Wet wpe ‘sonyad ' 1 a 4 isi ‘ Ar isa rf 7 / » qf eee i ' \pha af stand , efi ote tay ps : sy isdety Tebpbateh Getant “mee vee - ate enlta geen: sifellay sae Aci f Tewepheae V ete bewaee! : is piabey egy tes pnt et yee ts eee gop aatets PAG ts hao: Haid ae Ch pate Ld Rg At Ta sats pa et ay! 3 ail Leip tegaeiae CPU eam Bates! he hae a go Celaya betaine J ; eaonetads ey aa i day, sor pglabepprater A pg Bate ete pea ya foratgnctbe Heth : miehitereie & ; aeatetera tare’ : a's whale he ow ‘ ‘ VAG be (ete row eters Leyte ca aka deiaial, gates a : oe oty at tere the were vate hae g ttehe a bpei eae ey tae ae Lied te te eee a State (bs ale genre to Hy : pha y Ce tae 1 rr us : 1 [ily aha phase ab ee , ’ t W . 73 7 i 4 ds s/s 1 : , ‘ t F ae ape { 4 : de F 22 075 74 . aoe “ ‘ 3 we ' A ae ’ ‘ ' bia 4 i : ‘ . 4 thet a | ia # ’ i ; } ij : fie? bag t ‘ : ; a6 , fo f ‘ »t a y ‘ta G44 , oo: peug (bos el { sity fs 7 re : i : 7 j ; j , ; ' rot tl, 4 * 5 \ - : j a ae i 1% itp {4 : i ae | ' Soe . ; ats rp , : . . 4 ag , Fi ‘ 7 [eeeda yes aren head be pe, ) bei k ae ie Pasi rigs 5 1 ; aye {eres , § aS cM eos a ee f eee (Photograph by Iay-Hart, N. Y.) AVEN NELSON Frontispiece to Volume 7, Maprono. MADRONO A WEST AMERICAN JOURNAL OF BOTANY VOLUME VII 1943 - - 1944 y on Ma 38 & Gy : National Muse Mf & ae Published by the California Botanical Society, Inc., 4004 Life Sciences Building, University of California, Berkeley North Queen Street and McGovern Avenue, Lancaster, Pennsylvania Board of Editors Hersert L. Mason, University of California, Berkeley, Chairman. LeRoy Aprams, Stanford University, California. EpGar ANDERSON, Missouri Botanical Garden, St. Louis. Lyman Benson, Pomona College, Claremont, California. Hesrert F. Coperann, Sacramento Junior College, Sacramento, California. Ivan M. Jounston, Arnold Arboretum, Jamaica Plain, Massachusetts. Mivprep E. Marutas, 2851 North Lake Avenue, Altadena, California. Bassett Macuire, New York Botanical Garden, New York City. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board ANNETTA M. Carter, Department of Botany, University of California, Berkeley. Business Manager Wititiam Hiesey, Carnegie Institution of Washington, Stanford University, California. To Aven Nelson, founder of the Rocky Mountain Herbarium, President, Emeritus, of the University of Wyoming, now in his eighty-fifth year—we dedicate this seventh volume of Madrojfio. His is a life, rich in experience, rich in accomplishment! He pioneered both in the field of botany and horticulture in the Rocky Mountain and Great Basin regions. He effected the training and greatly influenced the lives of many students who have taken their places among the able botanists of our time. Apart from his work as an educator his influence on the agri- culture of the northern Rocky Mountains has been notable. He wrote many horticultural bulletins issued by the state of Wyo- ming and was among the first to urge the growing of apples as an agricultural crop in that area. His great ability and kindly personality are reflected in his being appointed to serve as president of the University of Wyo- ming during the years 1917 to 1922, and, in 1935, being elected president of the Botanical Society of America. The genus Anel- sonia Macbride and Payson and such species as Salia Nelsoni Ball, Carex Nelson Mackenzie, Stipa Nelsoni Scribner, Delphinium Nel- soni Greene and Cirsium Nelsonii (Pammel) Rydb. are among the many plants named in his honor. We wish him good health and many more years of usefulness in his chosen field of study. iii Ryd yn CONTENTS Frontispiece: Aven Nelson Thalictrum ametrum Greene: an Interesting Nomenclatural Case. Leon Croizat Una Nueva Especie de Pinus Mexicano ...... sa ere esas Oe Maximino Martinez An Investigation of the Presence of Siliceous Rods in the Secondary Wall Oly WiOOGY, LISSUG: 42a 5 oe ciak Mae, dow dee eh lS See ies Se St eae Walter M. Schall A New Cliff-Rose from Arizona:.......:............. Thomas H. Kearney Hr anGcescoylrancesChs @ oe. uae uh. aw 2 oes bd ena Maks Se es John M. Tucker | ERAS ae IS nt a mE eC 27, 63, 93, 126, 226, chek Watherine (Crum 6.25% esi. asoans pe Guehhes a daa Ss Herbert L. Mason The Xerophyllous Species of Philadelphus in Southwestern North America. C. Leo Hitchcock A New Species of Phacelia from Saline Valley, California. Lincoln Constance Gilia multiflora Nutt. and Its Nearest Relatives. Thomas H. Kearney and Robert H. Peebles A New Species of Tauschia from the State of Washington. Mildred E. Mathias and Lincoln Constance Notes on the Flora of the Charleston Mountains, Clark County, Nevada. Wem aChaCedO 9 game eae oe ne Sp Reals A es Ira W. Clokey Noteworthy Plants from Idaho.. II .................... Arthur Cronquist The Legitimacy of Thalictrum polycarpum Wats. ....... Arthur Cronquist Valid and Legitimate Names—and Thalictrum polycarpum S. Wats. C. A. Weatherby The Anatomy of Redwood Bark .....................2 04. Iroing H. Isenberg Walham Albert Setchell’ 52. .2445.4 5) 0.008 de yes ae ands Herbert L. Mason Vigethia, A New Genus of Compositae Based on Wyethia mexicana Watson. William A. Weber A Revision of the Genus Fremontia .................... Margaret Harvey The Genus Styrax in Central and Western Texas .............. V.L. Cory The Story of Parthenium alpinum ................... George J. Goodman Two Types of Broad-leaf Erodium in California. Kenneth A. Wagnon and Harold H. Biswell Cytological Evidence for the Taxonomic Position of Schizachne purpurascens. W.S. Boyle ‘Noteworthy Plants of Texas. II. A New Species of Peltandra. Fred A. Barkley A Monograph of the North American Species of Fritillaria. Dorothy E. Beetle he Homonym Question . 2.064056. 004.508 dees wes RE i os Leon Croizat INOTES mala NIG WS: esate es nna ental: et cee Me Ree hee cna, ann, Racine & 160, 192, The Taxonomy and Cytology of the Subgenus Gormania of Sedum. Robert T. Clausen and Charles H. Uhl Jawa Stacey, Caricolocist) -.. so. 264..6hea cu daw Bi. as John Thomas Howell Dermatitis and Photosensitization Produced by Ptelea angustifolia. W.C. Muenscher and Babette I. Brown On the Shoot Apex of Chlorogalum pomeridianum (DC.) Kunth. Clarence Sterling Structure and Taxonomy of Taenioma, Including a Discussion of the Phy- logeny of the Ceramiales ..............5......-.-+ George F. Papenfuss Notes ‘on the-Algal Genus Paenioma 52 242 6 sec. 2 oe a es C.K. Tseng Some New Laurenciae from Southern California ..... Elmer Yale Dawson Nomina Conservanda Proposals for Ten Genera of Trees and Shrubs. Elbert L. Little, Jr. Forestiera in Southern and Southwestern Texas ............... V. L. Cory Me XeCOmNOlUUMIC DV silat ys eit aac stag eh Davamen tet eee LOT De Bene 1 fhe ole as co ah a ae Hen iB a ty VOLUME VII NUMBER 1 MADRONO A WEST AMERICAN JOURNAL OF BOTANY s ~s, fic Co. Hi ¢ iD e- | ! il i & be Uj Ij \ = / Pe fis eS - | SA as aw we b SS VNAL MU. Pa a — } i Contents THALICTRUM AMETRUM GREENE: AN INTERESTING NOMENCLATURAL CASE, EEOC OUU LS ae seu ey ea eer cet emer Nard CN, ae Rt ey Sea ORE 1 Una Noeva EspPecie pe Pinus Mexicano, Maximino Martinez ........... 4, Aw INVESTIGATION OF THE PRESENCE OF SILICEOUS Rops IN THE SECONDARY Watt or Woopy Tissuz, Walter M. Schall ..............0 0.0.00 ..0.... 8 A New Cuirr-Rose rrom Arizona, Thomas H. Kearney Francesco FrRANcESCHI, John M. Tucker ........000 00000 ccc ee es 18 Reviews: Maunsell Van Rensselaer and Howard EK. McMinn, Ceanothus. Part I. Ceanothus for Gardens, Parks and Roadsides. Part II. A Systematic Study of Ceanothus (J. Francis Macbride); Adriance S. Foster, Practical Plant Anatomy (Ira L. Wiggins); Andrew Denny Rodgers, John Torrey, A Story of North American Botany (Mildred Mathias) ; S. F. Blake and Alice C. Atwood, Geographical Guide to the Floras of the World (Herbert L. Mason); Harold N. Moldenke, EHrio- caulaceae, Avicenniaceae, Verbenaceae in Flora of Texas (Ira L. Wiggins) Published at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania January, 1943 | MADRONO A WEST AMERICAN JOURNAL OF BOTANY Board of Editors Hersert L. Mason, University of California, Berkeley, Chairman. LeRoy Asrams, Stanford University, California. Epcar ANpDERSON, Missouri Botanical Garden, St. Louis. Lyman Benson, University of Arizona, Tucson. Hersert F’. Copetanp, Sacramento Junior College, Sacramento, California. Ivan M. Jonnston, Arnold Arboretum, Jamaica Plain, Massachusetts. Miuzprep E. Maruias, University of California, Berkeley. Bassett Macurre, Utah State Agricultural College, Logan. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board—EtTHEt Crum Department of Botany, University of California, Berkeley Business Manager—Wiut.iam HIeEsEy North Queen Street and McGovern Avenue, Lancaster, Pennsylvania or Carnegie Institution of Washington Stanford University, California Entered as second-class matter October 1, 1935, at the post office at Lancaster, Pa., under the act of March 3, 1879. Established 1916. Published quarterly. Subscription Price $2.50 per year. Completed volumes I to V inclusive, $25.00; each volume $5.00; single numbers $0.75. Papers up to 15 or 20 pages are acceptable. Longer contributions may be accepted if the excess costs of printing and illustration are borne by the contributor. Range extensions and similar notes will be published in con- densed form with a suitable title under the general heading “Notes and News.” Articles may be submitted to any member of the editorial board. Manuscripts may be included in the forthcoming issue provided that the contributor pay the cost of the pages added to the issue to accommodate his article. Reprints of any article are furnished at a cost of 4 pages, 50 copies $4.10; 100 copies $4.50; additional 100’s $0.85; 8 pages, 50 copies $5.95; 100 copies $6.60; additional 100’s $1.30; 16 pages, 50 copies $8.35; 100 copies $9.35; additional 100’s $2.00. Covers, 50 for $2.75; additional covers at $1.65 per hundred. Reprints should be ordered when proofs are returned. Published at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania, for the CALIFORNIA BOTANICAL SOCIETY, INC. President: A. R. Davis, University of California, Berkeley. First Vice- President: Palmer Stockwell, Institute of Forest Genetics, Placerville, Cali- fornia. Second Vice-President: Reed C. Rollins, Stanford University, California. Treasurer: William Hiesey, Carnegie Institution of Washington, Stanford Uni- versity, California. Secretary: Clarence R. Quick, United States Department of Agriculture, 26 Giannini Hall, University of California, Berkeley. Annual membership dues of the California Botanical Society are $2.50, $2.00 of which is for a year’s subscription to Madrofio. Dues should be remitted to the Treasurer. General correspondence and applications for membership should be addressed to the Secretary. THALICTRUM AMETRUM GREENE: AN INTERESTING NOMENCLATURAL CASE Leon CroizatT The validity of Thalictrum polycarpum S. Wats. against the earlier 7’. polycarpum Loret, and the later T'. ametrum Greene has been affirmed by Wheeler (Rhodora 40: 318-320. 19388) in a dis- cussion which is exceedingly interesting under the standpoint of nomenclature. In this discussion Wheeler raises two issues, first, whether the parenthetic author must be indicated in every case; second, whether an earlier name can be duplicated, and if so under what conditions. Lack of space makes it necessary for me to discuss here only one of these issues. Accordingly, I shall discuss the second, which is more important. The first printed mention of Thalictrum polycarpum occurs in a paper by Loret (Bull. Soc. Bot. France 6: 16. 1859). To spare the reader the necessity of wading through Loret’s stiff French, but doing violence to accepted bibliographical standards, I shall quote here as if in the original my own translation of this publica- tion. Loret states: “I have collected in a hedge at Barcellonette (Basses-Alpes) at the end of July 1851 an interesting Thalictrum which is quite noteworthy on account of its short-ovoid carpels, 9-12 to 14 being borne upon a single receptacle. This plant is close to, but differs appreciably from T. Jacquinianum Koch and T. expansum Jord. I believe this Thalictrum to be a new species, but, fearing to augment the confusion already prevalent in this genus, I merely bring this plant to the attention of the botanists who may have the opportunity of collecting at Barcellonette, hop- ing on my part to see this plant again on the spot. If I were to be allowed to give this plant a name, I would gladly call it T. poly- carpum or, better still, T. multiflorum (S’il m’etait réservé de lui imposer un nom, je lui donnerais volontier celui de Th. polycarpum ou mieux multiflorum).” The binomials of Loret have been disregarded by practically every author, with the exception of Lecoyer. In his monograph of Thalictrum (Bull. Soc. Bot. Belg. 24: 78-324. 1885), Lecoyer treats T'. polycarpum as a synonym of T'. multiflorum (op. cit., 304), which he places in the synonymy of T. minus L. Lecoyer adds (op. cit., 297) that T. multiflorum is “une forme non décrite” of T. minus, believing T'.. multiflorum to be a nomen nudum or a nomen semi- nudum which has T. polycarpum as its synonym. Wheeler is of the opinion, on the contrary, that T’.. polycarpum and T. multiflorum “is an illegitimate name of the type known as a nomen provisorium,”’ without specifying which one of these two binomials he interprets as the provisional name. I do not believe it necessary to argue the status of these binomials at this point, because the matter has little immediate importance. The impor- ManproNno, Vol. 7, pp. 1-32. January 28, 1943. 2° MADRONO [Vol.7 tant side of Wheeler’s interpretation is in his statement that: “Provisional names are not only illegitimate but are not validly published. Since they are not validly published they cannot as earlier homonyms, invalidate a later name”’ (italics mine). This statement contains an unmitigated fallacy. Thalictrum polycarpum is illegitimate because its duplicates T. minus L. (Art. 16, Art. 60[1] Amsterdam Code), and is invalid, in addition, at least because it is published as a synonym of T. multiflorum (Art. 40). Article 61 in the current Rules states: “Even if the earlier homonym is illegitimate, or is generally treated as a synonym on taxonomic grounds, the later homonym must be rejected.” This is clear enough: T. polycarpum Loret is not only illegitimate and invalid but has been treated as a synonym of T. minus L. on taxo- nomic grounds by Lecoyer. Obviously, T. polycarpum Loret, 1859, as an earlier homonym renders illegitimate T. polycarpum S. Wats., 1879. The text of Article 61 and the status of the bino- mials both of Loret and Watson precisely and absolutely contra- dict Wheeler’s affirmations. Since Wheeler’s error rests upon assumptions which are unfor- tunately widespread, it is advisable to add here a few words of comment. Many are the taxonomists who believe that a name which is not “valid”? may be “ignored.” This belief involves a fundamental confusion between two different concepts, which can easily be illustrated by an example. Let us suppose that John Doe publishes in 1940 Planta una without a Latin description. This binomial is invalid, because Article 88 of the Amsterdam Code requires a Latin diagnosis for a valid publication. Accord- ingly, Jack Roe can freely use the type specimen of P. una and propose on it in 1942 a new binomial, P. quaevis. Roe can do this because the publication of Doe does not “‘exist’’ as valid nomen- clature on account of the lack of a Latin diagnosis. Of course, Roe, if he so wishes, can honor the earlier invalid binomial pro- posed by Doe, effectively publishing P. una with a Latin diagnosis. In this case (Art. 48), the species will be known as P. una J. Doe in (or ex) J. Roe. An entirely different state of affairs obtains if J. Roe attempts to publish in 1942 a new species, naming it P. una and basing it upon a type specimen other than the one originally used by J. Doe in 1940 for his P. una. Such a duplication is expressly forbidden by Article 61, as it has been seen. Under the Vienna Code (1905) and Bruxelles Code (1910) it was not permitted to reject a well known name, “Because of the existence of an earlier homonym which is universally regarded as non-valid or for any other motive either contestable or of little import’? (Art. 50, Vienna and Bruxelles Codes). The motives behind this Article were lofty, no doubt, but its practical application led to countless contro- versies and abuses, because the generality embodied in the Article was not accompanied by an elucidation of what was meant as an 1943] CROIZAT: THALICTRUM AMETRUM 3 homonym “universally regarded as non-valid,’ and what were “motives contestable or of little import.” Naturally, everybody thought of his own motives as being true and relevant, and of those of his opponents as “contestable or of little import.” To remedy this situation, the text now embodied by Article 61 was approved by the Cambridge Congress of 1930. Space forbids my entering into details, but I may at least point out that Miss L. Green, who is well informed on everything that was proposed and voted upon at Cambridge, states in her authoritative commentary on nomenclature (Emp. For. Jour. 10: 68. 19381) that: “All later homonyms should be rejected even if the earlier homonym is not an accepted name” (italics in Miss Green’s text). Much confusion reigns as to the meaning of invalidity as dis- tinct from illegitimacy in the sense of the Rules, for the very good reason that the Rules themselves use these terms in a loose and contradictory manner. Examples of this confusion are rife in the Amsterdam Code, and one at least may be cited here. Article 2 defines as illegitimate, names or forms contrary to an Article, and states that such names cannot be maintained. Article 638, on its part, prescribes that the name of a taxonomic group “must be re- jected when its application is uncertain’. Since such a name [nomen dubium] ‘“‘must be rejected,” it stands to reason that this name is illegitimate under the definition given in Article 2. How- ever, Recommendation xxxvii which immediately follows Article 63 authorizes the certification of a nomen dubium following an adequate taxonomic study made on the basis of new evidence (Art. 17, Rec. iii, Rec. xxxvii). Thus, Article 63 errs in stating that a nomen dubium “must be rejected,” branding it implicitly as illegitimate. Such a name is merely invalid, proof of this being the fact that this name can be used legitimately under certification. Since the Rules themselves are not clear as to the proper use of validity and legitimacy, it would be useless to argue here Wheeler’s contention that a nomen provisorium is both invalid and illegitimate. Sooner or later, a fundamental debate is bound to take place in a Botanical Congress about these concepts. Mean- while, I may contribute here a brief comment as to the meaning of validity and illegitimacy, once again using an example. As it is well known, the law orders that a testament must con- form with certain specified requirements, a part of the estate of the deceaésed going automatically to certain parties by reason of their being related with the author of the will. If the will is drawn against the law and, for instance, the estate is distributed in a manner which is forbidden by law, the will is illegitimate, and as such it cannot be maintained. A will, conversely, may be drawn according to the law, but before it takes effect it must go through the procedure of probating, and is not valid until probated. The probating of a will is exactly the same procedure as the valid publica- A MADRONO [Vol. 7 tion of a taxonomic name. Neither a will nor a taxonomic name is valid until it is probated or published according to the laws of the land or the Articles of the Rules of Nomenclature. Natu- rally, neither a will nor a taxonomic name is legitimate if it vio- lates the law of the land or the Articles. A will that violates the law and a name that violates an Article may be unimpeachable as to form, but can neither be probated nor maintained because they are faulty as to substance. This, in a nutshell, is the distinction that can briefly be made here between the concept of validity and that of legitimacy. It is high time that the Articles be carefully revised and amended in order that they be purged of pointless and confusing abuses of the proper terms ultimately leading to a flood of mistaken comments in the literature. The following synonymy is in order: THALICTRUM AMETRUM Greene in Muhlenbergia 5: 129. 