found to have no effect on the final titer. It had originally been planned to employ agglutination tests, and to run cross-agglutinations to determine more accurately the relationship of the corn samples. However, in view of the low titers obtained, cross-ag- glutination reactions would have been of little value. Therefore, it was decided to check for the presence of complement-fixing antibodies. COMPLEMENT FIXATION REAcTION. The theory of complement fixation is predicated upon the following facts: 1. Complement, a protein, is always present in normal serum. 2. Certain immunological reactions require the presence of complement. 3. These reactions are not visible. 4. Hemolysin, an antibody produced by injecting sheep red blood cells (RBC) into rabbits, will lyse sheep red blood cells only in the pres- ence of complement. 1956] DAVIDSON AND THOMPSON: PHYTOSEROLOGY 295i) Where complement-fixing antibodies are produced, their presence is detected in the following manner. The serum is heated for 30 minutes at 55°C. to inactivate its complement, and some antigen plus an optimal amount of pure complement is added. This mixture is incubated to allow the antigen and complement-fixing antibodies to tie up the complement. Then anti-sheep hemolysin, in known amount, and sheep red blood cells are added and incubated for 30 minutes at 37°C. If all of the complement has been tied up by the antigen-antibody reaction, none will be available to react with the hemolysin to lyse the red blood cells. If only a small amount of complement-fixing antibodies is present, some of the comple- ment will be available, and lysis will occur. The tube containing the low- est dilution of serum which shows complete lysis is taken as an indication of the titer. TITRATION OF HEMOLYSIN FOR OPTIMAL AMOUNT. Using 1 ml. of 1:30 dilution of purified complement, a series of dilutions of hemolysin were added to determine the dilution that gave complete lysis in 30 minutes at 37°C. This amount of hemolysin is one unit. Two units were used in the final test. TITRATION OF COMPLEMENT FOR OPTIMAL AMOUNT. Using two units of hemolysin, dilutions of complement were made to determine the dilu- tion that gave complete lysis in 30 minutes at 37°C. This amount of complement is one unit. Two units of complement were used in the final test. A further complication arising in the present work was due to the fact that the corn extracts apparently contained some substance which inacti- vated or destroyed the complement. This was controlled by diluting the antigen 1:1 and by adding sufficient complement to offset this reaction. TABLE 3. COMPLEMENT FIXATION TEST ON CORN SAMPLES 2 units Dilution Antigen* complem. Hemolvsin Red blood Tube ofserum Diluted1:1 1/30dil. Saline 2 units cells 3% z vo J 0. 25 ml. 1.75 ml. oO. mi. 0.25 ml. 025ml. 5 2 1112 E _ : ; o > 1:96 O 2 6 1:192 1 s A 1:384 e me 8 1:768 zg z 9 121536 © & a 10 1:3072 as S} Hemol. control 0 1.75 5 iD = Comp. control 0 1.75 1.00 0 - Sensit. cont. 0 ) & 2.50 0.25 0.25 = Cell control 0 0 ee eS 0 0.25 3 Serum control 0 HATES x, Bfies 0.25 0.25 ha *The first ten tubes were repeated for each antigen. 258 MADRONO [Vol. 13 | TABLE 4. RESULTS OF COMPLEMENT FIXATION TEST | STARE Ress OC a NLS Antigen Control 1/12 1/24 1/48 1/961/192 1/384 1/768 1/1836 1/3072 Titer ne ee ee EE ee 300 ttt HOE Ft tH 708 4 +++ -— = = = = + fe 4 (444) 3072 5 tthe Ht EH) Ft 1536 6 tte — — — — + (f4+) $44 4+ +44 384 Hemol. +++ Comp. — Sensit. — RBC — Serum +++ +++ complete lysis a complete inhibition ( ) titer Best results were obtained when the antigen and serum were incubated with the complement at 5°C. for 20 hours. After the preliminary titer from the injection of antigen had attained a value of over 1:3,000, the other antigens were tested as shown in Table 3. The results of this test appear in Table 4. From the above it would appear that, as far as similarity of proteins is concerned, samples 1 and 4 are most closely related, and that the re- mainder are, in order of similarity, 5, 3, with 2 and 6 being equally distant from 1. This might be shown graphically as: 1 2 42s? s ae ee DISCUSSION After the above results had been obtained, the genealogy of the sam- ples was requested from the Department of Agronomy. The samples represented the following plants: 1. Dent corn N 6 2. Dent corn hybrid L 289 « I 205 3. Dent corn 38-11 4. Pop corn SA 24 5. Dent corn N 6 (from a sister plant to sample 1) 6. Dent corn hybrid L 289 « I 205 (from the same plant as 2) The relationship of these plants, as given by the geneticists who know the genealogies is as follows: 1 and 5 are very closely related, being sister plants; next is the pure line, 3; then 2 and 6, being from the same plant; and most distantly related is the popcorn, sample 4. This can be shown graphically as: 3 Teas 3 6 4 It is interesting to notice that, with the exception of the popcorn (4), the serology reflects the genealogy quite well. While the serology was able 1956] BEATTIE: BURKE 259 to distinguish between the sister plants (1 and 5), it was unable to dis- tinguish between fruits from the same individual (2 and 6). These latter results would indicate that the methodology was sound, but because of the exception noted above, it would appear that serology of itself cannot be accepted as an adequate criterion of relationship. Within known groups, such as the dent corn samples utilized, it appears to be valid — at least from this limited series of experiments. As with most taxonomic tools, it would seem that serological data should be correlated with evidence from other fields to be significant. SUMMARY Six samples of corn of known genealogy were tested by serological meth- ods to determine whether or not the serological relationships were an ac- curate reflection of the genealogical relationships. The serological pro- cedures are given. Results indicate that within the dent corns the serology did deflect the genealogical relationships, but the inability of serological techniques to distinguish between a dent corn and a popcorn demonstrates that serology of itself cannot be used as a valid clue to genealogical rela- tionships in every case. Department of Botany, and Department of Bacteriology, University of Nebraska, Lincoln LITERATURE CITED Boyp, W. C. 1947. Fundamentals of Immunology, 2nd Ed. Interscience Publishers, London, England. 