1909. T. polycarpum S. Wats. in Proc. Am. Acad. Sc. 14: 288. 18793. Jepson, Fl. Calif. 1: 580. 1922; Munz, Man. South. Calif. Bot., 173. 19385; Wheeler in Rhodora 40: 318-320. 1938. Non Loret. Arnold Arboretum, Harvard University, Jamaica Plain, Mass., March 3, 1942. UNA NUEVA ESPECIE DE PINUS MEXICANO MaxiMino Martinez Pinus Douglasiana sp. nov. Arbor 20 m. alta; diametros 30— 50 cm.; coma densa rotundata. Cortex leviter scabris, rubescens, 2 cm. crassus, squamatus. Rami expansa; ramuli brunneo rubes- centis, valde scabri. Folia 5, triangularia, crassa, rectiuscula, pungentia, 25-33 cm. longa, marginibus denso serrulato, claro virore vel galbinus coloris, fulgentia, intus glauco in folia juniora. Hypodermo biformis usque endodermo penetrabilis, chlorenquima partitus; fascies-exterius endodermi incrassatus. Ductus resini- feri 3 in parenchymatis parte siti; fasces fibrovasculares 2, ap- proximati, patentibus. Vaginae persistentes, 20-30 mm. longae, squamatae, castanei rubescens, dein obscuro castanei. Strobilis junioribus erectis violaceo fuscus, subterminalibus, oblongis at- tenuatis, obtusis; squamae crassae, apex expansus vel erectus. Strobili maturi ovoidei, leviter asymetricus, deflexi, paulum in- curvati, in apex attenuati, fusco rubescens coloris, cadivus, 7.5— 10.5 cm. longis, terni vel quini. Pedunculi 12 mm. incurvi ad strobili adnatus. Squamae 28-30 mm. longae, 15 mm. latae; apex irregulariter, obtuso vel rotundato; umbo subquadrangulo vel polyangulatus, carina transversa patente, carina longitudinali depressa, fere complanata in basis strobili. Cuspide complanata, paulum patente, mucro cadivo. Semina obscura fere ovoidea, 5 mm. longa, ala 25 mm. longa 8 mm. lata, brunnea. Lignum mol- lis, album; resina fere nulla. 1943 | MARTINEZ: PINUS 5 Fic. 1. Distribucién del Pinus Douglasiana. Typum in Instituto de Biologia, Mexico; isotypi in Arnold Arboretum, Harvard University, Jamaica Plain, Massachusetts, United States National Herbarium, Washington, D. C. He tenido a la vista ejemplares procedentes de: Srinatoa. Batel, Concordia; Santa Lucia, Concordia, “‘pino real’; Cerro de Tecoripa, Sierra del Rosario; limites de Sonora y Chihuahua; Potrero de Bejarano, Badiraguato; Rosario. Nayarit. Juanacata, Jala, “pinabete.” Jatisco. San Martin Hidalgo; Tecolotlan; Cerca de Cuale, a 20 km. al E. de Bahia de Banderas; Ameca, ““pino blanco”; Atengo, “pino blanco o pino hayarin” ; Soyatlan, a 2200 m., “pino blanco” ; Concepcién de Buenos Aires. MicHoacan. Tiripitio. Oaxaca. El Barrio; Santiago Tlaxoyaltepec. Mexico. Cuautepec, Sultepec. Es arbol de unos 20 metros de altura, por 30 a 50 cm. de dia- metro a la altura del pecho; de copa redondeada y densa; con la corteza algo aspera de 2 cm. de espesor aproximadamente, rojiza y escamosa, dividida en placas irregulares. Ramas extendidas, agrupadas en la parte superior del tronco. Ramillas morenas con tinte rojizo y muy asperas, debido a la persistencia de la base de las bracteas, las cuales son anchas, salientes y contiguas. Se desca- man facilmente. Hojas en grupos de 5, triangulares, gruesas, casi derechas, y agudas, de 25 a 33 cm. de largo, con los bordes fina- mente aserrados, de color verde claro, algo amarillento, brillantes, con tinte glauco en las caras interiores, solamente visible en las hojas tiernas. El] hipodermo es biforme, muy grueso con 5 capas de células desiguales e irregularmente colocadas, y presenta entrantes, a veces dobles, que llegan al endodermo seccionando el 6 MADRONO [Vol.7 clorénquima; las paredes exteriores de las células endodérmicas son muy engrosadas. Tienen dos fasces vasculares contiguos, bien distintos, rodeados arriba y abajo de células de refuerzo; los canales resiniferos son medios y en numero de tres. Las vainas son persistentes, de 20 a 30 mm., escamosas abajo y anil- ladas arriba, de color castafio rojizo al principio y castafio obscuro después. Las yemas son coénicas, de color naranjado rojizo. Los conillos son moreno violadceos, erguidos, oblongos, subterminales, algo atenuados en ambas extremidades, romos, generalmente en grupos de tres, con escamas gruesas, armadas de puntas extendidas o dirigidas hacia el apice. Conos largamente ovoides, algo asimétricos, reflejados, ligeramente encorvados, atenuados hacia el apice, de color moreno rojizo, opacos, caedizos, de7.5a10.5cm. Se presentan en grupos de 8 a 5, sobre pedinculos de unos 12 mm., siempre encorvados, quedando con el cono cuando éste cae. Escamas de unos 20 a 30 mm. de largo, por 15 de ancho, de apice irregular, obtuso o redondeado, umbo irregularmente cuadrangular o poligonal, rugoso, quilla transversal patente y una saliente perpendicular poco marcada; apofisis irregular, sub- piramidal, algo levantada (casi aplanada en las escamas basales), cuspide aplanada o muy poco saliente, con espina pronto caediza. Semilla obscura, casi ovoide, de unos 5 mm., con ala de 25 mm. de largo por unos 8 de ancho, de color moreno. La madera es blanda, de color blanco; con muy escasa trementina. Se emplea en construcciones y para muebles. Este pino, quizé incluido por Shaw en el Pinus pseudostrobus var. tenuifolia (Benth.) Shaw, coincide con éste en la estructura de las hojas, pues el hipodermo forma entrantes, a veces dobles, que llegan al endodermo, pero dichas hojas son gruesas y fuertes, de 25 a 88 cm., tiesas y derechas, en tanto que las del Pinus tenuifolia son muy delgadas, flexibles y colgantes. Los conos en lo general coinciden con los del P. tenuifolia, pero las apofisis son mas gruesas. Teniendo en cuenta que las hojas no son delgadas, sino por el contrario, gruesas y fuertes, no puede convenirle la denomi- nacién de tenuifolia, ya que la caracteristica de éste, como clara- mente lo indica el nombre, es que las hojas son delgadas. El Arbol se encuentra en una zona relativamente amplia (desde Sinaloa a Oaxaca). Por tales razones he visto la conveniencia de considerarlo a parte con rango especifico. Su zona de vegetacién, como se ve, comprende Sinaloa, Nayarit, Jalisco, Michoacan, México y Oaxaca (probablemente también Guerrero), formando masas puras. Se le ve asociado con Pinus Lumholtziu, Pinus leiophylla y Pinus oocarpa. Se denominé en honor de la Sefiora Margaret Douglas, dama norteamericana, entusiasta admiradora de la Flora Mexicana, y protectora de los estudios de la misma. 1943] MARTINEZ: PINUS 7 f ay 7 Pirate 1. Pinus Dovetasiana. Fig. 1. Seccidn transversal de la hoja. Fig. 2. Rama yconos. (Dib. de M. Ornelas C.) 8 MADRONO [Vol.7 SUMMARY Pinus Douglasiana was perhaps included by Shaw under P. pseudostrobus var. tenuifolia (Benth.) Shaw, but it differs from this in its longer, stouter, leaves and larger apophyses of the cone scale. It occurs from Sinaloa to Oaxaca. It is named in honor of Mrs. Margaret Douglas. Morelia 61, México, D. F. Mayo de 1942. AN INVESTIGATION OF THE PRESENCE OF SILICEOUS RODS IN THE SECONDARY WALL OF WOODY TISSUE* Watrer M. ScHALL In 1920, Forrest B. H. Brown (3) proposed an explanation of differential wood shrinkage by stating that a skeleton of siliceous rods existed within the secondary wall of wood elements. He assumed that these rods, acting as a restraining framework, kept longitudinal shrinkage at a minimum. Since the presenta- tion of this explanation, many workers in cell wall structure (6, 7, 12) have referred to this siliceous skeleton or have tacitly assumed its presence, notwithstanding the fact that the micelle theory advanced by Nageli (13) in 1863 and substantiated by subsequent workers (1, 2,7, 11, 12, 17) is now generally accepted as the logical explanation for the shrinkage behavior of wood. The present study was undertaken not to explain the mechanics of shrinkage but rather to investigate the procedure employed by Brown (3), first to check his results and second, if similar results could be obtained, to interpret them in the light of the accepted theories regarding wood shrinkage. The author wishes to give acknowledgment and thanks to Dr. R. A. Cockrell for his interest and suggestions, both in conducting the investigation and writing the final manuscript. In repeating Brown’s (3) work, the following species of wood were used: Swietenia mahogoni, Trochodendron aralioides, Quercus alba, Cedrela sp., Pinus strobus, and a lapachol-forming species of Tecoma. Tecoma, according to Record (15, p. 582), is divided into four groups, “Prima vera,’ “roble, ~ “ipe peroba,, vand “lapacho” or “pao d’arco.” The “lapacho” group is charac- terized by having wood that is very “hard and heavy, has an oily olive-brown color, and the vessels are more or less completely filled with yellow crystalline substance (lapachol), which may give the surface the appearance of having been dusted over with sulfur. Ripple marks are always present and usually regular.” Lapachol (C,;Hi,0;) ‘““‘when moistened with ammonia or dilute 1 Contribution of the Department of Forestry, University of California, Berkeley. 1943] SCHALL: SILICEOUS RODS IN WOODY TISSUE 9 sodium carbonate, turns a deep wine-red, thus providing a reliable diagnostic feature.’ In order to give c.mparable results with Brown (3), the results of tests made on Tecoma sp. are com- pletely reported although the same tests with comparable results were made on the other material. If a siliceous skeleton were present in the secondary wall of woody tissue, a change in dimension might take place upon desilicification. To test this, following Brown’s (3) procedure, blocks of the foregoing species were placed in hydrofluoric acid to dissolve silica thereby eliminating its reputed restraining effect on shrinkage. In addition, duplicate control blocks were placed in hydrochloric acid which does not dissolve silica but which in other respects should have an effect on the cell wall substance similar to that of hydrofluoric acid. The blocks, ranging in size from one-half to one inch in linear dimension, with true radial, tangential, and longitudinal faces, were measured with a microm- eter and placed in boiling water. After five and one-half hours of boiling, during which the blocks were measured at one-half hour intervals, duplicate blocks of each species were immersed in separate solutions of concentrated hydrochloric and 52 per cent hydrofluoric acid. Measurements of the change in dimension were made after eighteen, twenty-three, twenty-eight and thirty- one hours’ immersion. After washing in running water for ten hours, and measuring again, the blocks were placed in an oven at 103° C. to remove the excess moisture. The moist blocks were then allowed to air-dry for eight days. Since the swelling values for each of the species followed the same general trend, only the values for Tecoma sp. are included. These are recorded in Table I. In order to supplement the block measurements, individual fibers obtained by maceration with concentrated nitric acid and potassium chlorate (4) were treated with both hydrofluoric and hydrochloric acid. The fibers were mounted in alcohol and measured both in length and width under the microscope. The alcohol was then allowed to evaporate and the acid was added. In no case did the fiber length change after contact with either acid for an hour. If silica in the form of rods was present, the time in hydrofluoric acid was sufficient to dissolve the silica from the fiber. Since no measurable change was observed, it must be concluded that the acids did not remove the restraining force holding the fiber cell wall together, and the cell wall must have been fully swollen before the acids were added. The effect of chemical treatment on the physical characteris- tics of the wood was marked. In several cases the blocks showed a decided tendency to check and some even showed some collapse. These same blocks were brittle and a slight pressure on any of the faces caused splitting that took place tangentially along the annual rings as well as radially along the rays. It is not likely that merely drying the wood would cause such extreme stresses to 10 MADRONO [Vol. 7 TABLE I Percentage Swelling for Tecoma sp. in Water and Acid* Longi- Time Tangential rar 14, hour 5.4 7.0+ 1 hour 6.0 11.0+ 114 hours 6.3 12.7+ Per cent change 21% hours 61% 127% in boiling water 3% hours 7.0 41, hours T.0 13:07 51% hours 7.1 Per cent change 1814 hours 12:2 73 in concentrated hydro- 23 hours 12.8 7.6 chloric acid at room 28 hours 12.8 7.5 temperature 31 hours 12.8 thes Per cent change at air dry moisture 8 days —4.7 —43 content Per cent change 1é6hours | 8.2 15.0¢ | 44 11.0¢ | 0.5 0.1+ in 52 per cent hydro- 20%) hours | 7.7 85.0t | 46 43.0+ | 0.5 —18.0+ fluoric acid at room 271% hours T5510} 4.4 29.0+ 0.3 —22.0+ temperature 301% hours 7.5 4.3 0.4 Per cent change at air dry moisture 8 days 0.1 —13.8¢ | 2.2 —19.0+ | 0.0 -—40.0+ content ; | * Values represent averages for at least four determinations. Negative values represent shrinkage. + Results as obtained by Brown (3) for similar treatment. be set up but it is probable that the splitting action was caused by a material weakening of the cell wall by chemical action. These checked blocks were disregarded and only apparently sound blocks were used for measurement. In determining the air-dry moisture content before and after acid treatment, it was found that the hygroscopicity had de- creased. After treatment with hydrochloric acid, the moisture content of the blocks was reduced to three-fourths of the original value and after treatment with hydrofluoric acid to one-third. Again, this would indicate that the acids caused a chemical and physical change in the minute structure of the cell wall so that its original equilibrium moisture content was sig- nificantly reduced. | Each of the untreated blocks was sectioned on a microtome, using a jet of a steam to soften the woody tissue. The sections, mounted in water, were examined at 1300 diameters under a Zeiss binocular microscope equipped with an oil immersion apochro- matic objective (N.A. 1.3) and 10x compensating eyepieces. Likewise, sections mounted in glycerine were examined and even after contact with the liquid for twelve hours, in no case was there any indication that small isolated areas, supposedly the cut ends 1943] SCHALL: SILICEOUS RODS IN WOODY TISSUE 11 of siliceous rods, were present. The photomicrograph of a cross section of Tecoma sp. (fig. 1) taken at 810 diameters shows no discontinuities in the cell wall. The index of refrac- tion of the cell wall, as found by Brown (3), was verified using Mc- Lean’s Solution (4) of known indices on cross sections of Tecoma sp. Since the index of re- fraction of the secon- dary wall and silica are practically the same, the proximity of the two indices may have obscured any possible difference due to the presence of silica. Both the cross sec- tions and the individual fibers of Tecoma sp. were incinerated to see if visual evidence of the rods could be found during the course of in- Fic. 1. Transverse section of a lapachol- cineration or in the ash. forming species of Tecoma. x 810. Individual cells were isolated from a section 14 microns thick and incinerated over an alcohol flame. The heating time was lengthened for successive sections so that examination could be made at varying degrees of incineration. Under the microscope, the ash showed irregulari- ties but no regular arrangement of bodies was noted. According to Uber (19), these irregularities are probably a result of a natu- ral tendency of the ash to check upon shrinkage. Macerated fibers were incinerated the same way but again no visual evidence of the localization of silica was obtained. In addition to determining the effect of the presence of a siliceous skeleton on the shrinking and swelling of wood, the actual amount of silica present was determined. Following Brown’s (3) procedure, this determination was carried out in two steps. The first was to determine the ash content and the second to determine the per cent silica in the ash. An oven-dry sample of Tecoma sp. was accurately weighed and ashed in an electric muffle at moderate red heat. The weight of ash thus obtained was 0.25 per cent of the dry weight of wood. A silica determination based on the weight of ash was made using hydrochloric and per- chloric acid in the quantitative analysis (9). The weight of 12 MADRONO [Vol. 7 silica thus obtained gave an average of 0.85 per cent of the dry weight of ash. The values obtained for ash and silica content are consider- ably below those given by Brown (3). He states that, “after combustion, 1.8 per cent (of the dry weight of the fiber) of min- eral matter was obtained and 0.1 per cent (of the dry weight of the fiber) of silica or silicic acid.” These percentages are of little value since they are based on the dry weight of fiber which can be obtained only arbitrarily. The maceration procedure is ex- tremely variable since neither the time nor the temperature of reaction is standardized. A trial maceration using concentrated nitric acid and potassium chlorate (15), gave a dry weight of fibers for Tecoma sp. equal to 42 per cent of the dry weight of wood. Based on the dry weight of fibers, this would give values for ash and silica content approximately twice as large as results based on the dry weight of wood if the assumption was made that all the mineral matter was retained by the macerated material. A comparison of ash and silica content cannot be made with any marked degree of accuracy, however, since within a single tree, the values vary from periphery to pith and from the base to the top of the tree. It is not outside the realm of possibility that inorganic mate- rials are centralized within certain areas in the cell wall. Kerr and Bailey (11, p. 285) state that “the central layer of normal tracheids, fiber-tracheids, and libriform fibers is composed, in all cases, of a complex and firmly coherent matrix of cellulose with elongated, intercommunicating interstices. Within these interstices more or less ‘lignin’ and other non-cellulosic constit- uents may be deposited.” Bailey (1) recently states specifically that minerals may be deposited within these interstices. Since the mineral content occupies such a small percentage of the weight of materials within the interstices, it is improbable that silica would be so localized as to form a continuous rod. In view of these findings, it is not likely that a highly silicified skeleton is present in the secondary wall of woody tissue. The results of swelling tests alone should be conclusive evidence of this fact since several species of wood were used and the blocks treated with two mineral acids, one of which would dissolve silica while each should have similar effects on the cell wall structure. If siliceous rods were present they would certainly have been disintegrated after immersion in 52 per cent hydro- fluoric acid for 80 hours. As indicated in Table I, a small increase in size was observed but no appreciable difference was noted be- tween the blocks placed in hydrofluoric acid and those placed in hydrochloric acid. Furthermore, the individual fibers failed to give a measurable change when each of the acids was added. In a schematic drawing of a fiber before and after hydrofluoric acid treatment, Brown (8) indicates that the diameter increases and the length decreases with the acid treatment. He explains the 1943 | SCHALL: SILICEOUS RODS IN WOODY TISSUE 13 change by stating that the siliceous rods had been broken, thus freeing the so-called homogeneous substance which could then swell without hindrance. If the cell wall is a homogeneous substance, the fiber should have increased in length and width in the same proportion and an inverse relationship could not have resulted. The acids probably caused a degradation of both lignin and cellulose as well as attacking any of the minerals in the wood. This is borne out by the mechanical weakening of the wood and by the reduced hygroscopicity. Chamberlain (5) and Sacc (17) state that the action of hydrofluoric acid is to soften wood but they do not point out the manner of softening. Plowman (14) in 1904 stated that the action of hydrofluoric acid was to remove silica and other mineral deposits from the wood. Kerr (10) and Har- low (8) point out that the action of hydrofluoric acid is to attack the cell wall substance and the removal of silica is a secondary operation. Harlow (8) explains the softening action as a de- gradation of lignin since the treated wood does not respond to the Maulé reaction which is employed as a test for the presence of lignin. Kerr (10), on the other hand, states that the action is due to a degradation of cellulose to hydrocellulose and explains the action by drying other acids into wood in order to soften the woody tissue. Rudiger (16) points out that swelling precedes the dissolution of cellulose in liquid hydrofluoric acid and that the lignin structure of the membranes was destroyed, although the lignin itself did not swell. Forsaith (6) in his explanation of dif- ferential swelling of wood with adsorption of water credits the siliceous skeleton as proposed by Brown (8) as exerting an influ- ence in conjunction with the micelle theory as proposed by Nageli (13). A more recent worker, Maby (12, p. 434), in his explanation of shrinkage and swelling, says, “On the other hand, the longitudinal siliceous strands in the cell wall, noted and described by F. Brown, might be expected to exert a binding effect over dimensional changes in the longitudinal direction.” Schorger (18, p. 9) states, ‘“Nageli, as a result of his study of the growth of starch grains and the cell wall, concluded that the cell wall consists of ultramicroscopic, crystalline, molecular com- plexes which he called micellae. By this assumption he was able to explain striation, stratification, swelling, double refraction, and other properties of the cell wall.’ Subsequent workers have followed this general idea. Bailey (1, 2) points out that the cellulose consists of chains of anhydrous glucose residues which tend to aggregate in a parallel fashion. He also states that the aggregation of chain molecules is not in separate groups but rather a part of a continuous system which is held together by over- lapping chain molecules and perforated by intercommunicating spaces. It is probably in these intercommunicating spaces that water is adsorbed and causes the changes in the dimension of wood. The mineral content is also probably localized here but 14 MADRONO , [Vol.7 evidence of a continuous bond between inorganic material of the cell wall has not been found. CoNCLUSIONS 1. Silica in the form of continuous siliceous skeletons is not present in the secondary wall of woody tissue. 2. The silica content is such a small percentage of the total weight of wood that it could not have an appreciable effect, greater than other minerals, on the differential swelling or shrink- ing of wood. 3. Other than bringing about more rapid degradation of the wood substance, hydrofluoric acid is similar to hydrochloric acid in its action on the cell wall. University of California, Berkeley, May 28, 1942. LITERATURE CITED 1. Battey, I. W. The cell and protoplasm. Publ. Amer. Assoc. Advance- ment Sci. no. 14: 31-43. 1940. 2 . The microfibrillar and microcapillary structure of the cell wall. Bull. Torr. Bot. Club 66: 201-213. 1939. 3. Brown, F. B. H. The siliceous skeleton of tracheids and fibers. Bull. Torr. Bot. Club 47: 407-424. 1920. A. The refraction of light in plant tissues. Bull. Torr. Bot. Club 47: 243-260. 1920. 5. CHampertain, C. J. Methods in plant histology. Univ. of Chicago Press. 1938. 6. Forsairu, C. C. The technology of New York State timbers. Tech. Publ. 18. N. Y. State Coll. Forestry. 1926. 7. Hawtey, L. F., and L. E. Wise. The chemistry of wood. The Chemical Catalog Co., Inc., New York. 1926. 8. Hartow, W. H. Contributions to the chemistry of the plant cell wall. IV. Tech. Publ. 26. N. Y. State Coll. Forestry. 1928. 9. Hittesrann, W. F., and G. KE. F. Lunpextt. Applied inorganic analysis. John Wiley and Sons, New York. 10. Kerr, I. The action of hydrofluoric acid in softening. Tropical Woods 40: 37-41. 1934. 11. Kerr, I. and I. W. Batrey. The visible structure of the secondary wall... . Journ. Arn. Arb. 16: 273-300. 1935. 12. Masy. J. C. Micellar structure of the tracheid wall in certain woods in relation to morphogenetic and mechanical factors. New Phyt. 35: 432- 455. 1936. 13. NAcet1,C. W. Die Anwendung des Polarisationsmikroskops auf die Unter- suchung der organischen Elementartheile. Beitr. Wiss. Botan. 3: 1-126. 1863. 14. Prowman, A. B. The celloiden method for hard tissues. Bot. Gaz. 37: 456- 461. 1904. 15. Recorp, S. J. Timbers of tropical America. Yale Univ. Press. New Haven, Conn. 1924. 16. Rupicer, W. Papierfabrickant. 38: 9-13. 1940. 17. Sacc, J. E. Elements of botanical microtechnique. McGraw-Hill. New York. 1940. 18. Scuorcer, A. W. The chemistry of cellulose and wood. McGraw-Hill. New York. 1926. 19. User, F. M. Microincineration and ash analysis. Bot. Review 6: 204-226. 1940. 1943 ] KEARNEY: NEW COWANIA FROM ARIZONA 15 A NEW CLIFF-ROSE FROM ARIZONA Tuomas H. Kearney It was noted in a recent publication’ that an apparently unde- scribed Cowania occurs in western Arizona. The writer is now convinced that this plant merits recognition as a species. Cowania subintegra sp. nov. Frutex parvus, ramulis ad- scendentibus-patentibus 80-75 cm. longis, cortice albo-griseo; folia nec viscida nec distincte glanduloso-punctata, integra vel nonnunquam apicem versus 1—2-dentata, usque ad 15 mm. longa et 3 mm. lata, spathulata, apice obtusa, basi attenuata, margine valde revoluta, supra laetevirentia et parce araneosa, subtus dense albo-lanata; ramuli, pedicelli, hypanthium, et calycis lobae tomen- tosi sine glandulis stipitatis ; pedicelli 4-11 mm. longi hypanthium subaequantes vel valde superantes; hypanthium infundibuliforme 5—7 mm. longum; petala ochroleuca. A straggling shrub with stems up to 75 cm. long, the branches ascending-spreading, the bark pale gray, becoming somewhat shreddy; herbage not viscid; twigs, pedicels, hypanthium tube, and outer face of the calyx lobes whitish tomentose, without stipi- tate glands; leaves up to 15 mm. long and 3 mm. wide but usually shorter and narrower, mostly 1-veined and entire but occasionally with 1 or 2 subapical rounded teeth, oblanceolate, obtuse at apex, attenuate at base, the margin strongly revolute, thick, minutely and obscurely glandular-punctate, bright green and loosely arach- noid-pubescent over the whole upper surface, the lower surface densely and conspicuously white-lanate; pedicels 4 to 11 mm. long, nearly equaling to much longer than the hypanthium; hypanthium narrowly funnelform, attenuate at base, 5 to 7 mm. long, in anthesis about 8 mm. wide at summit; calyx lobes about 4 mm. long, broadly ovate, rounded at apex and sometimes ob- securely apiculate, denticulate-ciliolate, spreading at anthesis, becoming reflexed; petals ochroleucous, about 10 mm. long and 6 mm. in greatest width, rounded and slightly erose (occasionally shallowly cleft) at apex, cuneate at base; stamens about 40, the filaments about 4 mm. long, the anthers 1.25 mm. long, nearly orbicular; pistils 8 to 7, the ovary short-stipitate, densely se- riceous, the style 6 to 7 mm. long at anthesis, sericeous on the lower one-half to two-thirds, naked above; achenes about 6 mm. long, narrowly obpyramidal, glabrous except near the apex, the persistent style about 25 mm. long (perhaps longer at full ma- turity), silky-plumose with long antrorse hairs except the apical portion, this naked, 2 to 8 mm. long. The type was collected about two miles west of Burro Creek crossing on the road from Wikieup to Hillside, southeastern Mohave County, Arizona, near the Yavapai County line, altitude 1 Kearney, Thomas H., Peebles, Robert H., and collaborators. Flowering plants and ferns of Arizona. United States Department Agriculture Misc. Publ. 423: 405. 1942. 16 MADRONO [Vol. 7 2,500 feet, April 18, 1941 (Darrow & Benson 10891). The species is known only from the type locality, where it had been collected first on April 20, 1938 (Darrow & Crooks 3). On both dates of collection, only a few late flowers persisted, but the fruit was not yet fully mature. Dr. Lyman Benson stated (personal communi- cation) : “The base of the plant is perhaps as much as one to one and one-half inches thick, but the trunk continues for only a few inches above ground.” He also reported: ““We found the plant growing in a rather limited area on disintegrated material of a peculiar white rock and associated with a vegetation entirely dif- ferent from that on surrounding territory. However, although the distribution of the plant was restricted, it was locally abun- dant.” Cowania subintegra strikingly resembles C. ericaefolia Torr. of western Texas in habit, appearance, small stature, and very nar- row, mostly entire leaves, but the Texas species has linear, sharply cuspidate leaves not more than 6 millimeters long, stipitate glands on the hypanthium (often also on the pedicels), and darker colored bark. The flowers and fruit of C. subintegra apparently present no characters that are not within the range of variation of C. Stans- buriana Torr., but that is a much larger and more erect shrub, attaining (exceptionally) a height of seven and one-half meters. The branchlets are more stiffly ascending, the bark reddish brown or dark gray, and the herbage usually very viscid. The leaves are much larger, cuneate-obovate in outline, pinnately veined and deeply 8-cleft with the terminal lobe in turn 38-toothed or 3-cleft, and they are nearly always conspicuously punctate with few large glands. With the single exception, so far as the writer knows, of the specimen noted in the next paragraph, the pedicels and hypanthium are beset with stipitate glands and are not tomentose, or but thinly so. A collection from near Rye Creek, Gila County, Arizona (Collom 97 in 1933), has the herbage non-viscid, the leaves mi- nutely and inconspicuously punctate, and the pedicels and hy- panthium densely pubescent and lacking stipitate glands. In these characters it resembles C. subintegra but in other respects it is not distinguishable from C. Stansburiana. The writer believes that Jepson was justified in reducing C. Stansburiana to a variety of C. mexicana D. Don (Man. FI. PI. Calif. 498. 1925). The diagnostic characters given by Rydberg in his key and descriptions (N. Am. Fl. 22: 415, 416. 1913) are as follows: C. mexicana: Hypanthium campanulate, abruptly contracted into the pedicel. Glands of the pedicel sessile and often hidden in the tomentum. C. Stansburiana: Hypanthium funnelform, gradually contracted into the pedicel. Glands of the pedicel stalked. Standley (Contrib. U.S. Nat. Herb. 23: 326. 1922) mentions also that the leaf lobes are entire in C. mezicana, whereas at least 1943 | KEARNEY: NEW COWANIA FROM ARIZONA 17 the terminal lobe is cleft or dentate in C. Stansburiana, as was pointed out by Torrey in his original description (in Stansb. Expl. Great Salt Lake, 386. 1852). The color character of the bark given by Standley does not hold, many specimens of C. Stansburi- ana from Utah and northern Arizona having brown bark. The manner in which the several characters are associated in specimens from Mexico and from the United States is shown in Table 1. It is evident that whereas in most of the specimens the Taste 1. Association of characters in specimens of Cowania in the United States National Herbarium that have been referred, respectively, to C. mexicana and C. Stansburiana. ; Stipitate glands Ea leat Hypanthium on hypanthium, sel etc. Species and Campan- collection ulate, + | Funnel: Enti Toothed | abruptly 4 ntire Berk Bone attenu- | Present | Absent tracted ae au at base a C. mexicana Rose 11659, Cusi- huiriachic, Chih. .. x xX x E. Palmer 12, Tepe- huanes, Dur. ...... x xX xX E. Palmer 71, Papa- squiaro, Dur. ..... , X xX x E. Palmer 4669, Papa- squiaro, Dur. ..... xX xX X Dugés in 1899, moun- tains, Guanajuato . xX —- —- — — C. Stansburiana Most specimens from Utah, Arizona, etc.. xX xX x Hartman 276, Nacori, SOM te, Ae fhe ge cats xX X X M. E. Jones 5586c, | Provo, Utah ....... | X xX x V. Bailey 1457, near St. John’s, Ariz. ... xX xX x Knowlton 238, Grand Canyon, Ariz. ..... X xX x Collom 97, Rye Creek, PNIZ et MO ok a bic xX X X characters considered by Rydberg and by Standley to be diag- nostic of C. mexicana and C. Stansburiana, respectively, tend to be associated as indicated by them, there are several marked excep- tions. Thus two of the five specimens referred to C. meaicana have the hypanthium funnelform or intermediate,? rather than 2D. Don, in his description of the genus Cowania, based solely upon C. mexicana (Trans. Linn. Soc. London 14: 575. 1825), states: “calyx [hypanthium | obturbinatus basi attenuata tubulosus.” His illustration (Tab. XXII), how- ever, shows the hypanthium as campanulate and abruptly contracted. 18 MADRONO [Vol.7 campanulate, and three specimens from Utah and Arizona that are C. Stansburiana in all other characters have a campanulate hypanthium that is abruptly contracted at base. One of the Mexican specimens (Dugés in 1899) that has the entire primary leaf lobes of C. meaicana, is intermediate in shape of the hy- panthium and the latter is conspicuously glandular, although the glands are sessile or nearly so. The specimen from Rye Creek, Arizona, as was noted in a preceding paragraph, although con- forming to the characterization of C. Stansburiana in shape of the leaves and of the hypanthium, lacks the stipitate glands. It is also aberrant in having the leaves obscurely and minutely punc- tate, not conspicuously and coarsely so, as in all other specimens of C. Stansburiana and in all specimens of C. mexicana examined by the writer. Bureau of Plant Industry, United States Department of Agriculture, Washington, D. C. June 27, 1942. FRANCESCO FRANCESCHI JOHN M. TucKErR This paper is a brief account of the life and work of the man who stands out above all others in the history of horticulture in southern California—Dr. Emanuele Orazio Fenzi, known to his associates in this country in later life as Dr. Francesco Franceschi. In gathering data I drew upon a number of sources, and take this opportunity to express my appreciation to the following persons for the assistance they have given me: to Dr. Emily O. Lamb, who lived with the Fenzis at Santa Barbara as a member of the family for fourteen years, to Mr. Peter Riedel and to Mr. H. M. Butter- field, for much of the information contained in the following pages; to Mr. Butterfield, Miss Annetta Carter, and Mr. M. Van Rensselaer, for the loan of horticultural catalogues and journals —sources of much valuable data; to Dr. H. L. Mason, for placing at my disposal a collection of Franceschi’s business correspon- dence (a fund of information of which I have scarcely scratched the surface), and to Dr. Howard S. Reed, for guiding my efforts in preparing this paper. Particularly informative also, were the following two articles: “Una gloria dell’orticoltura italiana. I] Dott. Emanuele Orazio Fenzi,’ by Mario Calvino—(L’Agricol- tura Coloniale, 22: 122-128. 1928.) and “Dr. Fenzi’s Contribu- tions to American Horticulture,’ by F. W. Popenoe (Journ. Hered. 18: 215-220. 1922.). Emanuele Orazio Fenzi was born March 12, 1848, in Florence, Italy. His grandfather was a very wealthy banker of that city and a senator, a man of an aggressive dominating personality. The Fenzi family were patrons of the arts and sciences, and fol- lowed the latest developments in these fields with great interest. 1943 | TUCKER: FRANCESCHI 19 The renowned pianist, Anton Rubinstein, was a friend of the fam- ily, and on his visits to Florence, used to come to their home to practice, because their piano was the finest in the city. Thus, the background of young Fenzi was one of culture and wealth, with its attendant advantages. In early youth he lost his parents, and his grandfather took it upon himself to see that he was properly educated. The grand- father, seeing his young charge as a future man of affairs, sent him to the University of Pisa to study law. Although he would far rather have studied botany, he complied with his grandfather’s wishes and in 1864 received the degree of Doctor of Laws. After leaving the university, he turned to botanical and horti- cultural pursuits, despite the fact that his grandfather wanted him to start upon a business career. Endowed with a large estate which made him financially independent, he was able to indulge his tastes as he chose in the years that followed. At his country place near Florence, he formed an arboretum of rare trees, and on the estate of a relative near Rome, he assembled a large collec- tion of plants from countries all over the world. He was the first to introduce bamboos to Italy; Genista monosperma was another of his introductions. He did considerable work in the improvement of grape and olive culture in Italy, and frequently contributed horticultural and botanical articles, not only to Italian periodicals, but to the English journal, “The Gardener’s Chronicle,’ as well. Fenzi travelled a great deal, and at one time or another visited all the principal botanical gardens of Europe. In May, 1874, he served as secretary of the International Agricultural Exposition at Florence and edited its catalogue. Shortly thereafter, he became the first secretary of the Royal Tuscan Society of Horti- culture, an organization of which he was later president. His activities in later years were not confined to the fields of horticulture and agriculture, however. When his grandfather died, he took charge of the affairs of the bank and, in addition, managed an estate. He was instrumental in establishing an elec- tric tram line to Fiesole, the first in Italy, and the steam lines at San Casciano and Greve. But he had so little enthusiasm for business, that he soon turned over the management of the bank to a cousin. Then, during the economic crisis of 1889-90 Fenzi was forced to close the bank. In order to settle accounts with its creditors he found it necessary to liquidate virtually everything he owned, so that, finally, he and his family were left with only a very small fraction of their once large fortune. Because of his losses, he was no longer hampered by a multi- tude of business matters, and saw a chance to put his interest in plants to work. His ambition was to gather together in one area plants from countries all around the globe. He decided to go to southern California because the climate was well suited to his purpose. So, in 1893, he came to Los Angeles, his wife and fam- 20 MADRONO [Vol.7 ily remaining in Italy. He was in California six years before they joined him. From a strong feeling of family pride he dropped the surname, Fenzi, lest the stigma of his bank failure follow him to America, and adopted in its place a family name, Franceschi. During the twenty years he spent in California, he was known to all but his intimates, as Dr. Francesco Franceschi. In Los Angeles, he met J. C. Harvey, the elder Mr. Howard, E. D. Sturtevant, and other horticulturists and nurserymen. He re- mained there only a year, however, before he moved to Santa Bar- bara where he entered into partnership with C. F. Eaton in raising nursery stock. After a short time, the partnership was dissolved and Franceschi started a nursery business of his own, calling his organization the Southern California Acclimatizing Association. To obtain new species he wrote to botanical gardens, collec- tors, and plantsmen in all parts of the world, and soon developed an extensive correspondence. Here the unusual linguistic ability Franceschi possessed stood him in good stead, for he read, wrote and spoke not only his native Italian, but also English, German, French, Spanish, and modern and ancient Hebrew. Yet his interest was far from being confined to exotics. He was ever on the lookout for any elements of the native flora that might possess striking ornamental qualities. Indeed, in Novem- ber, 1894, having been in Santa Barbara less than a year, he made a week’s trip to Santa Cruz Island, the largest of the Santa Bar- bara Channel Islands. While there he obtained seeds of several different species which were at that time unknown in the horti- cultural trade. The most noteworthy of these was Lyonothamnus floribundus var. asplenifolius, the Santa Cruz Island ironwood. Unable to find any seedlings of this tree, he laboriously dug up a living stump and gathered some seed. With considerable dif- ficulty, he managed to carry his prize back to camp, and on his return to the mainland planted it in his lathhouse in Montecito, a few miles from Santa Barbara. In five or six months it had started to sprout. The next year, when he moved his nursery to Santa Barbara, he transplanted it to his new location where, in a few years, it developed into a fine tree. From the seed, Fran- ceschi obtained several trees, one of which is the fine specimen to be seen today in the grounds of the old botanical garden north of the library, on the campus of the University of California at Berkeley. He had been in Santa Barbara only a year, when, in 1895, he published a small book entitled, ‘““Santa Barbara Exotic Flora.” This book contains a good deal of meteorological and climatic data, and observations on the soil and native flora of the region. It includes notes on the history of plant introduction in the region, and mentions the two oldest introduced trees of the town, Casi- miroa edulis, the White Sapote, and Prunus Capuli, the Capulin Cherry, both natives of Mexico. This volume gives an appar- ently comprehensive review of the exotic plants then cultivated in PLatTeE 2. FRANCESCO FRANCESCHI. : Ni Uh sare t 4 nate / : han } : if x : fi So ‘ } 1943] TUCKER: FRANCESCHI 21 Santa Barbara and evidences the keenness of Franceschi’s obser- vations, and the breadth of his botanical knowledge. David Fairchild, in his recent book, “The World Was My Garden,” tells of meeting Franceschi at Santa Barbara in 1898. He speaks of Franceschi’s enthusiasm in the following words: “Santa Barbara in 1898 was but a simple, small town. Residents of the beautiful hillside villas today would not credit their eyes could they visualize the bare, sparsely settled roads where I drove with Dr. Franceschi. . . . Santa Barbara was so undeveloped that I considered him visionary and over-optimistic. However, he foresaw the future more clearly than I, and lived to see Santa Barbara become a great winter resort containing hundreds of beautiful villas like those on the Riviera.’ Fairchild, at that time with the United States Bureau of Plant Industry, was impressed with the work Franceschi was doing, and had numerous new plant introductions of the Bureau sent to him from time to time for trial in Santa Barbara. In 1904 Franceschi acquired forty acres of land on Mission Ridge, at that time a dry, barren hillside entirely outside the bounds of the city. Here he built the house in which. he lived until he left Santa Barbara, a place he named “‘Montarioso.” He established another nursery here, and, in order to have an ample water supply, built a small reservoir near the top of the hill. A perusal of his business correspondence leaves one with the impres- sion that he was not blessed with much financial success during the following years. In June, 1904, his propagating house burned and he appears to have been considerably in debt in the months that followed. In 1907 he went into partnership with Mr. Peter Riedel, and incorporated the Southern California Acclimatizing Association, hoping thereby to be relieved of some of the burden of routine business matters. This arrangement did not work out well, however, and after little more than a year, in 1909, they decided to dissolve their partnership. Following this incident, Franceschi continued his business independently on Mis- sion Ridge, calling it the Montarioso Nursery. By offering his services in landscaping and maintenance of the grounds of a num- ber of estates in Montecito, he was able to bolster his income. However, the continued cost of introducing new plants, a work Franceschi carried on despite his reverses, and the limited demand for his exotic rarities, made financial disaster inevitable. It is not surprising, therefore, that upon receiving an offer from the Italian government in 1912, to take a post in the African colony of Libya, he decided to accept. He was to introduce new plants having agricultural and horticultural value, and to do what he could to develop agriculture in the colony, the government furnishing the land and facilities for his introduction grounds. Accordingly, on July 21, 1918, he bade farewell to Santa Barbara, and with his wife and daughters, started out for his native Italy. 22 MADRONO [Vol. 7 His two sons remained in California, the older carrying on the business as manager of the Montarioso Nursery. For the next year and a half, Franceschi lived in seclusion on the Italian Riviera, preparing the manuscript of a book, “Frutti Tropicali e Semi-Tropicali,” which was published at Florence in 1915. A short time later, he made a reconnaisance trip to Libya, accom- panied by his nephew, Guido Corsini, and in February, 1915, laid out his establishment in the city of Tripoli. Thus, at an age when most men would have been long since retired, Franceschi, at 72, was setting out on a fresh venture. He held this post for several years, and then retired to carry on the same work at his own expense. In 1922, he was awarded the third Meyer Memorial Medal by the council of the American Genetic Association. This is a medal presented periodically to persons who have accomplished out- standing work in introducing plants to American horticulture. The medal was sent to Tripoli, in 1923, by the governor, Count Volpi. Franceschi continued his literary activity to the end of his life, contributing articles on agriculture in Tripoli to the Italian jour- nal, “L’Agricoltura Coloniale.” He died in Tripoli on November 5, 1924, at the age of 81. Franceschi’s oldest daughter has car- ried on his work in Tripoli, and today maintains a successful nursery there. She, apparently, had, in addition to a deep inter- est in things botanical, considerable business acumen. Probably the most important accomplishment of the two in Tripoli, has been the importation and propagation of large numbers of euca- lyptus trees of various species. HortTicuLtTURAL ACHIEVEMENTS During the years that he lived in Santa Barbara, Franceschi wrote numerous articles for the local newspaper, and was a regu- lar contributor to such journals as “Pacific Garden” and “Rural Californian.”’ He did much toward developing an appreciation of the beauty as well as the economic value of the large number of ornamental plants and tropical and subtropical fruits that could be grown in the region. His reputation spread throughout horti- cultural circles in this country, and every botanist or plantsman who visited Santa Barbara, was certain to pay him a call to see his collection of rare plants. The esteem in which his judgment and opinions were held is shown by the frequency with which he is quoted in the literature of tropical and subtropical plants. Among the multitude of new plants that he brought to the gardens of Santa Barbara, the following are some best suited to the climate there. A number of these have attained the popular- ity they deserve, although some, none the less fine ornamentals, are still rather rare. 1943] TUCKER: FRANCESCHI 23 Acacia obliqua Harpephyllum caffrum Acacia podalyriaefolia Hibiscus heterophyllus Aglaia odorata Jasminum simplicifolium Aleurites Fordii Lippia repens Alée Salm-Dyckiana Lithraea Gilliesii Anthyllis Barba-Jovis Lyonothamnus floribundus var. Asparagus decumbens asplenifolius Asparagus scandens var. de- Metrosideros tomentosum flexus Myoporum acuminatum Bauhinia grandiflora Myoporum tomentosum Bauhinia tomentosa Pithecoctenium clematidium Bauhinia variegata Pithecoctenium muricatum Benthamia fragifera Pittosporum heterophyllum Bocconia frutescens Pittosporum rhombifolium Buddleia madagascariensis Pittosporum viridiflorum Carica quercifolia Psidium lucidum Convolvulus florida Rhynchosia minima Dioclea glycinoides Schinus terebinthifolius Dombeya natalensis Schotia latifolia Dombeya punctata Solanum Guatemalense Erythrina tomentosa Sterculia discolor Eugenia edulis Stigmaphyllon littorale Feijoa Sellowiana Taxodium mucronatum Ficus altissima Tecoma garrocha Ficus infectoria Tipuana speciosa Ficus retusa Tricuspidaria dependens Genista monosperma _ Vitis capensis Of all the new plants Franceschi introduced, none has be- come better known than Lippia repens. Its popularity is probably due in large part to the publicity given it by Franceschi, but the importance he attached to this particular introduction appears to have been based on an erroneous idea that he held. In 1904, in an article that he wrote for the Los Angeles Times, Franceschi states that he first introduced this species from Italy in 1898. He says: “From the Director of the Botanic Garden in Rome I obtained by mail a small tin box of Lippia plants, less than 12 ounces weight. Now, after six years, there are hundreds and hundreds of acres planted with Lippia between California, Ari- zona, Mexico and Australia, and it all came out of that small time DOX.)... | In contrast to this remarkable statement, H. N. Moldenke, an authority on the Verbenaceae, in a personal communication of May 26, 1941, makes the following comment: “You can be very sure that the plants (of “Lippia repens’’) of Texas, New Mexico, and Arizona, and central, eastern, and southern United States have nothing to do with the plants introduced into California by Franceschi, but it is my belief that most of those of southern California ‘(at least all that I have seen so far) are descendants of the ones he introduced.” 24 MADRONO [Vol.7 The efforts Franceschi made to secure new plants, to make certain of their identity, and to obtain accurate information re- garding their culture and optimum growth conditions, were often great, and show a truly scientific spirit. Scattered throughout his correspondence are letters to and from such well-known botanists as Joseph Burtt-Davy, William Trelease, J. H. Maiden of Australia, Charles Sprague Sargent, Harvey Monroe Hall, Miss Alice Eastwood, and T. S. Brandegee, requesting and receiving identifications of specimens which he had submitted. As another illustration, his letter of October 80, 1908, to Mr. C. Wercklé at San Jose de Costa Rica, might be cited. Wercklé was the dis- coverer of Hidalgoa Wercklei, the “Climbing Dahlia,’ which Fran- ceschi had introduced at Santa Barbara. Having had difficulty in bringing it to flower, we find him, in this letter, requesting information regarding its mode of growth and climatic require- ments from the man who, logically, would know most about such matters—its discoverer. This thoroughness was typical of Franceschi. A good example of his persistence and determination in effect- ing the introduction of a desirable species, is the case of Tarodium mucronatum, the ““Montezuma Cypress.” Franceschi had often seen the tree in the Botanic Garden at Naples, planted by Tenore, the botanist who described the species, and its beauty had made a lasting impression upon him. In 1898 he decided to obtain seeds and try it out in Santa Barbara, and accordingly, sent to Naples for seed, which failed to germinate. Year after year, seeds from there and elsewhere persistently refused to germinate, but at last, in 1908, after ten years of failure, his patience was rewarded. Through his friend, Professor C. Conzatti of Oaxaca, he obtained, from the Federal Park at Chapultepec, Mexico, seeds which, much to his gratification, germinated successfully. Today, many fine trees grown from these seeds may be seen in the parks and gardens of Santa Barbara. The culture of tropical and subtropical table fruits was a sub- ject of special interest to Franceschi. He carried on much correspondence on aspects of this subject with men of the United States Bureau of Plant Industry, and horticulturists and fruit growers (the Popenoes of Altadena, California, particularly), throughout the warmer parts of this country and of many foreign countries. He introduced several new species of Anona, and from a superior tree of A. Cherimola growing in Altadena, he raised plants in 1910 which he subsequently sent out under the name of A. Cherimola mammillaris. His best known fruit introduction was the “Pineapple Guava,” Feijoa Sellowiana. This he introduced in 1901, obtaining seeds from France, where it had been introduced previously from its native South America. Although it was given much publicity at the time, Feijoa has not gained the popularity that Franceschi had hoped for it, and certainly has not attained any economic importance as a fruit in southern California. 1943 | TUCKER: FRANCESCHI 25 Of much greater consequence in this field, however, were the frequent articles he wrote for newspapers and _ horticultural journals. His book, “Frutti Tropicali e Semi-tropicali,” probably embodies the findings of his long years of experience in this field. It is a work of some 260 pages, contains numerous illus- trations, and descriptions of 727 species. The majority of these descriptions are rather brief, although for those which are of more economic importance he gives much more detailed information. An English translation of this book from the Italian would be a real contribution to the literature of subtropical horticulture in this country. The total number of different kinds of plants that Franceschi grew during his stay in Santa Barbara has not yet been fully worked out. Among the miscellaneous papers in the collection of his business correspondence, stored at the Herbarium of the University of California, is a typewritten list of plant names, con- tained on 114 sheets, with the following inscription pencilled on the first sheet: “List of Seeds and Plants Tried Out in Santa Barbara, California, by Dr. F. Franceschi.” The number of genera is approximately 796, of species, varieties, and horticul- tural forms, approximately 2,129. Exactly what significance may be attached to this list is a questionable matter, however. There is nothing to indicate who compiled it or when this was done. It appears to have been prepared without the exercise of very much care, and in no case is information given as to the results of the trial introductions. The total number of his new plant introductions is a debatable matter. In the catalogues of the Southern California Acclima- tizing Association which he published at irregular intervals, and later, in the price lists of the Montarioso Nursery, he points out the plants that he claims were first offered by him in the horticul- tural trade in this country. The total number of these—species, varieties, and horticultural forms—mentioned in his catalogues and price lists from 1896 to 1914 (see bibliography) reaches nearly 900. These claims, however, cannot all be accepted at their face value. There are occasional instances where species that he claimed to have introduced (“plants first offered by us in the United States’’) can be shown actually to have been offered by nurserymen prior to his coming to California. A single example is Phoeniz canariensis, the “‘Canary Island Date Palm.” Franceschi, in his “Condensed Catalogue and Price List’? for 1908, indicates that this species was one of his introductions, but, according to Mr. H. M. Butterfield this palm was offered by John Rock at San Jose as early as 1877. Moreover, the lack of consistency with re- gard to some of his claims, in catalogues of different years, casts a shadow of doubt on their validity. For instance, in his cata- logue for 1908, he claims to have been the first in this country to 26 MADRONO [Vol.7 offer the following species (among numerous others): Cinnamo- mum Camphora, Euphorbia pulcherrima, Leucadendron argenteum, Magnolia grandiflora, and Persea gratissima. In his catalogues for the years 1896, 1897, and 1900, however, he does not make this claim for any of these species, although they are all listed. How- ever, the number of errors of this sort is probably small, although a careful check of his catalogues against earlier horticultural literature is necessary before their extent can be ascertained. After making due allowance for these errors, the remainder is still truly remarkable. F. W. Popenoe, said of his work (Journal of Heredity, 13: 215, 1922) “His introductions are more numer- ous than those of any one man (in the United States), and many of them are now widely grown in the land of their adoption.” No evaluation of Dr. Franceschi’s contributions to horticulture in this country would be complete without a consideration of the influence of his personality on those of his contemporaries with whom he came in contact. His wealth of horticultural and botanical knowledge, gained from years of travel, observation, and practical work, coupled with his untiring interest in plants, were a constant source of inspiration to plant lovers wherever he went. The plants he introduced stand as living reminders of his un- tiring work, and his memory will long be perpetuated at Santa Barbara in his old home, “‘Montarioso,” which today is a city park, named in his honor, “Franceschi Park.”’ BIBLIOGRAPHY The following list, although not complete, probably contains a substantial part of the literary work of Franceschi. Bamboos. Gard. Chron. Ser. 2, 6: 773-774. 1876. Bamboos in California. Pac. Gard. July, August, September, 1908. Bauhinias. Pac. Gard. November, 1908. Behavior of alien plants at Santa Barbara. Ann. Rep. Am. Breeder’s Assoc. 6. 1911. . Colletia cruciata. Gard. Chron. Ser. 2, 9: 243, fig. 43. 1878. Dasylirion longifolium. Soc. Tose. Ort. Bull. 15: 112-113, pl. 6. 1890. Erythea armata. Gard. Chron. Ser. 3, 20: 424, fig. 74. 1896. Erythea edulis at home. Gard. Chron. Ser. 3, 13: 507-508. 1893. Erythea edulis at Santa Barbara. Gard. Chron. Ser. 3, 22: 157. 1897. Eugenias for California. Rural Californian. April, 1910. Euphorbia pulcherrima at Santa Barbara, California. Gard. Chron. Ser. 3, 21: 124, fig. 35. 1897. Fifteen years experience in southern California. Pac. Gard. December, 1908, January, February, March, 1909. Fioritura del Rhododendron longifolium. Soc. Tose. Ort. Bull. 8: 63-65. 1883. Frutti Tropicali e Semitropicali. Instituto Agricolo Coloniale Italiano, Florence. 1915. Introductions from Australasia. Pac. Gard. April, 1909. Japanese art of gardening. Pac. Gard. April, 1908. La Domenica delle Palme. Soc. Tose. Ort. Bull. 3: 116-121, fig. 12. 1878. Le Palmier de Vile de Guadalupe. Rev. Hort. 297-299. 1893. New and little known trees suitable for southern California avenues. Pomona Coll. Journ. Econ. Bot. 1: No. 4. December, 1911. 1943] REVIEWS 27 Notes on hardy bamboos. Gard. Chron. 1228-1229. 1872; Gard. 2: 316-318. 1872; (Ueber die harteren Bambus-Arten. Hamburg Gart. Blumenzeit 28: 417-423. 1872). Olea fragrans. Soc. Tose. Ort. Bull. 1: 273-275. 1876. Passato, presente e futuro della olivicultura in Tripolitania. L’Agricoltura Coloniale. 19: 201-204. 1925. Pittosporum. Pac. Gard. October, 1908. Pritchardia filamentosa Wendl. Soc. Tosc. Ort. Bull. 1: 116-118, fig. 1. 1876. Progressi di frutticultura a Tripoli nel 1923-24. L’Agricoltura Coloniale. 18: 309-313. 1924. Santa Barbara exotic flora. Santa Barbara. 1895. Trees from South Asia acclimatized in southern California. Forester. 4: 76-77, 129-130. 1898. Tropic and semitropic fruits for southern California. Pac. Gard. December, 1907, January, 1908. Weeds. Santa Barbara. Yucca angustifolia Pursh. Soc. Tose. Ort. Bull. 11: 10-11, pl. 1. 1886. Yucca filifera Chabaud. Soc. Tosc. Ort. Bull. 14: 278-280, pl. 9. 1889. CATALOGUES General descriptive catalogue. May, 1896, no. 4. General catalogue. May, 1897, no. 5. General catalogue. Easter, 1900, no. 6. Condensed catalogue and price list. 1908. Montarioso Nursery, descriptive price list. November, 1910, no. 2. Handbook and price list. 1911. Montarioso Nursery, descriptive price list. March, 1912, no. 3. Supplementary price list of the Montarioso Nursery. 1914. Department of Botany, University of California, Berkeley, May, 1941. REVIEWS Ceanothus. Part I. Ceanothus for Gardens, Parks and Roadsides, by Maunseryt Van Renssevarr. Part II. A Systematic Study of Ceanothus, by Howarp E. McMinn. Pp. xii+ 808. A Publication of the Santa Barbara Botanic Garden, Santa Barbara, California. Gillick Press, Berkeley, California. 