503 pp. CuesTER, K. S. 1937. A critique of plant serology. Quart. Rev. Biol. 12: 19-46, 165-190, 294-321. KasaTt, E. A. and M. M. Maver. 1948. Experimental Immunolchemistry, 1st Ed. Chas. C. Thomas, Chicago. 567 pp. Merz, C. 1936. Morphologie and Serodiagnostik. Bot. Arch. 16: 1-23. Miter, G. L. and E. E. 1948. Determination of nitrogen in biological materials. Analyt. Chem. 20: 481-488. JOSEPH BURKE IN 1853 R. KENT BEATTIE In her recently published work on the history of botanical exploration west of the Mississippi (McKelvey, Susan Delano. Botanical Exploration of the Trans-Mississippi West 1790-1850. Arnold Arboretum, pages 792-817. 1955), Mrs. McKelvey gives a very excellent and adequate ac- count of the botanical explorations of Joseph Burke in western Canada and the western United States in the years 1843 to 1847. For years, little has been known by botanists about Burke’s work except that he collected some plants around “Fort Hall” in what is now known as southern Idaho. We had learned that there was material about Burke at Kew, but no one 260 MADRONO [Vol. 13 had secured copies of it for study. Mrs. McKelvey obtained photostats and microfilms which are now at the Gray Herbarium. She traces his ex- plorations in America from 1 March 1844 to 6 November 1847. During this period he was east of the Rocky Mountains in Canada till the autumn of 1844, then crossed the Rocky Mountains and went down the Columbia to Old Fort Walla Walla and east to Fort Hall not far north of what is now Pocatello, Idaho. It has been assumed that he then returned to Eng- land. On page 817 of her book, Mrs. McKelvey says: “Tt would be interesting to know what Burke did in later years but the curtain falls at this point on what seems (to me) to have been a sorry story, for the unfortunate Burke certainly.” The author of this note wishes to report that there is in the New York Botanic Garden Herbarium a specimen collected by Joseph Burke in western Missouri in 1853 and sent to Dr. Torrey accompanied by a letter from Burke to Torrey. Torrey named the plant Hzbiscus militaris and it is today filed under that name in the Garden Herbarium. The letter is on file in the Library. The Garden very kindly had a photograph of the specimen and a photostat of the letter prepared for me, and for the cour- tesy I thank it and especially the Head Curator, David D. Keck, and the Librarian, Miss Elizabeth C. Hall. The letter from Mr. Burke to Dr. Torrey follows: Arrd. Oct. 14th Harrisonville, Cass Co., Mo. Sept. 29th, 1853 Dr. Torrey, Professor of Botany esq. Sir: I take the liberty of writing you concerning a plant I found a few days since. Although I am a total stranger to you — you are well known to me by repu- tation — knowing your great love of Botany encouraged me to write freely to you concerning the plant that interests me so much — had I your work I have no doubt I could satisfy myself without giving you trouble for I can scarcely hope it is a new sps. As near as I can remember it is Hibiscus — at present I have no work to refer to — About five years ago I lost all my books. They were fre'ghted & insured from New York to Oregon City. I have traced them to San Francisco and no farther. I have no doubt they have been burnt in that place — As they are insured with the Atlantic Mutual Insurance Co. I have no doubt I shall get my money after a time, and then be able to make up my book losses again — The Hibiscus? I forward you is not a good specimen. It was about the last flower of the season — I was several miles from home seeking horses, when I found it. As I had no proper means of saving it, it is very much damaged — I found it growing in a moist very rich alluvial soil, by the side of a prairie water hole, that is supplied by the rains — It is a perennial sending up many shoots of about 5 feet 6 inches in length — About the Christmas of the year 41 I found a very pretty Hibiscus in the interior of South Africa near the Cashan Mountains which the Dutch immi- grants call Macali’s berg from a chief named Macali that lives in that region 1956] GENTRY: GOSSYPIUM 261 of country — Near the place I found the Hibiscus & amongst the mountains I found many fine plants — the Hibiscus and a few others are figured in Hook- er’s continuation of Curtis. I think in the year 43 several are figured in Sir W. J. Hooker’s Icones plantarum. One is a leguminous tree which Sir W. J. H. has honored me by calling after my name. I have seen none of these figures, as I have scarcely been in England since. When the gold was first discovered in California, I went there — I spent all the time collecting, that I could not mine — My Herbarium was all lost — I sent a quantity of seeds to Europe according to former promises — I also sent a collection of reptiles and Coleopterous insects to the British Museum — I suppose they were also lost as I have heard nothing of them since — I have sent several collections of Coleoptera to Milby, the great Entomologist of Liver- pool — As I have never heard anything of them since — I have great reason to believe they were also lost — If you have a friend or acquaintance that is interested in Coleoptera I have a small bottle at his service, which I have collected on my own farm, & would feel happy to forward the first opportunity. I am sir, with much respect, your most obedient servant. Joseph Burke Evidently Burke was in California in 1849 or 1850 and later moved to Harrisonville, Cass County, Missouri and lived there on the 29th of Sep- tember 1853. This “raises the curtain” for Burke for five years, ten months and twenty-three days. Piney Branch Farm, Glen Mill Road, Route 1, Rockville, Maryland. A NEW GOSSYPIUM FROM MICHOACAN, MEXICO Howarp Scott GENTRY Gossypium lobatum sp. nov. Arbor 6—9 m. alta; truncus ca. 2 m. longus, 15-20 cm. diam. cortice griseo; rami badii, graciles, patuli, pauci- punctati; lamina folii 3—5-lobata, cordata, 7-12 cm. longa et subaequilata vel latior, stellato-pubescens, apice, acuminata; pedunculi 7-9 mm. longi, glabri, bracteolis ovatis obtusis, glabris, persistentibus, 4-5 mm. longis, lignosis; calyx 5-lobatus, campanulatus, 20-22 mm. longus, lobis 9—10 mm. longis, triangulo-lanceolatis; corolla punicea, 5—5.5 cm. longa, tubi- formis, intra ad basim purpurea; capsula 3-loculata, 25-30 cm. longa; semina oblonga angustissima ca. 1 cm. longa dense et longe pilosa pilis crispatis pallidis. Slender tree 6—9 m. tall with single stem ca. 2 m. high, 15-20 cm. in diameter, bark light gray on trunk, dark or reddish-brown on branches and young stems with conspicuous scattered lenticels, these transversely elongate or 1-2 mm. broad parallel to horizontal ridges; branches slender, flexuous, spreading, the nodes remote, sparsely dotted with light-colored 262 MADRONO [Vol. 13 round lenticels, glabrate, the young twigs stellate-pubescent; petioles stellate-pubescent, 2—4 cm. long, abscising at base; leaf blades 3—5-lobed, broader than long, mostly 7-12 cm. long (from apex of petiole) , 8-14 cm. broad, the central lobes acuminate, stellate-pubescent above and below, paler below; venation primarily 5-palmate, but lower pair conspicuously weaker and frequently anastomosing above without lobation; flowers axillary in the 2--3 distal nodes, 1—-several at a node; pedicels 7-9 mm. long, glabrous, abscising at the base; bracteoles ovate-obtuse from a tri- angular base, glabrous, striate and gland-dotted, 4-5 mm. long, persis- tent, lignified in age; calyx at anthesis campanulate, 20-22 mm. long, the tube ca. 12 mm. long, densely long-stellate-pubescent, the lobes 9-10 mm. long, triangular-lanceolate, equal, ochroleucous, sparsely stellate- pubescent, sparsely gland-dotted; corolla 5—5.5 cm. long, tubular in form with strictly imbricate petals, finely gland-dotted and stellate-pubescent externally, light pink with a deep purple fimbriated spot covering lower half within; androecium 2.5—3 cm. long; apical filaments longer than the proximal ones; capsules 3-locular, 25-30 cm. long, narrowly ovoid, acu- minately to obtusely beaked, densely rugose-punctate towards apex, less so below, the valves rather thin; seeds narrow, elongate, short-linted with fibers 2-4 mm. long, ca. 10 mm. long, 2—2.5 mm. in diameter at middle, tapered, the raphe forming a naked groove nearly as long as the seed; seed coat dark brown, smooth, waxy. Type. Canyon del Marques at about 500 m. elevation, along highway between Uruapan and Apatzingan, Michoacan, Mexico, October 22, 1952, Gentry 12314 (U.S. Nat. Herb. 2189410). Additional material used in the diagnosis from the same population is fruiting material collected May 12, 1951, Gentry 10450 (U.S. Nat. Herb. 2189412). Seed of this collec- tion grown at Iguala, Guerrero, produced the flower, Kerr s.n. (U.S. Nat. Herb. 2189413). The specific name, lobatum, refers to the lobes of the calyx, which are conspicuously larger than those of other members now known in the genus. This species belongs in Hutchinson’s section Erioxyla (1947) and is closely related to Gossypium aridum (Rose and Standley) Skovsted (1934, p. 422; 1935). It is distinguished from the latter by: (1) the lack of spur-branchlets, (2) the broad lobate leaves (vs. entire small leaves), (3) the glabrous pedicel and involucel (vs. densely pubescent), and (4) the large campanulate, deeply lobed, long-pubescent calyx (vs. smaller, cupulate, short-lobed, short-pubescent). It is distinguished from G. aridum and from all other species of Gossypium by the yellowish-green foliage, by the large deeply lobed, long-pubescent calyx, and by the slen- der elongate seeds, to list only the more distinct characters. Table 1 com- pares its calyx characters with closely and distantly related species. It is of interest that the polyploid G. hirsutum with n = 26 chromosomes more closely approaches the lobatum calyx in size and lobing than any other 1956] GENTRY: GOSSYPIUM 263 Fic. 1. Gossypium. A-G, Gossypium lobatum: A, foliage branch, xX 4%. B, sec- tion of bark, X 2. C, flower, x 1. D, calyx, X 1. E, involucel, < 1 and xX 2. F, cap- sule, X 1. G, seed, X 1 and X 2. H, Gossypium aridum, calyx, X 1. 264 MADRONO [Vol. 13 species. Gossypium barbadense also has 26 chromosomes, while the rest of the listed species are considered to have but 13 chromosomes. Calyces are compared at the anthesis stage as the persistent calyx of Gossypium is accrescent and frequently distorted by capsular growth. Measurements were taken from dried herbarium material of several speci- mens in each case, except for G. /obatum, where only one flower was avail- able. Gossvpium lobatum and G. aridum are together distinguished by densely pubescent calyces. The remainder are essentially glabrous, having in a few cases a few trichomes towards the base or distally on the lobes, as in G. hirsutum. It appears that all American species can be easily sep- arated by the characters of the involucel and calyx alone. TABLE 1. CALYX FEATURES IN SOME AMERICAN SPECIES OF GOSSYPIUM Tube Lobe Lobe Relative Species length length regularity pubescence lobatum 12 mm. 10 mm. equal long-pubescent aridum 5-8 1-3 variable short-pubescent armourianum 5 1-2 ie glabrous harknessii 5-7 1-3 ss - thur beri 2-3 0-1 $3 davidsonii 4-5 0 t hirsutum 7-10 2-4 subequal ys barbadense 6 0 s Gossypium aridum and G. lobatum, which have distinct single boles, are the only true tree forms known in the genus. Among American species they are also distinguished by the pink corolla, the elongate androecium (which they share in lesser degree with the Australian section Sturtiana), the simple persistent lignified bracteoles, and the slender long-pubescent seeds. Gossypium lobatum is more extreme in the latter character. Gos- svpium aridum is Populus-like in habit and is reported by one collector to reach 15 m. in height (Gentry 5498), exceeding its associates in the Sinaloa Thorn Forest. Gossypium lobatum in the juvenile stage is a low, spreading, broad-leaved shrub, a habit form quite distinct from the slen- der, monopodial, smaller-leaved, erect tree it becomes in maturity. The writer also observed it a few miles west of Apatzingan, where it was a scat- tered small-tree component of the semiarid Short-tree Forest over the dark, heavy, argillaceous soil derived from volcanic rocks. Further ex- ploration in the Balsas River basin would doubtless find it present be- yond the environs of Apatzingan. Both these trees are drought deciduous and become leafless during win- ter. The area of G. avidum in central Sinaloa is visited in some years by light winter rains. In such seasons the trees may respond with a flush of leaves, which for lack of moisture remain pauperized. Both trees flower 1956] KRAJINA: PSEUDOTSUGA 265 in the late winter and spring during the prolonged dry season while they are generally leafless, releasing their seeds previous to the summer rains beginning in June. Plant Introduction Section, Horticultural Crops Research Branch, Agricultural Research Service, United States Department of Agriculture, Plant Industry Station, Beltsville, Maryland LITERATURE CITED Hutcuinson, J. B., R. A. Strow and S. G. STEPHENS. 