1942. $2.50. If there is anyone who admits the place of decorative plants in the general scheme of things but thinks that botanists—above all taxonomists and their tools, herbaria and botanical gardens— have little or “nothing to do with the case,’ may he be given a copy of “‘Ceanothus”’! And may it be opened at the preface which, as G. B. S. long ago maintained is apt to be the most important part of a book. William Lassiter, Major General, United States Army, Retired, has written this one which not only reviews the work and outlines its purpose but places anyone who has anything to do with orna- mental plants in the mood to desire greater use, understanding and appreciation of them. The book is divided into two parts: the first by Van Rens- selaer is devoted to a consideration of the species, alphabetically arranged, known in cultivation; the second, by McMinn is a 28 MADRONO [Vol.7 detailed taxonomic account of the entire genus. The former is followed by an evidently carefully considered exposition of the propagation and cultivation of these shrubs prepared by A. J. Stewart, Horticulturist of the Santa Barbara Botanical Garden, and the latter by a chapter of twenty-odd pages, entitled “‘Dis- tributional History and Fossil Record of Ceanothus,’ by Herbert L. Mason. Even those primarily interested in the living plants will read Dr. Mason’s lucid account of the probable evolution of the group and its species with interest because he gives a glimpse of the relationship of old floras with those of today, and by his thoughtfully interpreted observations throws light on the prob- able development of a number of the forms treated with doubt as distinct entities. McMinn precedes his key to the fifty-five species and about half as many varieties (and the descriptive text) by a “General Discussion”; incidentally this includes the formal description of the genus. The discussion itself comprises about ten pages not counting some twenty more occupied by distributional maps, probable relationship charts, natural hybrid charts, chromosome “pictures” and a list of the specimens from which they were drawn (this commendable). There is also a plate depicting well the two sections into which the genus is readily divisible. The seven distributional maps are clear and conveniently show on each one several species and their variants. With the aid of these various charts the author explains, apparently with considerable confidence, his taxonomic conclusions and his reasons for arriv- ing at them. The reviewer has no knowledge of Ceanothus what- soever but anyone with general experience in classification will, it seems to him, have the impression from the author’s discussion, and that of Mason, that the taxonomy even with present knowl- edge could have been on a sounder basis. Perhaps a number of the entities recognized as species of equal merit could have been differently evaluated considering the fact that a simple geographi- cally limited, variable, often solitary character has frequently been accepted as “specific.” However the author believes that species can be proved by experiment. Granting that this is so, it is my impression that experimental taxonomic methods, so to speak, have more often proved than disproved the fundamental soundness of specific lines and apparent relationships as they were previously suggested by the more capable botanists of yesterday and entirely from morphological and geographical data. Mc- Minn’s interest ...I almost wrote enthusiasm ... in “‘ex- perimental methods” has, I cannot help but sense, inhibited his own expression of what he believes is actually happening (or has happened) in the history of these fascinating and plastic plants. This unfortunately (from the standpoint of practical simplicity) has resulted in his not anticipating facts (as he hints himself, for example page 191 and elsewhere) that the methods he advertises will probably at least in many cases be able to prove. In any 1943] REVIEWS 29 case his keys and remarks are clear, and others can follow his reasoning. His obvious knowledge of the group is little short of amazing even with due appreciation of his indebtedness to the work of previous students, notably Jepson and the latter’s pupil, J.T. Howell. To the former he pays the compliment of using his method of citation of references and specimens. In this connec- tion one may remark the excellent, distinctive typography. There is a lovely color plate of C. purpureus and innumerable photographs uniformly of exceptional beauty and value as well as a number of good drawings. Finally let us turn to the descriptive account of the seventy- odd species and a number of varieties distinguished by Van Rens- selaer in cultivation. He describes them in the idiom of the horticulturist, usually adding some remarks as to distribution, where cultivated and the growing conditions required. It would have been desirable if, besides the index to the entire book, page references after the descriptions had been given in each part to the other part. In this case some discrepancies in names used for the same plants would have been discovered. For example on page 14 we find the name C. austromontanus instead of C. foliosus in which it is included by McMinn, page 223. On page 30 the name C. exaltatus is given as “‘a new horticultural designation” while McMinn ignores it except as a variety of C. gloriosus, et cetera. Some of these slips, or they may be differences in opinion between the two authors, are going to confuse if not anger bibliographers, not to mention certain professional botanists who, of course, are almost God-like in the perfection of their own work! Above the rare mistakes, here is a living work, jointly con- ceived, jointly prepared, inspiring to everyone whether amateur or professional in the garden, herbarium or laboratory, and creating a closer bond of understanding, of friendship between all who have to do with plants, as Major General Lassiter has hap- pily phrased it. The contributors who made the book possible are to be congratulated and thanked for supporting so worthwhile a project that is destined to become a classic of its kind. Human- ity needs many similar books and from them will be born the realization that adequately financed herbaria and gardens must always be the basis for them.—J. Francis Macsrinz, Field Museum of Natural History, Chicago. Practical Plant Anatomy. By Avriance S. Foster. Pp. 1-155. D. Van Nostrand Company, Inc., New York. 1942. $2.50. This compact book of fourteen chapters or “Exercises” is spirally bound in flexible fabrikoid. Each exercise consists of a brief but accurate résumé of both early and recent papers dealing with the subject of the chapter, some discussion of the subject matter and different points of view thereon, and suggestions for study of selected materials and drawings to be made by the student. A short but well-chosen bibliography completes each 30 MADRONO [Vol. 7 exercise. The book is primarily a laboratory guide for a course in plant anatomy, although the discussions contain much factual material not usually included in a strictly laboratory guide. The suggestions for study, the material recommended, and the diagrams and drawings which the student is supposed to make, are all thoughtfully handled. The author has succeeded in reducing the number of drawings required to a minimum, a feature that will be welcomed by the immature students who want to confine their laboratory work to a definite, set period. This feature is not as reprehensible as it might seem to some pro- ponents of many drawings, for diagrams are substituted for the tedious, time-consuming detailed drawings to show relationships among various tissues. Of course, any student interested in mak- ing numerous detailed drawings of cellular types will find ade- quate hints throughout the text if he looks for them! The total absence of figures and illustrations mitigate against the use of the book as a guide to a study of plant anatomy by those who are unable to work under the direction of a trained instructor or where the library facilities do not afford extensive reference works. No attempt was made to give directions for the preparation of permanent microscope slides. But a far better point of view has been taken in that free-hand, temporary mounts of fresh materials are called for in nearly every exercise. No student using this method will get the idea that xylem is always stained red and that phloem and parenchyma cells are characterized by an affinity for a blue or green stain! Brief directions for macer- ating woody tissues and for the use of a few special reagents are included in the short appendix. The index is gratifyingly com- plete. The method of approach is analytical and classificatory rather than phylogenetic. The author’s reason for thus avoiding contro- versy is well stated in the following excerpt from Exercise IV (p. 89): “Since all methods for classifying plant tissues are open to objection, the writer has adopted a non-committal and ‘practi- cal’ attitude in this book. Instead of following any one scheme of classification, the emphasis is placed first of all upon the salient morphological features of the principal types of plant cells. These cell types recur in various regions, ‘tissues’ and organs of the higher plants, and a thorough knowledge of their form, structure, development, and presumable function(s ) must constitute the necessary analytical approach to anatomy.” The format and typography is good and errors are extremely few. The fabrikoid cover helps to protect the book from damage where liquids may be spilled on the laboratory table and the spiral binding permits the book to lie flat when opened to any page. The numerous references in the text and the bibliogra- phies following each exercise hold valuable keys to voluminous literature on plant anatomy.—Ira L. Wieains, Stanford University. 1943] REVIEWS 31 John Torrey, a Story of North American Botany. By ANDREW Denny Ropeers III. Pp. 1-352. Princeton University Press, 1942. $8.75. Mr. Rogers has included in one volume an amazing wealth of information on the history of botanical exploration in North America. He is to be congratulated on presenting such a com- prehensive picture of the life of John Torrey. Torrey, professor of chemistry at Columbia and Princeton, was by avocation a bota- nist. Through his pioneering efforts, the systematization of the flora of this continent was begun. He arranged and described the collections of numerous individuals and expeditions—Fre- mont, Emory, Owen, Whipple, Ives, the Mexican Boundary Sur- vey, the United States Exploring Expeditions, and many others. He was founder of the Torrey Herbarium, now housed at the New York Botanical Garden, and of the United States National Herbarium. His influence was extended by his association with the foremost botanists of his time, one of his first proteges being Asa Gray. The large amount of material from original sources which is made generally available for the first time in this book is intensely interesting as well as invaluable to a study of the his- tory of botanical exploration.—Mitprep Maruias, Department of Botany, University of California, Berkeley. Geographical Guide to the Floras of the World. Part I. By S.F. BuakE AND Atice C. Atwoop. United States Department of Agri- culture, Miscellaneous Publication 401, pp. 1-336. Washington, DAC.) June, 1942. $0.75. Here is the first part of a catalogue, unique in its field, that is destined to become one of the most useful books in the entire science of botany. Its value has been adequately demonstrated to the reviewer many times during the short time that the book has been in his possession. The work is a bibliographic catalogue listing in geographic order the floras and floristic accounts of the various geographic units of the world. In general, only complete works are included, but for little known regions collector’s lists are sometimes cited. Both general and local floras are included and most of the entries are annotated to indicate the content of the work. Part one deals with Africa, Australia, insular areas, North America and South America. To the authors are due the thanks of the entire botanical profession for performing so well a task that doubtless entailed much drudgery. It is to be hoped that this work may serve as a stimulus to inspire some bibliographically inclined individual to compile and publish a companion volume which will guide students to the literature dealing with taxonomic accounts, in whole or in part, of the families and genera of the flowering plants——Hersert L. Mason, 32 MADRONO [Vol. 7 Eriocaulaceae, Avicenniaceae, Verbenaceae. By Haroitp N. Mot- DENKE, in Flora of Texas, edited by C. L. Lundell. Vol. 3, pt. 1, pp. 1-87. University Press in Dallas. Southern Methodist Uni- versity. 1942. $1.50. This is the first part to be published of a proposed “ 10-volume work, each volume to contain approximately "700 pages. Volume 1 and 2 will contain the history of botanical exploration, the key to the families, a catalogue of all species, and maps showing distribution according to counties.”” The pro- posed work is unlike most modern floras in that each part will be a complete monographic unit in itself and will not, apparently, be arranged in taxonomic sequence by families. This first part by Dr. Harold N. Moldenke, of the New York Botanical Garden, is notable for the length of the generic and specific descriptions, complete synonomy, citation of numerous specimens, and extensive (probably complete) citation of refer- ences to publications where the species have been treated by earlier authors. The number of specimens cited for each family, together with an enumeration of the herbaria where they are deposited, is given in a footnote below each family description. The keys to genera and species are dichotomous and separate varieties as well as species. In most instances they use qualita- tive rather than quantitative characters, but sometimes make use of geographic ranges to supplement comparatively weak morpho- logical characters. Brief but interesting comments deal with vernacular names, geographic ranges, relationships, earlier misidentifications and uses by aborigines and the early settlers. The paper is of good quality and the printing clean-cut. Typo- graphical errors are few. Both the taxonomic and editorial work seem to have been done thoroughly, although Dr. Moldenke appears to draw specific lines pretty finely. The publication of this monograph of the three families as they are represented in Texas initiates an ambitious project which will be watched with interest by all plant taxonomists. Dr. Lun- dell is to be congratulated for undertaking to see through the tedious processes of writing and printing a work that will fill a long felt need—an adequate and exhaustive flora of the Lone Star State. It will be a source of great satisfaction to taxonomists if the high standards established in this initial publication hold throughout the entire work. It is the hope of the reviewer that Dr. Lundell may be able to secure the aid of enough collaborators to push the series through to completion within a reasonable length of time.—Ira L. Wicearns, Stanford University. MADRONO A West American Journal of Botany A quarterly journal devoted to important and stimulating articles dealing with plant morphology, physiology, taxonomy, and botanical history. These volumes should be a part of every botanist’s li- brary and should be made accessible to students of all universities and colleges. Volume I, 1916-1929... $5.00 Volume II, 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 Single numbers....... 0.75 The subscription price of MADRONO is $2.50 per year. We solicit your pat- ronage, Address all orders to: William Hiesey Bus. Mgr. Carnegie Institution of Washington Stanford University, California VOLUME VII NUMBER 2 MADRONO A WEST AMERICAN JOURNAL OF BOTANY APR 3 0 i843, ‘.. : NE = Contents —————— ay ges XEROPHYLLOUS SPECIES OF PHILADELPHUS IN SOUTHWESTERN NortTH _ AMERICA, ORE CON WERLECICOGIR iUi (a aa Gna tdcs ai ey tee eG ULI) Biya Ts Ado 35 i New Species oF PHACELIA FROM SALINE VALLEY, Ca tirornia, Lincoln OTT EAE) URLS RICOEUR AUD ES Oe a A 56 Giza mourririora Nurr. anp Irs Nearest Revatives, Thomas H. Kearney AGU EE EL SCE CCOLESIVON SA Uh sa CON ku eee cs ABE Pome pay Won Uh a enablers byaeNeey 59 Review: Viola Brainerd Baird, Wild Violets of North America (Ethel Crum), 46424. Cie RP IORI RANE UNS To Col oe A veg Eek CELL DR Sea SP 63 Achaea at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania April, 1943 MADRONO A WEST AMERICAN JOURNAL OF BOTANY Board of Editors Hersert L. Mason, University of California, Berkeley, Chairman. LeRoy Asrams, Stanford University, California. Epcar AnpeERsoN, Missouri Botanical Garden, St. Louis. Lyman Benson, University of Arizona, Tucson. Hersert F. Copetanp, Sacramento Junior College, Sacramento, California. Ivan M. JoHnston, Arnold Arboretum, Jamaica Plain, Massachusetts. Muprep EF. Marutias, University of California, Berkeley. Bassett Macume, Utah State Agricultural College, Logan. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board—ANNeEtTTA CARTER Department of Botany, University of California, Berkeley Business Manager—Wutiiam Hirsey North Queen Street and McGovern Avenue, Lancaster, Pennsylvania or Carnegie Institution of Washington Stanford University, California Entered as second-class matter October 1, 1935, at the post office at Lancaster, Pa., under the act of March 3, 1879. Established 1916. Published quarterly. Subscription Price $2.50 per year. Completed volumes I to V inclusive, $25.00; each volume $5.00; single numbers $0.75. Papers up to 15 or 20 pages are acceptable. Longer contributions may be accepted if the excess costs of printing and illustration are borne by the contributor. Range extensions and similar notes will be published in con- densed form with a suitable title under the general heading “Notes and News.” Articles may be submitted to any member of the editorial board. Manuscripts may be included in the forthcoming issue provided that the contributor pay the cost of the pages added to the issue to accommodate his article. Reprints of any article are furnished at a cost of 4 pages, 50 copies $4.10; 100 copies $4.50; additional 100’s $0.85; 8 pages, 50 copies $5.95; 100 copies $6.60; additional 100’s $1.30; 16 pages, 50 copies $8.35; 100 copies $9.35; additional 100’s $2.00. Covers, 50 for $2.75; additional covers at $1.65 per hundred. Reprints should be ordered when proofs are returned. | Published at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania, for the CALIFORNIA BOTANICAL SOCIETY, INC. President: H. S. Reed, University of California, Berkeley. First Vice- President: N. T. Mirov, California Forest and Range Experiment Station, Uni- versity of California, Berkeley. Second Vice-President: Jack Whitehead, Uni- versity of California Botanic Garden, Berkeley. Secretary: Clarence R. Quick, United States Department of Agriculture, 26 Giannini Hall, University of Cali- fornia, Berkeley. Treasurer: William Hiesey, Carnegie Institution of Wash- ington, Stanford University, California. Annual membership dues of the California Botanical Society are $2.50, $2.00 of which is for a year’s subscription to Madrofio. Dues should be remitted to the Treasurer. General correspondence and applications for membership should be addressed to the Secretary. 