1947. The evolution of Gossy- pium. Oxford University Press. 160 pp. SxovsTepD, A. 1934. Cytological studies in cotton. II. Two interspecific hybrids be- tween Asiatic and New World cottons. Jour. Genet. 28:407-424. . 1935. Some new interspecific hybrids in the genus Gossypium L. Jour. Genet. 30:447-463. A SUMMARY OF THE NOMENCLATURE OF DOUGLAS-FIR, PSEUDOTSUGA MENZIESII VLADIMIR J. KRAJINA Epiror’s Note. In January, 1956, the Editors of MADRONO received from Dr. Krajina a manuscript presenting a careful and exhaustive treatment of ‘“‘The No- menclatural Problem of Common Douglas-Fir,” and substantiating the adoption of the combination PsEupotsuGA Menzirsi1 (Mirb.) Franco, made in 1950. While Dr. Krajina’s manuscript was in the hands of the Editorial Board, a series of papers, treating various facets of the same subject and finally reaching the same conclusion as had Dr. Krajina, appeared in “Taxon” [5(1, Febr.): 4—7, 18, 19, 1956; 5(2, Apr.): 38-39, 1956; 5(3, May): 41-43, 43-46, 1956]. Dr. Krajina has consented, therefore, to the deletion of the first two-thirds of his manuscript in which he covered in great detail the history of the problem. The balance of his paper, in slightly revised form, is published herewith.—Eb. There are not many species nomenclaturally so interesting as the com- mon Douglas-fir. A resume of the events involved with the first collec- tions made of this tree sheds light on some of the sources of the confusion concerning the proper name which should be applied to it. About the discovery of Abies Douglasii (Sabine mss.) Thomas Nuttall (1865) wrote as follows: “This species was originally discovered by Mr. Menzies at Nootka Sound, in 1797, during the voyage of Captain Van- couver, and from a specimen without cones or flowers was published a description by Mr. Lambert, under the name of Pinus taxifolia, . . .” Sargent (1898), who, in his “Silva of North America,” accepted the name Pseudotsuga mucronata (Raf.) Sudw., makes the following statement about its early history, “Pseudotsuga mucronata was discovered in 1791 on the shores of Nootka Sound by Archibald Menzies, the surgeon of Van- couver in his voyage of discovery; it was first described in the journal of Lewis and Clark. Rediscovered by David Douglas in 1827, it was intro- duced by him into the gardens of Europe, where it has become one of the 266 MADRONO [Vol. 13 best known and most valuable coniferous trees for park plantations.”’ Similarly, Little (1952; 1953) published that Menzies’ type-specimen of Pinus taxifolia Lamb. was collected at Nootka Sound on Vancouver Island in 1791. However, H. J. Elwes and Augustine Henry (1909) pub- lished the following reliable historical data: ‘“‘The Douglas fir was dis- covered by Menzies at Nootka Sound in 1797. Seeds were, however, first sent home by Douglas in 1827, from which plants were raised by the Horticultural Society of London and distributed throughout the country.” The first (illegitimate) name, Pinus taxifolia Lamb. (1803), was based upon the collection made by Menzies in 1797. Similarly, the second (ille- gitimate) name, Abzes taxifolia Poir. (1805), as well as the correct spe- cific basonym, Abies menziesu Mirb. (1825), are based upon Lambert’s description of the material collected by Menzies. In 1832 were published two synonyms of Abies menziesii Mirb.: the first, Pinus douglasit Sabine ex D. Don in Lamb., was based on the original collection by Menzies, complemented by the fruiting material sent to Europe in 1827 by David Douglas; the other synonym is the name pub- lished by Rafinesque in the autumn of 1832, Abies mucronata. It was based upon the field description by Lewis and Clark (cf. Coues, 1893), the specimens collected by Meriwether Lewis in 1805 and 1806 having been sent to Pursh (1814) who accepted the name Pinus taxifolia for them. Thus, common Douglas-fir was discovered by Archibald Menzies in 1797 at Nootka Sound on Vancouver Island. Another collection was made by Meriwether Lewis on the banks of the Columbia River in 1805-1806. From the same region, cones and other material were collected by David Douglas and sent to Europe in 1827. On these three collections rest the eighteen names that have been applied to this tree. The complete synonymy [omitting that pertaining to Pseudotsuga menziesii var. glauca (Beissn.) Franco, which was thoroughly discussed by Little (1953) | follows: PsEUDOTSUGA MENZzIESII (Mirb.) Franco, De Coniferarum duarum nominibus. 4. May, 1950. — Franco, Cedrus libanensis et Pseudotsuga menziesii. Bol. Soc. Brot. (Coimbra), sér. 2, 24: 74. July, 1950. — Franco, Notas sobre a flora lenhosa da Mata do Bug¢aco. Bol. Soc. Brot. (Coim- bra), sér. 2, 25: 206. 1951. — Little, Check list of native and naturalized trees. Agr. Handb. 41: 305. 1953. Pinus taxifolia Lamb., Descr. Genus Pinus 1: 51, pl. 33. 1803; nomen illegit. Non Pinus taxifolia Salisb. Prodr. 399. 1796, a synonym of Abies balsamea (L. 1753) Mill. 1768. Abies taxifolia | (Lamb.) | Poir. in Lamarck, Encycl. Meth. Bot. 6: 523. 1805; nomen illegit. Non Abies taxifolia Mus. ex Du Tour, Nouv. Dict. Hist. Nat. 20: 114. 1803, pro syn. (a name not validly published). Non Abies taxifolia Desf. Tabl. Ecole Bot. Mus. Hist. Nat. 206. 1804, a syno- 1956 | KRAJINA: PSEUDOTSUGA 267 nym of Abies alba Mill. 1768. Non Abies taxifolia Raf., New Flora N. Am. 1: 38. 1838, asynonym of Tsuga canadensis (L. 1763) Carr. 1855. Abies menziesu Mirb., Paris Mus. Hist. Nat. Mem. 13: 63, 70. 1825 (as “Menziezi’’). Non Abies menziesu (Dougl.) Lindl., Penny Cyclop. 1: 32. 1833, based on Pinus menziesii Dougl. ex D. Don in Lamb., Descr. Genus Pinus, ed. 3 (8°), 2: unnumbered page between pp. 144 and 145. 1832, a synonym of Picea sitchensis (Bong. 1832) Carr. 1855. Abies douglasii Hort. ex Loud., Hort. Brit. 388. 1830; nomen nudum. Pinus douglasit Sabine ex D. Don in Lamb., Descr. Genus Pinus, ed. 3 (8°), 2: unnumbered page between pp. 144 and 145, pl. [47]. 1832. Abies mucronata Raf., Atl. Jour. 1: 120. 1832 (autumn). Abies douglasii (Sabine ex D. Don in Lamb.) Lindl., Penny Cycl. 1: 32, illus. 1833. Abies douglasu var. taxtfolia Loud., Arboretum et Frut. Britannicum 4: 2319, fig. 2231. 1838. Picea douglasi (Sabine ex D. Don in Lamb.) Link in Linnaea 15: 524. 1841. Tsuga douglasu (Lindl.) Carr., Traité Gen. Conif. 192. 1855. Pseudotsuga douglasu (Lindl.) Carr., Traité Gen. Conif., ed. 2, 256. 1867. Pseudotsuga taxifolia |(Lamb.) | Britton, N.Y. Acad. Sci. Trans. 8: 74. 1889. Tsuga taxtfolia |(Lamb.) |] Kuntze, Rev. Gen. Pl. 2: 802. 1891. Pseudotsuga mucronata ( Raf.) Sudw. in Holz., Contr. U.S. Nat. Herb. 52 206. 13895, Pseudotsuga taxifolia (Poir.) Britton ex Sudw., U.S. Dept. Agr., Div. Forestry Bull. 14: 46. 1897. Abietia douglasn (Lindl.) Kent, Veitch’s Man. Conif., ed. 2, 476. 1900. Pseudotsuga vancouverensis Flous, Soc. d’Hist. Nat. Toulouse Bull. 66: 340, pl. 1934; Lab. Forest. Toulouse Trav. tome 1, v. 2, art. 6: 12, pl. 1934. Department of Biology and Botany, University of British Columbia, Vancouver REFERENCES Boivin, B. 1954. Pseudotsuga menziesii (Mirbel) Franco versus Pseudotsuga taxi- folia (Poiret) Britton. Bol. Soc. Brot. 2-a série 28: 63, 64. Coves, E. 1893. The history of the expedition under . . . Lewis and Clark to the sources of the Missouri... during 1804-6. 3: 831. Exwes, H. J., and A. Henry. 1909. The trees of Great Britain and Ireland. 4: 824. Franco, J. po AMARAL. 1953. On the nomenclature of the Douglas fir. Sociedade Astoria, Lisbon, pp. 6. GLeason, H. A. 1955. Pedanticism runs amuck. Rhodora 57: 332-335. Littie, E. L., Jr. 1952. The genus Pseudotsuga (Douglas-fir) in North America. Leafl. West. Bot. 6: 181-198. . 1953. Check list of native and naturalized trees. U. S. Agric. Handb. No. 41. NutTtTatLi, T. 1865. The North American sylva. 2: 189. PursH, F. 1814. Flora Americae Septentrionalis. 2: 640. SARGENT, C. S. 1898. The silva of North America. 12: 91. 268 MADRONO [Vol. 13 NOTES AND NEWS AN EXTENSION OF RANGE FOR TRIFOLIUM OWYHEENSE GILKEY. In the January issue of the current volume of Madrono this new species of clover was described and cited from Malheur County, Oregon, the only locality known at that time. Since then, an unnamed specimen of clover in the Oregon State College herbarium, collected by Percy Train in 1935, has been found to be this species. The label reads as follows: STAG OLIN see oe eee ae 5/30/35. Alt. 4000’. Belt, Artemisia. Habitat, dry shale hillsides, Steins Mts., Harney County, Oregon, east base, near Folly- farm.” Since Mr. Train’s collections of that year were, I believe, widely distributed, it is probable that this species is present in undetermined material in other herbaria. The Harney County site is approximately 50 miles west of the type locality, with similar topography, sparse vegetation, and blue-gray diatomaceous soil. Further field work is necessary to determine whether the distributional range of the clover coin- cides with that of the combination of ecological factors thus far known to be asso- ciated with it—Hertren M. GILxey, Oregon State College. MeEmMoriAL ENDOWMENT Funp. The Executive Council of the California Botanical Society, at its meeting on October 18, 1956, voted to establish a Memorial Endowment Fund to honor the memory of deceased members and friends of the Society. The income from this Memorial Fund is to be used in helping to defray the expenses of the Society, the major portion of which is the publication of Madrono. In perpetually serving the cause of botanical science, the Memorial Fund will thus carry on the work to which these members and friends in whose memory the contributions are made devoted their lives. This Memorial Fund was first conceived in memory of LeRoy Abrams who passed away in August of this year. Since then, Roy Clausen and Thomas H. Kearney have also passed away. It is hoped that members and friends of the California Botanical Society will wish to commemorate in this way those who have done so much to further the science of Botany. Unless a donor stipulates to the con- trary, the Society will send a card to the family of the person in whose name a con- tribution to the Memorial Endowment Fund has been made. Dr. Ira L. Wiggins of Stanford University has been appointed Chairman of the Memorial Endowment Fund Committee. Some publications of interest follow: Drawings of British Plants, by Stella Ross-Craig. Part IX. Rosaceae (2). 39 pls. 1956. 8 s./6 d. net. G. Bell and Sons, Ltd. London. With Part IX we have the comple- tion of the third volume of Miss Ross-Craig’s monumental work which aims at illus- trating all the species of flowering plants growing naturally in the British Isles. Guide to the Flora of Malta, by Guido B. Lanfranco. 1-66, 300 figs. grouped in 33 pls. 1955. 6 s./6 d. Progress Press Co., Ltd. Malta. A handbook for the layman, of interest to us because of the occurrence in Malta of many species which have become commonly established here as weeds of Mediterranean origin. Check List of Vascular Plants of San Luis Obispo County, by Robert F. Hoover. 1-53. 1954. California State Polytechnic College, San Luis Obispo, California. A mimeographed species list with some annotations. Leguminosae of Nevada, Part I — Astragalus and Oyxtropis, by R. C. Barneby. Contributions toward a Flora of Nevada, No. 38. 1-86. 1 fig. 1956. Plant Industry Station, Beltsville, Maryland. Unbound, printed in offset. This is the most recent issue of this series of publications on the flora of Nevada prepared under the supervision of W. Andrew Archer, United States National Arboretum. Originally started in 1940, many parts of the series were published between 1940 and 1942; work was resumed in 1954, and several issues have appeared since then, among which are: Scrophu- lariaceae of Nevada, Part I—Mimulus, by Gabriel Edwin, 1955; Portulacaceae of Nevada, by Arthur H. Holmgren, 1955. 