1943] MASON: ETHEL CRUM 33 ETHEL KATHERINE CRUM It is with regret that we announce the death of Miss Ethel Crum, Secretary to the Editorial Board of Madrofio and Assistant Curator of the Herbarium of the University of California. She passed away on January 5, 1943, at her family home in Lexing- ton, Illinois. Her botanical career began with the attainment of the degree of Master of Arts at the University of California in 1929 and her appointment as research assistant to Dr. W. L. Jep- son at the time that the manuscript for volume two of “A Flora of California” was in preparation. She assisted in the organiza- tion of the material for many of the larger genera in this volume but adopted as her own, the genus Potentilla, the manuscript of which was accepted with some revisions. Further evidence of her work may be noted in the Capparidaceae and in the Legu- minosae. In 1933 she accepted a position on the curatorial staff of the Herbarium of the University of California where she re- mained until her death. Here, in addition to her duties, she in- terested herself in the genera Monolopia and Pseudobahia and for the Editorial Board of Madrofio she assumed the burden of edit- ing and preparing manuscript for the printer. It is in this capac- ity that she made her most significant contribution to botany. Her early training in the classics, her command of the English language and her experience in handling manuscript combined to make her particularly suited to these tasks. Her ideas of editorial procedure were well conceived and fairly consistently adhered to. She believed that the first func- tion of an editor of a botanical journal is to protect the author from himself by preventing him where possible from making hasty and unconsidered statements, from being misinterpreted or misunderstood because of his lax use of grammar and from in- jecting personalities into an article to the point of creating offense. She edited to reduce the hazards of future embarrass- ments to the author. She had a positive concept of where the rights of the author leave off and the rights and duties of the editor begin. These rights fluctuated only with the age and experience of the author. She believed, for instance, that the text of a manuscript, while subject to such editing as is necessary to make it conform to the style of the journal and the standard of fitness set by its editors, is the concern of the author and when once the article is accepted by the editors any changes affecting the argument or conclusions may have the effect of impeaching the author and are wholly unjustified. On the other hand, the precise wording of the title of an article should be left to the editor subject to suggestion and approval by the author. This concern stemmed from experience in the use of bibliographies wherein a tremendous amount of time is wasted investigating the subject matter of articles whose titles ManroXo, Vol. 7, pp. 33-64. April 28, 1943. apR 29 ‘43 34 MADRONO [Vol. 7 are inadequate or misleading. She insisted that a title should be sufficiently specific to indicate not only the subject of an article but something of its scope. She insisted that it should be ex- pressed in language that can be readily translated into any foreign tongue and still convey its precise meaning. This entails the elimination from the title of all abbreviations and all forms of speech that are colloquialisms or professionalisms. As to the use of footnotes, it was her firm conviction that any- thing worth saying in a footnote was worth incorporating into the text. Footnotes, she maintained, interrupt the continuity of thought of the reader and in addition spoil the appearance of the printed page. They should, therefore, be avoided or reduced to emergency use. | } She believed that articles on isolated nomenclatorial problems were largely a waste of time and space and constantly urged the Editorial Board to adopt a rule refusing to print them unless they were written in conjunction with a monograph of the group con- cerned or were being used to urge the necessity of changes in or additions to the rules of botanical nomenclature. It was her further conviction that the available space in any journal is too precious in these days to waste it on preliminary taxonomic revi- sions and that no revision is worthy of publication until the author has completed his work upon it. She deplored the word “‘pre- liminary” because it is often taken as a promise of a more com- plete treatment to follow, whereas the author is content to allow his decisions to rest with the preliminary account. She argued that if the author intended that the account would be final then his use of the word “preliminary” in the title becomes a form of dishonesty and can only be construed as an excuse to cover mis- takes and decisions resulting from inadequate study. She continually urged upon young authors the avoidance of making in print promises as to the publication of future re- searches. Too often circumstances do not permit the fulfillment of such promises and the fact that they have been made may cause some investigator to make fruitless search of the literature. On behalf of the training of youthful and inexperienced authors she was a crusader; she would spare no effort to see to it that they were at least shown the paths of righteousness. Most of them responded in good taste and evidenced a sincere appreciation of her efforts to help them. A few have become soured on editors for life. Her work on Madrofio has set a standard of excellence in editorship the maintenance of which will serve as a challenge to her successors. She, perhaps more than any other person, has influenced its recent editorial policy.. Miss Crum was born in Lexington, Illinois, on March 18, 1886. She was educated at the University of Illinois and graduated with the class of 1907. Her education stressed English literature and 1943] HITCHCOCK: PHILADELPHUS 35 the classics. From 1909 until 1929 she taught in the public school systems of America, an experience of which she spoke as “traversing an intellectual desert of the most barren sort.” This may serve to explain better than more precise words her reasons for shifting her career to botany. During the years from 19382 to 1988 she served as Secretary of the California Botanical Soci- ety, relinquishing these duties to devote more time to her work as Secretary to the Editorial Board. Upon acquaintance one soon became impressed with the fact that Miss Crum was a woman of outstanding intellectual bril- liance, that she had an engaging personality and a ready flow of wit and humor. She was devoted to her work and her several hobbies crowded one another for her attention. Her passing ends a career of uncommon usefulness; her life was a milestone in the history of a journal HersBert L. Mason. THE XEROPHYLLOUS SPECIES OF PHILADELPHUS IN SOUTHWESTERN NORTH AMERICA C. Leo HircucocK There are, in southwestern United States and northern Mexico, several species of Philadelphus which are xerophytic. These plants have a rather heavy indumentum on the lower sur- faces of the leaves as well as on the calyces and on the epidermis of the twigs. Their leaves are small (1-3 cm. long), rather thick and leathery, and entire. Whereas the flowers of Philadelphus are usually borne in cymes or panicles of from three to many blos- soms, the flowers of these members of the genus usually occur singly (rarely in two’s or three’s) at the ends of short leafy lat- eral branches. They have been placed in the group Microphylli by Rydberg (No. Amer. Fl. 22: 168. 1905) with no indication whether this group is of sectional or subgeneric rank. Since the precise taxonomic status of this and corresponding groups of the genus is not pertinent to this paper, the term “group” which was frequently used by Rydberg will be used. Floristically, these small-leaved species are particularly inter- esting, since they occur chiefly in the lower levels of the larger mountain ranges from Texas to California. Since localized populations of each of the species are isolated by intervening deserts, geographical races have become differentiated from one another. As a unit they are readily distinguished from all other North American species of Philadelphus, yet it is quite apparent that they have been derived from, and are very closely related to, certain species of the Mewicant (Rydb. op. cit.). A general idea of the relationship of these sections as well as the geographic distribution of the Microphylli is expressed in the accompanying diagram (fig. 1). Of the various species in the Meaicani, Philadelphus affinis 36 MADRONO [Vol. 7 Se =— ae SONORA 1. RAFFINIS 2. MEXICANUS 3, COULTERI 4, PURPUS!} 5. MICROPHYLLUS SuBsP ARGENTEUS MACULATUS CRINITUS STRAMINEUS F. ZIONENS!IS PUMILUS ARGYROCALY X 5 TYPICUS oe 3 : OCCIDENTALIS oe 6 SERPYLLIFOLIUS 7, MEARNSII A. Sv8SP BIFIDUS LO MNMOODS Fic. 1. Phylogeny and distribution of Philadelphus (Microphylli). Schlect., a species of central Mexico, has leaves that are large, thin, and denticulate, 6-10 cm. long, and rather sparsely pubes- cent. Its petals are 1.5—2 cm. long, its styles are densely hairy, 1943 | HITCHCOCK: PHILADELPHUS 37 and its inflorescence is a five- to eleven-flowered cyme. In gen- eral appearance it is very suggestive of the species of Californici, Coronarii, and Grandiflori. Another species of the same group, P. mexicanus, is quite similar to P. affinis in most particulars, but the plant is more pubescent and the inflorescence is greatly re- duced, usually consisting of one, or of three, flowers at the ends of short lateral branches. A third species, P. Coulteri, has much smaller leaves (3-6 cm. long) which are more leathery and, although denticulate, are very densely hairy; the petals are con- siderably smaller (12-17 mm. long) and the inflorescence is usually one-(three-) flowered. The styles are densely hairy as is the upper part of the ovary. Philadelphus Purpusi (Microphylli) although known from only one collection, is a well-marked species that is apparently closely related to P. Coulteri, having leathery, hairy leaves, and flowers that are quite similar to those of its supposed relative, but differ- ing in that it has entire, small (2—3.5 cm. long) leaves, petals that are but 10-15 mm. long, and sparsely hairy styles. Philadelphus microphyllus, with its various subspecies, has apparently been de- rived from P. Purpusii, probably through the subspecies argyro- calyx, argenteus, and crinitus, all of which have the pubescence and leaves characteristic of P. Purpusi. In the first two of these sub- species there is even a trace of the pubescence of the styles and ovary that characterizes P. Purpusw and P. Coulteri. The relationship of P. serpyllifolius and P. Mearnsii to the other species of Microphylli is not so apparent, but it is possible that they have been derived from P. microphyllus. From the map it can be seen that the two subspecies of Phila- delphus Mearnsit apparently are widely separated geographically. It seems probable, however, that more thorough collecting in northern Mexico will disclose that the plants are much more widely distributed than extant collections indicate. The typical form of the species was collected in 1892 and it was not until 1924 that the next collection was made, and that in a region remote from the type locality. Only two collections have been seen that were made since 1924, yet it seems certain that the plant is to be found in other parts of New Mexico, Texas, or Mexico. The presence of what I consider to be a local population of P. microphyllus subsp. typicus in the Chisos Mountains of Texas is dificult to explain. As mentioned under the treatment of that subspecies, however, it is strongly suspected that these plants are really genetically different from the plants in central New Mexico, but aside from the fact that they are much greener in aspect, distinctive taxonomic characters are not discernible from the herbarium material examined. Until the treatment of the genus by Rydberg in 1905, but few specific names had been proposed in the group. Rydberg not only recognized five species that had been described previously 38 MADRONO [Vol. 7 but proposed seven additional specific entities. Since his work one or two additional names have been proposed. It was with the hope of evaluating the validity of these various species that this study was undertaken. Herbarium material has been made available to me from the following institutions, their abbreviations, as I have used them in citing material, being shown in parentheses. University of Cali- fornia (C), Gray Herbarium (G), New York Botanical Garden (NY), Pomona College (P), Stanford, Dudley Herbarium (S), United States National Museum (US), University of Washington (Wash), State College of Washington, Pullman (WSC). To the curators of these herbaria I express my sincere appreciation. Key To SPECIES A Lower surfaces of leaves grayish with short matted tomentum and (usually) longer straight ap- pressed hairs; stamens mostly 28 (26-32 or more ?)s ‘stylesless tham 1 mm. long? > 2... ee 3. P. serpyllifolius AA Lower surfaces of leaves green or grayish but never with tangled matted tomentum; stamens and styles various. B Base of styles and adjacent ovary densely pilose; leaves coarsely strigose on both surfaces; CalyCeSceray 2 oo nec ee ahr eee eee 4. P. Purpusii BB Base of styles and adjacent ovary glabrous or (very rarely) with few hairs; leaves and calyces various. C Stamens 16-24, filaments distinct; styles less than 1 mm. long; pubescence of leaves very coarse, the hairs thick in cross SECTION \ pinch Se aunt ee eae ee 2. P. Mearnsii CC Stamens more than 25, filaments often united at the bases; styles usually at least 1 mm. long; pubescence of leaves of slender DAIS. 4-2 soties til cae as eee ee eee 1. P. microphyllus 1. PHILADELPHUS MICROPHYLLUs Gray, Mem. Am. Acad. Sci. I, 4:54. 1849. Much branched, rounded shrub 1-2 m. tall; young branches densely pubescent with appressed hairs, often silvery, bark more or less reddish-brown to tan, usually exfoliating the second year, older branches grayish to grayish-brown; petioles 1-3 mm. long, leaf-blades ovate, ovate-lanceolate, or lanceolate, to lanceolate- elliptic, 8-35 mm. long, 83-15 mm. broad, entire, often slightly revolute, the apices rounded to acute, 3-nerved from base, pubes- cent with hairs that are minutely papillate their entire length, the upper surfaces either strigose with closely appressed hairs or hir- sute with short erect hairs, but sometimes both strigose and hir- sute, rarely glabrate, lower surfaces usually more densely hairy, the hairs either fairly long and partially appressed, or shorter, straight, and closely appressed; flowers borne singly or in threes (twos) on pedicels 0.5—3 mm. long, at ends of short leafy shoots; 1943] HITCHCOCK: PHILADELPHUS 39 calyx-tube 2—3 mm. long in flower, enlarging to 3—5 mm. in fruit, sparsely to densely strigose (glabrate), less commonly quite sil- very with dense indumentum of tangled somewhat curled hairs, lobes 3-5 mm. long, acute to acuminate, outer surface glabrous to pubescent like the tube, inner surface always lanate; petals white to cream, 6-17 mm. long, usually rounded but often emarginate ; stamens (30) 32 to about 70, the filaments usually partially united at least at base, but sometimes entirely free; styles (0.5) 1-2 mm. long, free above, or more commonly, united to tips; stigmas 1.5— 3 mm. long, usually partially united. Philadelphus microphyllus has been treated taxonomically in various ways. Many authors, like Rydberg, considered the plant as it occurs in central New Mexico to be specifically distinct from the many closely related forms of California, Nevada, Utah, and Arizona. It is true that it is largely a matter of interpretation whether these other entities be considered varieties, subspecies, or species, but certain characteristics upon which Rydberg based his species appear to me to be but normal variations largely ac- countable for by habitat. Still other differences which were used to characterize some of these segregates are not at all constant even though they may be genetic in origin, so that the mainte- nance of some of Rydberg’s species, even as subspecific entities, does not seem feasible. One such characteristic is that of the degree of union of styles and stigmas. Although there is some variation in the length of the styles and in the degree to which they are united, I can find no consistent variation which can be correlated with other distinctive morphological characters or with geographical distribution. In some cases the styles of dif- ferent flowers on the same specimen may be “united” and “‘par- tially free’ (Hitchcock et al. 4148, C). Although Rydberg listed definite and distinctive numbers of stamens for several of his closely related species (P. stramineus 30- 40, P. microphyllus ca. 40, P. minutus ca. 60, P. pumilus ca. 30, and P. argyrocalyz, P. ellipticus and P. occidentalis “‘many’’), the sta- mens are so inconstant in number that species cannot be distin- guished in this manner. The normal number of stamens is seem- ingly a multiple of four, P. stramineus, for example, may have any- where from thirty to forty, but usually has thirty-two, whereas P. occidentalis, P. argenteus, and P. argyrocalyx usually have from forty to sixty. There is often some variation in the stamens of flowers of the same branch, so that neither P. pumilus nor P. stra- mineus can be maintained on the basis of this character alone. Leaf size, too, varies greatly, but the variation appears to be due chiefly to the age of the branch on which the leaves are pro- duced or to the amount of shade in which the plant grows. As would be expected, plants which have grown in partial or com- plete shade have large leaves and rather sparse pubescence. The variation in leaf size due to branch age is best shown, perhaps, on 40 MADRONO [Vol. 7 a plant collected by Rydberg and Garrett (no. 9608). The leaves of a young shoot are well over 3 cm. in length, those of an older shoot are scarcely 1.5 cm. long. Ina Heller and Heller collection (no. 3792) nearly all plants have leaves from 2.2 cm. to 8.5 cm. in length; however, the sheet at Stanford bears a plant with leaves less than 2 cm. long. Another collection (Harrison 6604) made from a plant with leaves 6 to 8 mm. long had a two to three year old branch with leaves 9 mm. long. The three twigs of this branch bore new growth with leaves all 12 to 19 mm. long. I can see no significant variations in the color, pubescence, or degree of exfoliation of the bark. It seems, therefore, that the characters that are dependable for subspecific delimitation are the following, listed in order of significance: type and amount of pubescence, number of stamens, leaf size, and flower size. Key To SUBSPECIFIC VARIATIONS OF P. MicropHyLLus A Petals with large purple spot at base ......... lb. P. microphyllus subsp. maculatus AA Petals not spotted with purple. B Calyx-tube silvery, completely covered with long hairs. C Pubescence of calyx matted, consisting of long straight hairs mixed with more slender intertwined hairs; leaves rather densely pubescent, the hairs of lower surfaces rather long and often not appressed, the upper surfaces much less pubes- cent and usually greenish ........ 1d. P. microphyllus subsp. argyrocalyx CC Pubescence of calyx usually not mat- ted, but if so, the hairs all of same type, often the pubescence too sparse to be tangled. D Calyx shaggy with long matted slender hairs; both surfaces of leaves grayish with long slender hairs, these sometimes somewhat matted on the lower SUPPACES UI. taco e ene ee ne Ic. P. microphyllus subsp. crinitus DD Calyx with appressed pubescence chiefly, the hairs not tangled; leaves mostly distinctly green- ish above, if grayish the hairs not at all matted. E Upper surfaces of leaves with hirsute pubescence, the hairs erect or nearly so. F Stamens usually 40 or more; leaves mostly over 15 mm. long, the pubescence of lower surfaces not tightly appressed ".......... ld. P. microphyllus subsp. argyrocalyx 1943 | HITCHCOCK: PHILADELPHUS Al FF Stamens usually 32; leaves often less than 15 mm. long, the pubes- cence of lower sur- faces rather tightly appressed. G Hairs of upper sur- faces of leaves all essentially of the same length .... GG Hairs of upper sur- faces of leaves mainly short and erect, the remain- der less numer- ous, longer and more appressed . EE Upper surfaces of leaves with strigose pubescence, the hairs appressed. H Stamens usually 40 or more, some usually united half their length; leaves mostly over 15 mm. long .. HH Stamens usually 32, free or united only at base; leaves seldom over 15 mm. long ... BB Calyx-tube not silvery, or at least not com- pletely covered by pubescence. I Leaves hirsute above, the hairs erect or nearly so. J Leaves usually 8-12 mm. long, the pubescence of lower surfaces of appressed hairs, that. of the upper surfaces consisting of many short erect hairs mixed with less numerous, longer, sub- appressed thairs’ <.4 9020: 4045 254 JJ Leaves 8-30 mm. long, the hairs of lower surfaces not appressed, or those of the upper surfaces all of one type. K Stamens usually 40 or more; leaves commonly at least 15 mm. long, the hairs of lower surfaces not always tightly appressed ........ KK Stamens usually about 32 (to 40); leaves 8-15 mm. long, the hairs of lower surfaces APPLESSEG. che cea. eo eee If. ig: la P. microphyllus subsp. stramineus forma zionensis . P. microphyllus subsp. pumilus . P. microphyllus subsp. argenteus le. P. microphyllus lg Id subsp. stramineus . P. microphyllus subsp. pumilus . P. microphyllus subsp. argyrocalyx le. P. microphyllus subsp. stramineus 42 MADRONO [Vol. 7 II Leaves not hirsute above, the hairs appressed. L Petals 11-17 mm. long; hypanthium with but little pubescence, this ; chiefly at the angles ......... lh. P. microphyllus subsp. typicus LL Petals seldom over 10 mm. long; hypanthium various but often quite densely pubescent. M Filaments usually united at least one-third their length; calyx-tube - sil- very; leaves mostly 20-30 mm. long, not hirsute ... la. P. microphyllus subsp. argenteus MM Filaments free or united at base only; calyx-tube not silvery, or if so, leaves usually less than 20 mm. long. N Stamens usually about 32; entire outer sur- face of calyx quite densely pubescent with stiff straight ap- pressed hairs, the outer surface almost completely covered by shairs ee le. P. microphyllus subsp. stramineus NN Stamens usually 40 or more; the outer sur- face of calyx sparsely pubescent or _ even partially glabrous, in any case not hidden by JNaibS eo. ar eee li. P. microphyllus subsp. occidentalis la. PHILADELPHUS MICROPHYLLUS subsp. argenteus (Rydb.) comb. nov. P. argenteus Rydb. N- Am. Bl 22: 17109 1905" ak: argyrocalyx var. argenteus (Rydb.) Engl., Engl. & Prantl, Nat. Pflanzenf. II, 18a: 193. 