1956] INDEX 269 INDEX TO VOLUME XIII Classified entries: Chromosome numbers, Reviews. New scientific names are in bold- face type. Un-annotated taxa in floral lists are omitted from Index. Abies: alba, 266; balsamea, 266; doug- lasii, 267, var. taxifolia, 267; menziesii, 267; mucronata, 267; taxifolia, 266 Abietia douglasii, 267 Acanthochiton: 6; Wrightii, 44 Acnida: 6; alabamensis, 15; altissima, 18, var. prostrata, 18, var. subnuda, 18; australis, 15; cannabina, 11, var. aus- tralis, 15, var. concatenata, 18, var. cuspidata, 15, var. lanceolata, 11, var. salicifolia, 11; concatenata, 18; cus- pidata, 15; Ellioti, 11; floridana, 25; obtusifolia, 11; rhyssocarpa, 11; ruso- carpa, 11; salicifolia, 11; subnuda, 18; tamariscina, 26, var. concatenata, 18, var. prostrata, 18, var. subnuda, 18, var. tuberculata, 18; tuberculata, 18, var. prostrata, 18, var. subnuda, 18 Acrolasia gracilis, 103 Actinolepis: 226, 227; nubigena, 227; multicaulis, 227; Pringlei, 227 Adiantum: tracyi, A natural hybrid, 195 Adiantum: capillus-veneris, 196; jordanii, 195, 201, fig. 203; pedatum, 195, figs. 201, 203; X tracyi, 198, figs. 197, 201, 203 Aloina rigida var. pilifera, 101 Amaranthus: 5; Acanthochiton, 44, figs. 8, 9, map, 45; altissimus, 18; ambigens, 18; annectens, 42; arenicola, 39, figs. 8, 9, map, 41; australis, 15, figs. 8, 9, map, 17; Bigelovii, 37; cannabinus, 11, figs. 8,9, map, 14; cuspidatus, 15; flori- danus, 25, figs. 8, 9, map, 24; Greggii, 42, figs. 8,9, map, 43; var. Muelleri, 42; Haughtii, 5; macrocaulos, 11; mia- miensis, 18; myrianthus, 42; Palmeri, 31, figs. 8, 9, map, 33, var. glomeratus, 31, 36; spinosus, 6; tamariscinus, 26, figs. 8, 9, map, 29; Torreyi, 36, 39, var. suffruticosus, 36; tuberculatus, 18, figs. 8, 9, map, 23; Watsoni, 36, figs. 8, 9, map, 37; Wrightii, 44 Amaranths, Revision of the dioecious, 5 Amblogyne Torreyi, 36, 39 Antheropeas: 227; lanosum, 227, fig. 232; morphology and anatomy, 234, 237; Wallacei, 227, figs. 232, 233 Arceuthobium: americanum, 172; cam- pylopodum, 170, fig. 171; douglasii, 172 Arizona: Two fungi associated with, Co- leosporium crowellii Cummins, 189 Babcock, Ernest Brown, 81 Baerla, 226, 227 Bahia, 226 Baja California, Mexico: A new species of Bouvardia from, 140; Muhlenbergia Brandegei, a new species from, 244; Prosopis globosa Gill. in, 172 Bealia: 245 ; mexicana, 244, 250; speciosa, 246 Beattie, R. K.: Joseph Burke in 1853, 259 Bell, C. R.: Review, How to know the grasses, 48 Birdsey, M. R.: The status of Pseudo- homalomena pastoensis, 47 Blake, S. F.: Terms used to designate type material, 207 Bonar, L.: Review, The Polyporaceae of the United States, Alaska and Canada, 80 Bouvardia: Alexanderae, 142; figs. 141, 143; from Baja California, Mexico, 140 Brandegea, 84, 121 Burke, Joseph, in 1853, 259 California Botanical Society: Report of the treasurer for 1955, 208 California: Monochoria vaginalis in, 112; New species of Elatine in, 239; Mosses of Pterigoneurum ovatum and P. sub- sessile, 100 Calla aethiopica, 47 Carlquist, S., On the generic limits of Eriophyllum (Compositae) and related genera, 226 Carter, A.: A new species of Bouvardia (Rubiaceae) from Baja California, Mexico, 140 Chromosome counts in the section Simio- lus of the genus Mimulus, 107 Chromosome numbers: Acnistus arbores- cens, 205; Antheropeas lanosum, 231, Wallacei, 231; Argythamnia Brande- gel var. intonsa, 205; Baileya multira- diata, 206; Brickellia peninsularis var. amphithalassa, 206; Capsicum macro- phyllum, 205, maculatum, 205, steno- phyllum, 205, stramoniifolium, 205, tetramerum, 206; Clermontia oblongi- folia, 206; Cyanea angustifolia, 206; Eremalche exilis, 241 ; Eriophyllum am- biguum, 231, confertiflorum, 230, var. tanacetiflorum, 230, Congdonii, 231, Jepsonii, 230, lanatum var. achillae- oldes, 230, var. arachnoideum, 230, var. grandiflorum, 230, latilobum, 230, mul- ticaule, 231, Nevinii, 230, Pringlei, 231, staechadifolium, 230; Furcraea Roezlii, 205; Laurentia carnosula, 206; Lycian- thes multiflora, 206; Malvastrum, 241; Marah fabaceus, 120, macrocarpus, 120, oreganus, 120; Mentzelia affinis, 104, albicaulis, 105, arborescens, 105, con- gesta, 104, gracilenta, 104, gracilis, 105, involucrata, 105, laevicaulis, 105, lind- leyi subsp. crocea, 104, subsp. lindleyi, 104, micrantha, 104, multiflora, 105, 270 MADRONO nitens, 104, pectinata, 104, veatchiana, 104; Mimulus guttatus, 108, var. graci- lis, 108, var. grandis, 108, var. typicus, 108, glabratus var. utahensis, 108, lu- teus, 108, nasutus, 108, trigrinoides, 108, tigrinus, 108, tilingii, 108; Monolopia gracilens, 231, lanceolata, 231, major, 231, stricta, 231; Nolina Beldingii, 205 ; Perityle socorrosensis, 206; Pseudoba- hia bahiaefolia, 231, Heermannili, 231, Peirsonii, 231; Phacelia capitata, 215, corymbosa, 215, frigida, 215, hetero- phylla, 215, leptosepala, 215, leucophyl- lay 215, mutabilis, 215; nemoralis, 215; Purshia tridentata, 205; Solanum Cer- vantesii, 206, ciliatum, 206, jamaicense, 206, parcebarbatum, 206, salviifolium, 206, torvum, 206, umbellatum, 206; Sphaeralcea, 241; Syntrichopappus Fremontii, 231, Lemmonii, 231; Tetra- molopium humile, 206 Chromosome numbers of plants, Docu- mented: 205 Cladosporium aecidiicola, 189, 192, figs. 192, 193 Cleome: hemsleyana, 183; lutea, 187; macrorhiza, 186; mexicana, 181; steno- phylla, 186; tenuis, 186; wrightii, 186 Cleomella: 178; angustifolia, 187; hem- sleyana, 183; longipes, 178, figs. 185, 186, map, 187; medicagineae, 182 ; mex- icana, 181, figs. 185, 186; obtusifolia, 178; perennis, 180, figs. 185, 186; the Mexican species of, 177 Clomena: 245 Coleosporium: crowellii, 189, figs. 191, 192, 193 Colorado: Flora of the Crested Butte Quadrangle, 64, check list, 71 Copeland, H. F.: Observations on Prasi- ola mexicana, a freshwater alga of un- known relationships, 138 Crepis, 81 Cronartium quercuum, 194 Darluca filum, 190, figs. 191, 192, 193 Davidson, J. F. and T. L. Thompson, Phytoserology versus genealogy in Zea mays, 252 Dimeresia, 238 Douglas-fir, Pseudotsuga menziesii, A summary of the nomenclature of, 265 Dressler, R. L.: Prosopis globosa Gill. in Baja California, 172 Drymaria debilis, 142 Dudleya nubigena, 142, Echinocystis: 84, 121; and Echinopepon in United States and northern Mexico, 84; coulteri, 90; echinata, 85; fabaceus, 128, var. inermis, 130; gilensis, 137; guadalupensis, 124; horrida, 131; iner- mis, 130; lanatus, 88; lobata, 85, 207, fig. 89, map, 98; macrocarpa, 132, var. [Vol. 13 leptocarpa, 132; marah, 122; minima, 96; muricatus, 126; oregona, 122; sca- brida, 130, 207; watsonii, 126; wrightii, 94 Echinopepon: 84, 87, 113; cirrhopedun- culatus, 97, figs. 93, 99, map, 98; con- fusus, 90, 207; coulteri, 90, figs. 91, 99, map, 98; horridus, 88, figs. 89, 99, map, 98; in United States and northern Mex- ico, 84; jaliscanus, 88; lanatus, 88; minimus, 95, 207, map, 98, var. mini- mus, 96, figs. 93, 96, var. peninsularis, 97, fig. 96; nelsoni, 92, 207; parvifolius, 92; peninsularis, 97; wrightii, 94, figs. 91, 99, map, 98 Edwin, G.: Notes on Nevada Mimulus, 110 Elaterium: 87; coulteri, 90; minimum, 96; wrightii, 94 Elatine: ambigua, 239; americana, 239; brachysperma, 240; gracilis, 240; het- erandra, 240; in California, New spe- cies of, 239; obovata, 240; triandra, 239, var. obovata, 240 Elytroderma deformans, 194 Endowment Fund, 268 Epicampes, 246; mexicana, 250 Eremalche: 241; exilis, 242; kernensis, 243; key to the taxa, 242; Notes on, 241: Parryi; 243 3 rotundifolia, 242 Eriophyllum: ambiguum, 227, figs. 228, 236: On the generic limits of, 226; ba- hiaefolium, 227; confertiflorum, 229, fig. 228, var. tanacetiflorum, 229, fig. 228; Congdonii, 227, fig. 236; Heer- mannii, 227; Jepsonii, 229, fig. 236; lanosum, 227; lanatum var. grandi- florum, fig. 235, var. arachnoideum, fig. 228° latilobum, 2292 minus, -227.mo-= havense, 227, fig. 235; multicaule, 227, figs. 228, 233, 236; Nevini, 229: nubi- genuim, 227% .Prmelet, 22 /,) ties 232% sect. Eriophyllum, 229, cytology, 229, morphology and anatomy, 234, 237: Sect. Actinolepis, 229, morphology and anatomy, 234, 237; staechadifolium, 226, fig. 233; Wallacei, 227 Funanus cusickii, 110 Euphorbia antisyphilitica, discovered in southern New Mexico, 176 Eusphaeralcea, 241 Gentry, H. S.: A new Gossypium from Michoacan, Mexico, 261 Gilkey, H. M.: A new Trifolium from Oregon, 167; An extension of range for Trifolium owyheense Gilkey, 268 Gonium: behavior in culture, 60; discoi- deum, 61; multicoccum, 56, 61, figs. 55, 59, octonarium, 49, 57, figs. 50, 53; pectorale, 54, var. pectorale, 54, 58, fig. 55, var. praecox, 55, 57, fig. 55; sociale, 58; The genus in North America, 49 1956] Gossypium: from Mexico, A new, 261; aridum, 262, fig. 263; armourianum, 264; barbadense, 264; davidsonii, 264; harknessii, 264; hirsutum, 264; loba- tum, 261, fig. 263; sects. Erioxyla, 261, Sturt’ana, 264; thurberi, 264 Greene, E. L., and Howell’s ‘“‘Flora of Northwest America,” 1 Heiser, C. B., Jr.: Biosystematics of He- lianthus debilis, 145 Helianthus: agrestis, 150; annuus, 150, subsn. texanus, 160; argophyllus, 150, 152; Bolanderi, 147; chromosome num- bers and pairing, 147; cucumerifolius, 160; debilis, 151, fig. 159, subsp. cu- cumerifolius, 160, fig. 163, map, 153, subsp. debilis, 151, figs. 155, 159, map, 152, subsp. hirtus, 162, fig. 163, map, 153, subsp. praecox, 160, fig. 163, map, 153, subsp. runyonii, 161, fig. 163, map, 153, subsp. silvestris, 158, fig. 159, map, 153, subsp. tardiflorus, 156, figs. 157, 159, map, 152, subsp. vestitus, 154, 158. hes, 157, 159; map, 152, var: cu- cumerifolius, 160, var. praecox, 160; lindheimerianus, 160; petiolaris, 150, 162, map, 153; praecox, 160; procum- bens, 151; vestitus, 150, 154 Hexameria: 84, 85 Howell’s “Flora of Northwest America,” Edward L. Greene and, 1 Hyponema, 178 ltis, H. H.: Studies in the Capparid- aceae Il. The Mexican species of Cleo- mella: Taxonomy and evolution, 177 Isopara: 178; mexicana, 181 Kearney, T. H.: Notes on Malvaceae, VIII. Eremalche, 241 Keener, P. D.: Two fungi associated with a microcyclic rust, Coleosporium crow- ellii Cummins on needles of Pinus edu- lis Engelm. in Arizona, 189 Kennedy, C. A. and P. JT. Leyendecker: Euphorbia antisyphilitica discovered in southern New Mexico, 176 Kech, L. F.: Mosses of California V. Pterigoneurum ovatum and P. subses- sile, 100 Kratina, V. J., A summary of the nomen- clature of Douglas-fir, Pseudotsuga menziesii, 265 Kruckeberg, A. R., Notes on the Phacelia magellanica complex in the Pacific Northwest, 209 Kuijt, J., A new record of dwarf mistle- toe on lodgepole and western white pine, 170 Lange, FE. F.: Edward L. Greene and Howell’s “Flora of Northwest Amer- icay~ 1 INDEX 2 Langenheim, J. H.: Flora of the Crested Butte Quadrangle, Colorado, 64 Lewis, H. and H. J. Thompson, Chro- mosome numbers in Mentzelia (Loa- saceae), 102 Leyendecker, P. J. and C. A. Kennedy: Euphorbia antisyphilitica discovered in southern New Mexico, 176 Malvastrum: 241; exile, 241; Parryi, 241; rotundifolium, 241 Marah: 84, 121; fabaceus, 128, map, 129, var. agrestis, 130, 207, figs. 114, 115, 127, var. fabaceus, 128; gilensis, 136, fig. 136, map, 129; guadalupensis, 124, ne. 125; horndus, 87, 1315 hes: “114. 133, map, 129; inermis, 130; lepto- carpa, 132; macrocarpus, 131, 207, map, 129, var. macrocarpus, 132, figs. 114, 133, var. major, 134, fig. 135, var. mi- cranthus, 134, fig. 135; major, 134, 207; micranthus, 134; minima, 95; muricatus, 121, 122, 126; oreganus, 121, 122, figs. 114, 125, map, 129; Some tax- onomic and ecological considerations, 113, watson, 126, 207, fies. 114, 127, map, 129 Mason, H. L., New species of Elatine in California, 239 McCaskill, B. J. and J. M. Tucker: Mon- ochoria vaginalis in California, 112 Megarrhiza: 121, californica, 128, 132; gilensis, 136; guadalupensis, 113, 124; muricata, 126; oregona, 122 Mentzelia: affinis, 102, fig. 106; albicaulis, 103; arborescens, 107; Chromosome numbers, 102; congesta, 103; dispersa, 102; gracilenta, 103; gracilis, 103; in- volucrata, 106; laevicaulis, 105, fig. 106; lindleyi, 103, subsp. crocea, 103, subsp. lindleyi, 103; micrantha, 102; multiflora, 106, fig. 106; nitens, 103; pectinata, 103; veatchiana, 103 Mexico: A new Gossypium from, 261; genera Echinocystis and Echinopepon in northern, 84; Studies on Capparida- ceae of, 177 Micrampelis: 85, 113; coulteri, 90; echi- nata, 85; fabacea, 128, var. agrestis, 130; guadalupensis, 124; lanata, 88; leptecarpa, 132; lobata, 85; marah, 122: minima, 96: oregona, 122; wat- sonli, 126: wrightii, 94 Minulus: bigelovii, 110, var. cuspidatus, 110, var. ovatus, 110; Chromosome counts in section Simiolus, 