1930. P. microphyllus var. argenteus (Rydb.) Kearney & Peebles, Journ. Wash. Acad. Sci. 29: 480. 1939. P. Palmert Rydb. N. Am. FI. 22: 178. . 1905; type: sierra Madre, 40 miles south of Saltillo, Mexico, Palmer 2122. P. ma- drensis Hemsl. Kew. Bull. 251. 1908; type: Sierra Madre, Durango, Mexico, Seemann 2167. Leaves mostly 20-35 mm. long, ovate-lanceolate or lanceolate, pubescence of lower surfaces grayish-strigose, upper surfaces sometimes glabrate; calyx more or less canescent with long straight appressed hairs which are often mixed with shorter more slender ones; petals 8-11 (12) mm. long, often pubescent on outer surfaces near the base; stamens 40 or more, filaments usually united at least at the base and often united to near the 1943] HITCHCOCK: PHILADELPHUS 43 tip, usually grouped in irregular phalanges; styles sometimes sparsely hairy near base. Type. Fort Huachuca, Arizona, Dr. T. E. Wilcox in 18938. Range. Mountains of southeastern Arizona, south into So- nora, Chihuahua, Coahuila, and Durango, Mexico. Material seen. Arizona. Coronado Mts., Goldman 2379 (US); Santa Rita Forest Reserve, Griffiths 4194 (US). Santa Cruz County: near Patagonia, Harrison 7180 (C, NY). Cochise County: Bisbee, Carlson in 1915 (US); Fort Huachuca, Wilcox in 1898, type collection (NY), Wilcoz in 1873 (NY), probably should read 1893, surely part of type collection; Wilcox in 1892 (NY), and Wilcor 2384 (US); Huachuca Mts., Ramsey Canyon, Jones 24918 (P) and Millers Canyon, Goodding 147 (G, NY, US) ; between Fort Huachuca and San Pedro River, Mearns 1540 (US); Chiricahua Mts., Barfoot Park, Blumer 1303 (G, NY, S, US); Barfoot Fire Station, Eggleston 10766 and 10794 (US); Upper Pine Canyon, Burrall in 1906 (US) ; Dos Cabezas, Lemmon in 1881 (C). Graham County: Mt. Graham, Peebles, Harrison, & Kearney 4449 and 4450 (US). Pima County: Santa Catalina Mts., Lemmon & Lemmon in 1881 (C); Santa Rita Mts., Pringle in 1881 (G, NY, P, US) and Peebles & Harrison 3000 (US) ; Rincon Mts., Spud Range, Blumer 3565 (C). Mexico. Cuinuanua: Guayanopa Canyon, Sierra Madre Mts., Jones in 1903 (P); San Luis Mts., Goldman 1432 (US). Sonora: San Jose Mts., Mearns 1617 (US). Coauuiza: Sierra Madre, 40 miles south of Saltillo, Palmer 2122, type collection P. Palmeri (G); Sierra de Parras, Purpus 4592 (G, US). Philadelphus microphyllus subsp. argenteus approaches subsp. stramineus and subsp. pumilus in the pubescence of the calyx, but is larger-leaved than either and the pubescence of the leaves is all appressed. It is also quite similar to subsp. argyrocalyx but dif- _ fers because the leaves are not hirsute, their lower surfaces have more appressed pubescence, and the pubescence of the calyx is more uniform. There is scarcely any difference between the type of P. Palmeri and the type of subsp. argenteus except that the filaments are more completely united in the former. Pubescence, leaf size and shape, and flower size are alike in the two. Rydberg surely had seen only the type specimen when he described P. Palmeri, but the Purpus collection, made in 1910, is annotated by Rydberg as P. Palmeri and shows that he must have had to change his concept of the species considerably, since the flowers of that collection have petioles that are but 1.5 mm. long (original description of P. Palmeri specified 1.5 cm., although this is probably a typo- graphical error), petals that are 9-12 mm. long (12 in original description), styles that are not distinct, and a hypanthium that is as long as it is in most of the other subspecies of the complex. The only possible basis for maintaining P. Palmeri even as a sub- AA MADRONO [Vol. 7 species would be the union of the filaments. However, much of the material from Arizona that is surely P. argenteus has filaments as completely united as are the filaments of the flowers of the two collections of P. Palmeri (Pringle in 1881, Santa Rita Mts., and Blumer 1303). Philadelphus madrensis is being reduced to synonymy even though the type (Seemann 2167, Sierra Madre, Durango) has not been seen. That plant, it will be noted, was collected very near the type locality of P. Palmeri, and the original description of P. madrensis fits P. Palmeri quite well. The ‘1—3-flowered inflores- cence, and subsessile, acute, entire, 8-nerved, sericeous-pilose leaves” are similar to those of the type of the latter species and therefore to those of subsp. argenteus. There is nothing distinc- tive about the flowers, the stamens being more or less connate into four phalanges and the ovary either glabrous or slightly pilose on top. In fact one could scarcely compose a description that would apply more closely to Philadelphus microphyllus subsp. argenteus than does the description of P. madrensis except that the stamens are said to be about thirty-two, whereas in the specimens of P. argenteus I have seen there have been forty or more. This dis- crepancy notwithstanding, it seems safe to assume that P. madrensis is identical with P. Palmeri and therefore with subsp. argenteus. There is the possibility that this is the same plant as P. asperi- folius (Koern. Gartenflora 16: 78. 1867) but in the absence of proof of this identity, that name is not being used. Rydberg’s separation of P. asperifolius on the basis “‘bark of previous year’s growth not exfoliating” is inconclusive. 1b. PHILADELPHUS MICROPHYLLUs subsp. maculatus subsp. nov. Foliis lanceolatis vel ellipticis-lanceolatis, 1.5—2.5 cm. longis, ca. 4 (3-8) mm. latis, viridis, appresso-pubescentibus; calicibus purpureis, cum pilis adpressis tenuibusque pubescentibus, cinera- ceis, tubo ca. 3.5 mm. longo, lobis 4-5 mm. longis, abrupto-acumi- natis; petalis oblongo-ovatis, ca. 12 mm. longis, albis, petalo inferiore purpureo-maculato; staminibus 32-40, plus minusve in phalangibus ex 2-8 filamentis conjunctis, interdum ad proximum apicem connatis; stigmatibus ca. 8 mm. longis, saepe absoluto- connatis; stylis ca. 1.5 mm. longis, ad basum sparse pilosis. Leaves lanceolate to elliptic-lanceolate, 1.5—2.5 cm. long, ca. 4 (38-8) mm. broad, both surfaces greenish, the pubescence ap- pressed; calyces purple, finely appressed pubescent and rather cineraceous, the tube ca. 3.5 mm. long, the lobes 4-5 mm. long, abruptly acuminate; petals oblong-ovate, ca. 12 mm. long, white with distinct purplish blotches at base, sometimes with this purple spot extending to upper half or even to tip of petal; stamens 32— 40, irregularly united into phalanges of 2—7 or 8 filaments, these sometimes connate to near tip; stigmas ca. 3 mm. long, usually 1943] HITCHCOCK: PHILADELPHUS 45 completely united, styles ca. 1-5 mm. long, sparsely strigose at base. Type. In forest of large pines, forest floor of low vegetation, mountain top, 7 kilometers southwest of Miquihuana, Tamaulipas, Mexico, lat. 23° 42’ N., long. 99° 45’ W., elev. 3430 m., August 5, 1941, Stanford, Retherford, and Northcraft 690 (Wash). This showy-flowered form of Philadelphus microphyllus is closely related to subsp. argenteus, but differs not only in the purple-spotted petals, but also in the narrower, greenish leaves. This greenness may be due, in part at least, to the fact that the plants were probably partially shaded. It is known only from the type collection but the collectors re- port that it was fairly abundant, that all plants seen had the spotted petals, and that their collection, of about ten herbarium sheets, was made from about four different shrubs. lc. PHILADELPHUS MICROPHYLLUS Gray subsp. crinitus subsp. nov. Folia 15-25 mm. longa, supra alba, adpressa-strigosa, subtus simplicatio-strigosa, calicibus extra albis implicatis-strigosis ; petalis ca. 11 mm. longis; staminibus 48-68, filamentis basi con- natis; stylis 1-1.5 mm. longis, connatis; stigmatibus connatis. Leaves mostly 15-25 mm. long, pubescence of the upper sur- faces grayish, strigose, the hairs appressed, pubescence of the lower surfaces densely long strigose, the hairs often somewhat tangled; calyces gray with long more or less matted slender hairs; petals ca. 11 mm. long; stamens 48-68, united at base; styles 1-1.5 mm. long, united; stigmas 3—4 mm. long, united. Type. Rocky ground near top of Mount Livermore, Davis Mountains, Jeff Davis County, Texas, June 4, 1928, E. J. Palmer 34347 (NY). Other material seen. Texas. North side Mount Livermore, Palmer 34364 (NY); Livermore Peak, Ferris & Duncan 2529 (NY, S The status of this entity is puzzling. By some workers it has been mistaken for P. serpyllifolius, but it differs so markedly from that species in pubescence, stamen number, style and stigma length, and general appearance, that a close relationship between the two seems doubtful. It is probable that it is much more closely related to P. argyrocalyz or even to P. argenteus, both of which it strongly resembles in general aspect, but the pubescence of the calyx and leaves is so unique that it cannot be combined with either. Since the flowers are so like those of the other sub- species of P. microphyllus there is not an adequate basis for ac- cording the plant specific recognition. 1d. PHILADELPHUS MICROPHYLLUs subsp. argyrocalyx (Wooton) comb. nov. P. argyrocalyx Woot. Bull. Torrey Bot. Club 25: 452. 1898. P. serpyllifolius var., Gray, Pl. Wright. 2: 64. 1852. P. ellipticus Rydb. N. Am. Fl. 22: 172. 1905; type: Mesilla Park, 46 MADRONO [Vol. 7 New Mexico, Tinsley in 1896, probably wrong locality (Woot. & Standl., Contr. U.S. Nat. Herb. 19: 300. 1915). Much like subsp. argenteus in general appearance but leaves hirsute on upper surfaces and usually with less appressed, but somewhat longer, looser, hairs on lower surfaces; calyces typi- cally grayish with matted mixture of long straight and slender curled hairs, but pubescence sometimes less abundant and calyces then scarcely grayish; petals 8-11 (17) mm. long, often pubes- cent near base on the outer surfaces; stamens usually 40 or more, filaments commonly partially united, often united to near tips; styles united, sometimes sparsely hairy near base; stigmas at least partially united. Type. Eagle Creek, White Mountains, Lincoln County, New Mexico, Wooton 524. Range. Mountains of south central and western New Mexico into southeastern Arizona, south into Chihuahua, Mexico. Material seen. New Mexico. Lincoln County: White Mts., Wooton 524, type collection (NY, US) and Wooton in 1901 and 1905 (US) ; West Peak, El Capitan Mts., Earle & Earle 220 (NY); Mesilla Park, Tinsley in 1893, type collection P. ellipticus (NY, US) ; Ruidoso Creek, Wooton in 1895 (US). Otero County: head of Rio Fresnal, Barlow in 1911 (C); Cloudcroft, Sacramento Mts., Rehder 372 (US), Eggleston 14541 (US), Wooton in 1899 (US), and Orcutt 1351 (US). Without definite locality, Wright 1101, the plant which Gray called P. serpyllifolius var., therefore from “mountain sides at the Copper Mines” (G, NY, US). When Wooton described P. argyrocalyx he included therein the Wilcox collection from Arizona. That plant was later made the type of Rydberg’s P. argenteus. The two entities are similar and undoubtedly they are closely related; in many areas they grade into one another. Yet subsp. argyrocalyx, as it occurs in Lincoln County, New Mexico, is very unlike any collections of Philadelphus from Arizona, the pubescence of the calyx being much more curled and matted, consisting not only of matted curled trichomes but also of long straight hairs. The leaves are conspicuously hirsute with short erect hairs, a condition which is characteristic of nearly all material from southwestern New Mexico but not of the plants from Arizona. However, material becomes more and more like subsp. argenteus near the Arizona border until, in the mountains of extreme southeastern Arizona, the material is intermediate in nature between the two subspecies. The following collections therefore are not quite identical with the type collection and some are at least as similar to subsp. argenteus as to subsp. argyrocalyz. All have hirsute leaves. NEw Mexico. Sierra County: Mimbres Mts., Hillsboro Peak, Diehl 626 (P). Soccoro County: San Mateo Peak, Goldman 1745 (US). Grant County: Burro Mts., Metcalfe 173" (€, G, NY, P)s, US), quite typical as to leaves, the calyx almost the same as in material 1943] HITCHCOCK: PHILADELPHUS AT from the type locality, the petals pubescent; Animas Peak, Gold- man 1870 (US). Valencia County: mountains west of Grants Sta- tion, Wooton 1109 (US). Arizona. Oak Creek, Goldman 2174 (US). Pinal County: Superstition Mts., Harrison 6604 (NY). Cochise County: Chiricahua Mts., Price in 1894 (S) and Jones 283827 (C, P), Barfoot Park, Blumer 1291 (G, NY, S), very close to P. argenteus, but calyx very sparsely hairy, the leaves hirsute, however; White River Canyon, Chiricahua Mts., Toumey in 1894 (C); Upper Pine Canyon, Burrall in 1906 (US). Pima County: Mt. Lemmon, Santa Catalina Mts., Livingston & Thornber in 1906 (NY); Santa Catalina Mts., Lemmon & Lemmon 186 (C) and Lem- mon 170 (G); near Mud Springs, Harris C16344 (US). Graham County: Pinaleno Mts., Bonita, Munz 1248 (P). Mexico. Cui- HUAHUA: Colonia Juarez, Sierra Madre Mts., Jones in 1903 (NY, Bs US). Other collections are more intermediate with subsp. typicus: Hillsboro Peak, Sierra County, New Mexico, Metcalfe 1323 (C, G, NY,P,US). The plants of this collection have silvery leaves and united stamens (characters of subsp. argyrocalyx but the calyces, as in subsp. typicus, are nearly glabrous. le. PHILADELPHUS MICROPHYLLUS Gray subsp. stramineus (Rydb.) comb. nov. P. stramineus Rydb. N. Am. Fl. 22: 172. 1905. Much the same as subsp. occidentalis, but the calyces usually canescent with stiff straight appressed hairs; leaves mostly 10-25 (probably averaging about 15) mm. long, pubescence of both surfaces usually appressed, if hairs more erect, then all approxi- mately the same length; petals 7-10 mm. long, their outer sur- faces sometimes pubescent at base; stamens 30—50 but usually 32, the filaments from scarcely united to united to near the tips. Type. White Mountains, Mono County, California, Shockley in 1888. Range. Foothills and lower levels of the Charleston Moun- tains of Nevada, White Mountains of California, and San Pedro -Martir Mountains of Lower California. Probably more wide- spread than present collections indicate. Although it is believed that subsp. stramineus is a valid entity, if it is to be maintained it must be on a different basis than that used by Rydberg. It was keyed by him on the following charac- ters: “Leaf blades equally grayish on both surfaces; sepals acute; bark of the old stems straw colored,’ whereas P. microphyllus, in contrast, had “leaf blades paler beneath; sepals acuminate; bark of the old stems gray.” The leaf-blades of the Shockley collec- tion are paler on the lower surfaces than on the upper; the sepals are just as sharp-pointed as are those of much of the material of P. microphyllus subsp. typicus; lastly, although it is true that the older bark of the type collection of P. stramineus is not grayish, in other collections of that entity from the White Mountains (Cassel 48 MADRONO [Vol. 7 262), the bark is as weathered and gray as that of any material from New Mexico. The unique characteristics of the entity are as follows: the calyces of subsp. stramineus are grayish with dense stiff straight hairs, in this respect being more similar to subsp. argenteus than to either subsp. occidentalis or subsp. typicus. Usually there are fewer stamens (thirty-two being the commonest number) per flower than in subsp. argenteus, subsp. occidentalis, or subsp. typicus, but since there are occasional plants with as many as forty stamens per flower (Percy Train 2178, from the Charleston Moun- tains), too much importance should not be ascribed to this pecu- liarity of subsp. stramineus. Material seen. Nevapa. Clark County: Charleston Mts., Clokey & Clokey 7133 (NY,S, Wash), Clokey 5458 (NY,S, Wash), Clokey 5490 (NY,S, Wash), Heller 10998 (C, NY, S, US), Clokey 70940 (G), Percy Train 2178 (NY), Jones in 1927 (P), and Jaeger in 1925 (P). Caxirornia. Inyo County: White Mts., Black Can- yon, Cassel 262 (S) and Duran 540 (C, NY, P, S, US), Wyman Canyon, Ferris 6966 (S); Crag Canyon, Grapevine Mts., Gilman 3262 and 3259 (P). Mono County: White Mts., Shockley in 1888 (NY, type) and Shockley 454, probably type collection also (C, G, S, US). Mexico. Lower Catirornia. San Pedro Martir Mts., Vallecitos, Goldman 1223 (US); San Pedro Martir, Brande- gee in 1898 (C) and Heller in 1902 (C); La Encantada, Wiggins & Demaree 4947 (C, G, NY, P,S, US) ; east of La Encantada, Melinge in 1981 (P,S, US). Some of these collections from Lower Cali- fornia approach subsp. pumilus in that the upper surfaces of the leaves tend to be hirsute with short erect hairs; further collections may prove that they more properly belong with that entity. Material from Zion National Park is unique in that the pubes- cence of the upper surfaces of the leaves is short hirsute rather than strigose. In this respect the plants resemble subsp. pumilus, but that entity has long, appressed hairs mingled with the short erect ones, whereas the material from Zion Canyon is hirsute with hairs of uniform length. Also, the leaves of the plants from Cali- fornia are smaller. This variant from Utah is therefore accorded minor status as follows. 1f. PHILADELPHUS MICROPHYLLUS Gray subsp. sTRAMINEUS (Rydb.) C. L. Hitche. forma zionensis forma nov. Planta a subsp. stramineus differens: folia supra pilosula. This form differs from subsp. stramineus, in the leaves which are hirsute above with short semi-erect hairs of uniform length. Type. Near summit of Lady Mountain, Zion National Park, Washington County, Utah, June 19, 1928, T’. Craig 1439, Pomona College Herb. no. 184362. Range. Known only from Zion Canyon, Utah. Material seen, in addition to the type. Zion Canyon, Jones in 1923 and in 1925 (P), Garrett R2669 (NY), and Woodbury 19 (US); “Utah,” Ward 699 (US). 1943] HITCHCOCK: PHILADELPHUS 49 lg. PHILADELPHUS MICROPHYLLUS subsp. pumilus (Rydb.) comb. nov. P. pumilus Rydb. N. Am. Fl. 22: 173. 1905. Much like subsp. stramineus, especially as to pubescence of the calyces, but leaves usually averaging no more than 8 mm. long, their upper surfaces hirsute with many short stiff erect hairs mixed with fewer and longer more appressed hairs; petals 6-10 mm. long, usually slightly pubescent on the outer surfaces near base; stamens 32 (to 50), the filaments free or united at base only; styles from completely united to almost entirely free. Type. Tamarack Valley, San Jacinto Mountains, Riverside County, Hall 2500. Range. Apparently restricted to the San Jacinto and Santa Rosa Mountains of southern California at 7000-8500 feet eleva- tion. Possibly also in Lower California (see under subsp. stra- mineus ). Material seen. Catirornia. Riverside County: Santa Rosa. Mts., Munz 15392 (Wash) ; San Jacinto Range, Hall S00 (S); San Jacinto Mts., near Tamarack Valley, Hall 2500, type collection (C, NY,S, US); Dark Canyon, Munz 8762 (P) and Munz & John- ston 8738 (P); Long Valley to Palm Springs Trail, Jaeger 1015 (C, P) ; above Chino Canyon, Jaeger in 1922 (S, US). The variation in the degree of union of the styles is easily seen in this phase of the species. The flowers of the type collection have styles completely united, even the stigmas being united over half their length in a few cases. The collection made by Munz and Johnston is so similar to the type that Rydberg surely would have called them conspecific, yet the styles of this collection are free nearly to the ovary. It is because of such instances that the union or non-union of the styles is considered of secondary impor- tance as a diagnostic character. lh. PHILADELPHUS MICROPHYLLUS Gray subsp. typicus nom. nov. P. microphyllus Gray, l.c.; Rydb. N. Am. Fl. 22: 172. 