107; cu- sickii, 110; eastwoodiae, 112; glabratus var. utahensis, fig. 109; guttatus, fig. 109; Notes on Nevada, 110; tilingii, 110, fig. 109 Mistletoe: on lodgepole and_ western white pine, A new record of dwarf, 170 Momordica echinata, 85 272 MADRONO Monochoria: vaginalis, 112, in California, 112 Monolopia: 227; bahiaefolium, 227; gra- cilens, 227; Heermannii, 227; major, 227; minor, 228; morphology and anat- omy, 234, 237 Montelia: tamariscina, 26, var. concate- nata, 18 Mosses of California. Pterigoneurum ovatum and P. subsessile, 100 Muhlenbergia: biloba, 250, fig. 249; Brandegei, 248, fig. 248, new species from Baja California, Mexico, and re- lationship to M. biloba, 244; peruviana, 245; speciosa, 246 Muller, C. H.: The distribution of Quer- cus boyntoni, 221 Nevada: Mimulus, Notes on, 110 New Mexico: Euphorbia antisyphilitica discovered in southern, 176 Notes and News: 80, 112, 176, 207, 268 Oregon: A new Trifolium from, 167 Phacelia: alpina, 216; capitata, 209, fig. 210; corymbosa, 212; heterophylla, 211, 213; leptosepala, 213, 217, 218; leucophylla, 212, 217, var. alpina, 216; magellanica complex in Pacific North- west, 209; nemoralis, 213 Physostemon: 183; hemsleyanum, 183; medicagineum, 182 Phytoserology versus genealogy in Zea mays, 252 Picea: douglasii, 267; sitchensis, 267 Pinus: contorta, 170; douglasii, 267; edu- lis, 189, fig. 192; menziesii, 267; monti- cola, 170; taxifolia, 266 Pocock, M. A.: Studies in North Ameri- can Volvocales, I, The genus Gonium, 49 Prasiola: mexicana, 138, fig. 139, fresh- water alga of unknown relationships, 138 Prosopis: globosa Gill. in Baja Califor- nia, 172; globosa, 172, var. mexicana, 174) fe 735 stiiata, el 72 Pseudobahia: 227; bahiaefolia, 232, fig. 228; Heermannii, 232, figs. 228, 232; morphology and anatomy, 234, 237; Peirsonii, 232 Pseudohomalomena pastoensis, status, 47 Pseudotsuga: douglasii, 267; menziesii, 266, summary of nomenclature, 265; mucronata, 267; taxifolia, 267; van- couverensis, 267 Pterigoneurum: ovatum, 100; subsessile, 100 Puccinia conspicua, 194 Quercus: boyntoni, 222, fig. 223; drum- mondii, 225; stellata, 221 [Vol. 13 Reeder, C. G., Muhlenbergia Brandegei, a new species from Baja California, Mex- ico, and its relationship to Muhlenber- gia biloba, 244 Reviews: Harrington, Manual of plants of Colorado, 78; Muenscher, Weeds, 175; Overholts, Polyporaceae of Unit- ed States, Alaska and Canada, 80; Pohl, How to know the grasses, 48 Robbins, G. T.: Review, Manual of the plants of Colorado, 78 Sauer, J.: Revision of the dioecious Am- aranths, 5 Sharsmith, H. K.: Review, Weeds, 175 Sicyoides: 84 Sicyos: 84; angulatus, 113; lobata, 85; Oreganus ells 3121. 122 Sphaeralcea, 241 Spilanthes repens, 151 Stebbins, G. L.: Ernest Brown Babcock, 81 Stocking, K. M.: Some considerations of the genera Echinocystis and Echinope- pon in the United States and northern Mexico, 84; Some taxonomic and eco- logical considerations of the genus Marah (Cucurbitaceae), 113 Syntrichopappus: 228; Fremontii, 228, figs. 232, 233; Lemonnii, 228, figs. 232, 235; morphology and anatomy, 234, 23d Thompson, H. J. and H. Lewis: Chromo- some numbers in Mentzelia (Loasa- ceae), 102 Thompson, T. L. and J. F. Davidson, Phytoserology versus genealogy in Zea mays, 252 Tortula brevipes, 101 Trifolium: from Oregon, A new, 167; owyheense, 169, fig. 168; owyheense Gilkey, extension of range, 268 Tsuga: canadensis, 266; douglasii, 267; taxifolia, 267 Tucker, J. M. and B. J. McCaskill: Mon- ochoria vaginalis in California, 112 Types, Terms used to designate, 207 Vaseyanthus, 84, 121 Vickery, R. K., Jr.: Chromosome counts in the section Simiolus of the genus Mimulus (Scrophulariaceae), 107 Volvocales, Gonium, in North America, 49 Wagener, W. H., Jr.: A natural hybrid, x Adiantum tracyi C. C. Hall, 195 Wislizenia: 185 Zantedeschia aethiopica, 47 Zea mays, Phytoserology, 252 INFORMATION FOR CONTRIBUTORS Manuscripts submitted for publication should not exceed an estimated 20 pages when printed unless the author agree to bear the cost of the ad- ditional pages at the rate of $15 per page. Illustrative materials (includ- ing “typographically difficult” matter) in excess of 30 per cent for papers up to 10 pages and 20 per cent for longer papers are chargeable to the author. Subject to the approval of the Editorial Board, manuscripts may be published ahead of schedule, as additional pages to an issue, provided the author assume the complete cost of publication. Shorter items, such as range extensions and other biological notes, will be published in condensed form with a suitable title under the general heading, ‘‘Notes and News.” Institutional abbreviations in specimen citations should follow Lanjouw and Stafleu’s list (Index Herbariorum. Part 1. The Herbaria of the World. Utrecht. Second Edition, 1954). Articles may be submitted to any member of the Editorial Board. MADRONO A WEST AMERICAN JOURNAL OF BOTANY A quarterly journal devoted to the publication of botanical re- search, observation, and history. Back volumes may be obtained from the Secretary at the following rates: VolumeI,1916-1929 . . . . . $5.00 Volume IJ, 1930-1934. . . .) . 5.00 Volume ITI, 1935-1936 . . . . 5.00 Volume IV, 1937-1938 . . . . 5.00 Volume V, 1939-1940. . . . . 5.00 Volume VI, 1941-1942 . . . . 5.00 Volume VII, 1943-1944 . . . . 5.00 Volume VIII, 1945-1946 . . .. . 5.00 Volume IX, 1947-1948 . Sener 6.00 Volume X, 1949-1950. . . . . 7.00 Volume XJ,1951-1952 . . ..) 8.00 Volume XIT, 1953-1954 . . . Sg 8.00 Volume XIII, 1955-1956 . 2. tt 8.00 Single numbers, Volumes: and:2.0 fj.) Seuss 90 Succeeding numbers, (except Volume XI, No.2). .. 1.00 Volume Xl No::2°-. 2a) 4 1.50 The subscription price of MapRONo is $4.00 per year. If your institution does not now subscribe to MApRoNo, we would be grateful if you would make the necessary request. Since there is no extra charge for institutional subscriptions, an individual may hold membership in the California Botanical Society on the basis of his institution’s subscription. Address all orders to: G. THomas Rossins, Corresponding Secretary Department of Botany University of California, Berkeley 4, California Pg i or i | _- . ) Ros ’ . a! ; ' * ‘ 1 i , . _ fy ee 7 | : SMITHSONIAN INSTITUTION LIBRARIES AMM ian SS ———————_ —————— —— ——————— ¢