1905. Leaves (12) 17-35 mm. long, ovate-lanceolate to lanceolate, strigose on both surfaces to glabrate above, the hairs all appressed but the upper surfaces usually greenish; calyces glabrate to mod- erately strigose, the tube 2—4 mm. long, usually strigose on the angles, the lobes often glabrous on the outer surfaces, moderately lanate on inner surfaces; petals 11-17 mm. long, not pubescent on the outer surfaces; stamens mostly 40-52 (32), practically free except for few that are geminate. Type. Eleven miles above Santa Fe, on Santa Fe Creek, New Mexico, June to July, 1847, Fendler 266. Range. Central and northern New Mexico, southern Colo- rado, and Apache County, Arizona; also from the Chisos Moun- tains of Texas. This, the typical form of the species, characterized chiefly by the large flowers and sparse pubescence of the calyx, is the only phase of the species found in northern New Mexico. It inter- 50 MADRONO [Voleia grades with subsp. occidentalis on the north and west, and with subsp. argyrocalyx on the south. Although most of the flowers of the type collection have about thirty-two stamens, the number of stamens present in the bulk of the material seen is between forty and fifty-two. Material seen. New Mexico. Rito de las Frijoles, Cockerell in 1912 (US). Socorro County: Beartrap Canyon, San Mateo Mts., Eggleston 18654 (NY, US); Hop Canyon, Magdalena Mts., Diehl 463 (P) and Herrick 608 (US), Copper Canyon, Magdalena Mts., Goldman 1669, intermediate with subsp. argyrocalyx (US). Bernalillo County: Albuquerque, Jones in 1884 (P). Valencia County: Grant’s Station, Wooton in 1892, approaching subsp. argyrocalyx (NY). Sandoval County: Balsam Park, Sandia Mts., Ellis 107 (NY, US) ;.Sandia Mts., Herrick in 1898 (US); Placitas, Sandia Mts., Wooton in 1910 (US); Ellis Ranch, Sandia Mts., Wooton in 1910 (US); Guadalupe Canyon, San Mateo Mts., Eggleston 18736 (US). Santa Fe County: Santa Fe Creek, Fendler 266, type collection (G, NY, US); Santa Fe Canyon, Heller § Heller 3792 (G, NY, P,S, US) and Wooton in 1910 (US) ; near Santa Fe, Arséne §& Benedict 15742 (US). San Miguel County: mouth of Indian Creek, Standley 4547 (G, NY, US). Taos County: Taos River Canyon, Nelson 11471 (C, G); Red Canyon, near Questa, Hitchcock et al 4148 (C, P, Wash, WSC). Cotorapo. Las Animas County: Brantley Canyon, Osterhout 2077 (C, NY, P); Mesa de Maya, 60 miles east of Trinidad, Rollins 1835 (NY, WSC). Montrose County: Cimarron, Nelson & Nelson 425 (C), with some doubt as no petals present. Fremont County: west of Parksdale, Jones in 1913 (S); Royal Gorge, Bacigalupi 1009 (NY, P, S), without petals, hence some doubt; Canyon City, Brandegee 84 (C). Utan. Grand River near Moab, Jones in 1915, intermediate with subsp. occidentalis (NY). Art- zona. Canyon de Chelly, Nelson 36, almost surely belongs here but no flowers on plant so identity cannot be certain (US). Several collections from the Chisos Mountains, Texas, must be placed here, even though they have smaller flowers than those common to the material from New Mexico. At first glance these Texan plants appear to be sufficiently different to be maintained as a separate entity, but it has been found to be impossible to construct a key that will separate them from subsp. typicus. Their leaves are quite green when dried, but this condition may be due to the care with which they were preserved; the stamens vary in number from twenty-nine to forty per flower; the flowers are about the size of those of subsp. occidentalis, but the leaves and calyces are more nearly glabrous than they are in that plant. It is entirely possible that these plants will prove to be genetically distinct from either subsp. typicus or subsp. occidentalis, but field study will be necessary in order to detect distinguishing charac- ters, if such there be. Trxas. Brewster County, Chisos Moun- 1943] HITCHCOCK: PHILADELPHUS 51 tains: Mt. Emory, Warnock 866 (US), Upper Boot, Cory 7076 (P) and Cory 7077 (G); 2 miles southwest of Boot Spring, Moore & Steyermark 3160 (C, G, NY, S); Lost Mine Peak, Ferris § Duncan 2856 (NY, S); with no further locality, Mueller 8013 (NY) and Havard 40 (G). li. PuHILaDELPHUS MICROPHYLLUS Gray subsp. occidentalis (Nelson) comb. nov. P. occidentalis Nelson, Bull. Torrey Bot. Club 25: 874. 1898. P. minutus Rydb., N. Am. Fl. 22: 178. 1905; type, Black Canyon of the Gunnison, Colorado, Baker 266. P. nitidus Nelson, Bot. Gaz. 42: 54. 1906; type, Sapinero, Colo- rado, Wheeler 425. Second cited specimen was Baker 266, hence P. nitidus is surely the same as P. minutus. Pubescence much the same as in subsp. typicus, the leaves often glabrate above, mostly 10-16 (to 25) mm. long, elliptic- lanceolate to lanceolate; petals 9-11 mm. long, not pubescent on outer surfaces; stamens usually about 40 (29-45), united at base only, if at all; styles wholly united or free one-third to one-half their length. Type. Near Rock Springs, Sweetwater Canyon, Wyoming, July 25, 1897, Smith 3595. Range. Central and northern Colorado, southwestern Wy- oming, and eastern Utah. The subsp. occidentalis merges with the subsp. typicus and is distinguished from that phase with some difficulty. In general, however, it has smaller flowers and leaves. In the western part of its range it also intergrades with subsp. stramineus but can usually be distinguished from that plant because of its greater number of stamens (forty or more as compared with thirty-two) and less pubescent calyces. There can be little doubt that P. nitidus and P. minutus are the same entity since Nelson cited three collections of P. nitidus, one of which was collected by Baker (no. 266), in Black Canyon, Colorado. This is the same collection which Rydberg, a year previously, had selected as the type of his P. minutus. Nelson specified that P. nitidus had “styles distinct down to the ovary.” Rydberg claimed P. minutus had “styles united one-half to two- thirds their length.” Although I have seen no flower which I would describe as having styles distinct to the ovary, I am convinced that the degree of union of the styles is not a particu- larly dependable character. One of the other two collections which Nelson cited under P. nitidus (Jones 6803), from Belknap, Utah, has the styles united to the stigmas in the flowers I have examined. Thus it seems evident that free or partially free styles are not a constant characteristic of subsp. occidentalis. Neither are the styles of subsp. typicus always united, for ex- ample, in a collection from the Sandia Mountains, New Mexico (Ellis 107), they are but incompletely united. Philadelphus minutus was separated from P. occidentalis in 52 MADRONO [Vol. 7 Rydberg’s key on the basis that the styles were partially free in the former; thus it can be seen that Rydberg and Nelson both believed their species to be unique and separable from P. occi- dentalis for the same reason. I hesitate, therefore, to reduce P. minutus to synonymy but can find no true criterion by which it can be maintained as distinct in any way from P. occidentalis. The calyx-tube, according to Rydberg, is 2 mm. long both in P. occidentalis and P. minutus, whereas it is “4-5 mm. long” in P. microphyllus. I do not believe that there is any such difference. Surely Rydberg could not have taken his measurements for the two species from flowers at the same stage of development as it is possible to find flowers of nearly equal size on plants which Rydberg would have called P. microphyllus, P. minutus, and P. occidentalis, respectively. According to Nelson, P. occidentalis has styles “‘free for from one-third to one-half their length.” Rydberg said of it “styles usually wholly and the oblong or clavate stigmas partly united.” Both conditions prevail and can be seen in the material here cited. Material seen. Cotorapo, S. Colorado, Palmer (US) and Popenoe in 1876 (US). Garfield County: Glenwood Springs, Palmer 38115 (NY, US). Montrose County: Cimarron, Jones in 1890 (P) and in 1925 (P); Black Canyon of the Gunnison, Baker 266, type collection of P. minutus (C, G, NY, P, US); Newcastle, Cary 153 (US). Fremont County: Canyon City, Brandegee in 1877 (C) and Osterhout 2092, very close to subsp. typicus and possibly really that form (P); Cotopaxi, Johnston & Hedgecock 740 (NY); Royal Gorge, Clokey 3791 (C, G, NY, P, S, US, WSC) close to subsp. typicus but calyx somewhat more pubescent; Oak Creek Canyon, Rollins 1240 (G, NY). Wyomine. Sweetwater County: near Rock Springs, Smith 6919 and Nelson 3595 (G, NY, US). This last collection is from the exact type locality and is the same number as the type but that collection was asserted to have been made by R. A. Smith rather than by Nelson. UrTan. Daggett County: 12 miles south of Manila, Hitchcock et al 3912a (C, P,S, WSC). Carbon County: Cottonwood Canyon, Graham 9520 (G). Beaver County: Beaver City, Palmer 151 (G, NY); S. Utah, Palmer 151 (US). Grand County: Thompson's Springs, Jones in 1913 (G, P, US, WSC); Grand Canyon, Graham 9916 (G). The following collections are also being referred to subsp. occidentalis, although they intergrade, in greater or lesser degree, with subsp. typicus. Utau. Grand County: Moab, Jones in 1913 (C, G, NY, S, US) and in 1891 (P), these plants are peculiar in that they have large thick leaves that are semi-glabrous on the upper surfaces; near Wilson Mesa, Rydberg & Garrett 5380 (NY). Summit County: Brush Creek Canyon, Uintah Mts., Goodding 1274 (G, NY, US); Brush Creek Gorge, Graham 10015 (US). Uintah County: Ashley Creek, near mouth of Dry Fork, Graham 1943] HITCHCOCK: PHILADELPHUS 53 6270 (US) ; 10 miles northwest of Vernal, Graham 7454 (G). San Juan County: Abajo Mts., Rydberg & Garrett 9608 and 9609 (NY, US); La Sal Mts., Purpus 6611 (C, P, US), near Clarke Lake, Maguire et al 5803 (C), road to Warner Ranger Station, Maguire et al 5802 -(C, US); locality uncertain, ‘“‘marvine laccolite”’ only data on label, hence probably also from the La Sal Mts., Jones 56680 (C, NY, P, US). Arizona. North end Carrizo Mts., Standley 7323 (US, with some doubt, as flowers lacking). Apache County: Luka-Chukai Mts., Goodman & Payson 2543 (NY, Wash). In the southwestern part of Utah and in northwestern Arizona subsp. occidentalis merges with subsp. stramineus. The following collections are intermediate between those two phases of the species. Most of them have the pubescence characteristic of subsp. stramineus but the stamen complement is more suggestive of subsp. occidentalis, in some few cases there are as few as thirty-two stamens but most flowers have about forty, fifty-two being the largest number seen. Urau. Sevier County: Belknap, Stokes in 1900 (NY, S) and Jones 63808 (NY, P,S. US); Burrville Canyon, Jones 5633 (C, NY, P, US). Piute County: Marysvale, Jones 5875p and 740547 (P, US); Bullion Creek, near Marysvale, Jones 5904d (NY, P). Arizona. Coconino County: Cape Royal, north rim Grand Can- yon, Peirson in 1927 (P); Grand View Trail to bottom of Grand Canyon, Ferris §& Duncan 2257 (S); Grand Canyon, Toumey (S), Toumey 1383 (US), Hitchcock 91 (US), and Knowlton 258 (US) ; Navaho Reservation, Vorhies 109 (C, G, NY). 2. PuirapELPHus Mearnsi Evans ex Rydb. N. Am. Fl. 22: 174. 1905. A low rounded shrub probably not much over 1 m. tall, the branches rigid, more or less spinescent, bark of young stems brownish, strigose, quickly exfoliating, the old branches dark gray; leaves elliptic or lanceolate, 6-25 mm. long, pubescence of very coarse hairs almost equally grayish-strigose on both sur- faces, petioles 1-3 mm. long; flowers mostly single at the ends of very short leafy stems, the hypanthium grayish-strigose, 1.5—2.5 mm. long in flower, 3 mm. long in fruit, sepals ca. 3 mm. long, acu- minate, strigose on outer surface, lanate within; petals whitish or ochroleucous, oblong-lanceolate, 7-11 mm. long, ovate-lanceolate to lanceolate, acute or sharply 2-toothed at apex; stamens 16-20 (24), the filaments short, free; styles less than 1 mm. long, united to top, stigmas 1.5—2 mm. long, almost completely united. Key To Supspecies oF P. MEARNSII Petals 8-10 mm. long; leaves mostly less than 15 mm. long, strigose on both surfaces with appressed pubes- cence; petioles 1-2 mm. long’ ..... 0.5... ..5.54045 _ 2a. P. Mearnsit subsp. typicus Petals 10-11 mm. long; leaves mostly over 15 mm. long, the upper surfaces with erect hairs; petioles 2-3 mm. NAYES cy & By ORS Ea RGR a 2b. P. Mearnsii subsp. bifidus 54 MADRONO [Vol. 7 2a. PuHinrapELPHUs Merarnsi subsp. typicus nom. nov. P. Mearnsu Evans ex Rydb., l.c. Characters as in key, the leaves rather thick, the pubescence of very coarse short hairs. Type collection. Upper Corner Monument, Grant County, New Mexico, April 28, 1892, Mearns 86 (G, NY,S, US). Other material seen. New Mexico. Eddy County: Guada- lupe Mts., near Three Forks of Rocky Arroyo, May 6, 1982, Wilkens 1833 (US) ; Carlsbad Cave, Bailey in 1924 (US). Texas. Culberson County: Guadalupe Mts., above McKittrick Canyon, July 17, 19381, Moore & Steyermark 3477. (C, G). Because of the uniform strigose pubescence, short styles and few stamens, this entity is readily distinguished from the other species and even though it is known only through three collec- tions, there can be no doubt of its validity. The Moore and Steyermark collection is in fruit and therefore lacks flowers, but the coarse hairs and short styles make it seem certain that it be- longs here. Although the species was described as having “‘about 15” sta- mens, those flowers of the type collection which I have examined have sixteen. The Wilkens collection also has sixteen stamens,, a fact which helps to convince one that it is conspecific with the Mearns collection. However, Bailey’s collection from Carlsbad has flowers with twenty-four stamens. A comparison of the Bailey and Wilkens collections cannot but convince one that they are not only conspecific but that they are too similar to be sepa- rated nomenclaturally, hence the species must be considered to have as few as sixteen (occasionally fifteen?) and as many as twenty-four stamens. 2b. PHirapeLPHUs Merarnsii subsp. bifidus subsp. nov. Planta P. Mearnsi subsp. typicus similis, hoc modo differens: foliis longioribus gracilioribusque, laminis ad 80 mm. longis, supra hirsuto-hispidis, pilis erectis; petiolis 2-3 mm. longis, petalis 10-11 mm. longis, apices 2-dentibus; staminibus 24, fila- mentis non connatis. Leaves longer and more slender than in subsp. typicus, the blades as much as 30 mm. long, pubescence of their upper sur- faces hirsute-hispid, the hairs slender, scarcely at all appressed, petioles 2-8 mm. long; petals 10-11 mm. long, the apices dis- tinctly sharply 2-toothed; stamens 24 (probably some variation), the filaments free. Type. Sierra Madre, near Monterrey, Nuevo Leon, Mexico, May 2, 1906, Pringle 13879 (G). Known only from the type collections at Gray Herbarium and United States National Museum, one branch of the latter speci- men with partially double flowers. Superficially this collection seems to bear little resemblance to the type of P. Mearnsu. How- ever, it is evident that that collection was an extremely small- 1943 | HITCHCOCK: PHILADELPHUS 55 flowered plant and that probably the other collections cited (Wilkens 1833) are more truly representative of the entity. Since the plant of the Wilkens collection is larger-leaved and larger- flowered, it is more suggestive of Pringle’s plant from near Monterrey. Because there is so much similarity between the flowers of the two entities, the plant from Mexico is being ac- corded subspecific rank. Toothed, oblong petals, few stamens, short styles, and united stigmas comprise an unusual combination of characters and their presence in both these entities makes it seem logical to assume that there is close relationship between the two, a relationship that is too close to be brought out by ac- cording both of them specific status, 3. PHILADELPHUS SERPYLLIFoLIUS Gray, Pl. Wright. 1: 77. 1850. A low rounded, rather rigid shrub 1—2 m. tall, young branches brownish, densely strigose-pubescent, bark‘of older stems exfoli- ating, the old stems grayish; leaves entire, ovate-lanceolate, 6—30 (averaging about 15) mm. long, 3-12 mm. broad, 3-nerved from base, the upper surfaces greenish and often shining, from sparsely strigose to rather densely strigose and short hirsute, lower surfaces grayish, the pubescence rather dense, of long straight appressed hairs and close covering of tangled tomentum, distinctly petiolate, the petioles 1-3 mm. long; flowers commonly 1 or 8 (or 2). at the ends of short leafy shoots, the pedicels 1 or 2 mm. long; calyx usually silvery with long appressed and shorter more or less curled hairs, but sometimes the indumentum more sparse and calyx greenish, the tube 2-3 mm. long in flower, 3—4 mm. long in fruit, the lobes lanceolate, acuminate, 4-5 mm. long, densely lanate on inner surfaces; petals apparently cream colored, 6-8 mm. long, more or less oval and usually somewhat emarginate or erose; stamens usually 28 (few more or less), the filaments free; styles less than 1 mm. in length, united, the stigmas 1.5—2 mm. long, almost completely united. Type. “Between western Texas and El Paso, New Mexico,” in 1851, C. Wright. Very closely matched by a specimen col- lected near Fort Davis, Texas, Palmer 34474. Material seen. New Mexico. Without definite locality, 1851-— 52, C. Wright 1100 (G, NY); Mexican Boundary Survey (NY), probably the same as the first.. Texas. Head of Seco, Reverchon 54 (G), one branch of this collection has leaves only 12 mm. long whereas a younger more vigorous shoot has leaves that are 30 mm. long. Jeff Davis County: deep canyon off Limpia Canyon, near Fort Davis. Palmer 34474 (NY, US). Edwards County: limestone bluffs, Upper Cedar Creek, Palmer 12333 (C). Kendall County: Spring Creek, near Boerne, Palmer 11492 (C) and Palmer 11595, 12900 (C, P) ; Cibola Canyon, Boerne, Pennell 10418 (NY), pubescence of leaves strigose above; Musquez Canyon, W. Texas, Havard 41 (G). Culberson County: McKittrick Canyon, Guada- lupe Mts., Moore & Steyermark 8479 (C, G, NY,S). 56 MADRONO [Vol. 7 Philadelphus serpyllifolius is, of course, closely related to P. microphyllus but differs strikingly because of the tomentum of the leaves and the extremely short styles. The collections from Kendall County are much less pubescent than is the rest of the material. 4. PuHILaADELPHUs PurRPusu Brandegee, Univ. Calif. Publ. Bot. A 200. ~ VONZ. A low spreading shrub 1-2 m. tall; young branches brownish, densely strigose-pubescent, older bark grayish, winter buds quite obviously not enclosed in the leaf bases, petioles 1-4 mm. long, blades entire, ovate-lanceolate to lanceolate, usually slightly mu- cronate, 20-35 mm. long, 3-nerved, pubescence of both surfaces grayish-green, almost equally strigose with short thick hairs; flowers single at the ends of short branches, the pedicels 3-6 mm. long; calyces grayish-strigose, the tube 8—4 mm. long, the lobes 4-6 mm. long, acute; petals obovate to oval, 10-15 mm. long, rounded, scarcely emarginate; stamens 40—50, filaments distinct ; styles 2-3 mm. long, united about one-half to four-fifths their length, basal portion and surrounding ovary grayish pilose; stigmas free, 2-3 mm. long. Type. Minas de San Rafael, San Luis Potosi, Mexico, Purpus 5368. Known only from two collections from the type locality, Pur- pus 5868 (C, G, NY) and Purpus 4910 (G, US). Philadelphus Purpusii is most easily distinguished because of the pilosity of the styles and upper ovary, but the grayish-green strigose leaves, large petals, and exposed buds are all features that help to make it the distinctive species that it is. University of Washington, Seattle, March, 1942. A NEW SPECIES OF PHACELIA FROM SALINE VALLEY, CALIFORNIA LINCOLN CONSTANCE Phacelia amabilis sp. nov. Herba annua vel biennis, omnino glanduloso-puberula et hispida praecipue in calycibus inflores- centibusque, circa 1 m. alta; caulis crassus ramosus; folia peti- olata oblonga oblongo-ovatave, 8-15 cm. longa, 3-5 cm. lata, pinnatifida, lobis oblongis dentatis, summa reducta minus alte divisa; inflorescentia corymbosa, cymis 5-12 cm. longis in fructu erectis; pedicelli in fructu 2-3 mm. longi; calycis lobae anguste lanceolatae, 3-5 mm. longae, 1-2 mm. latae, corolla plus quam dimidio breviores, capsulam leviter excedentes; corolla late cam- panulata, 7-8 mm. longa, 8-12 mm. lata, alba, lobis integris; appendiculae supra tubae basin minus quam 1 mm. insertae, parte libera lata; stamina exserta; stylus exsertus, pallido-lilacinus ; 1943] KEARNEY AND PEEBLES: GILIA MULTIFLORA 57 capsula ovoidea, 3-4 mm. lata; semina plerumque 2 vel 4, 3—4 mm. longa, tenuissima pallidissima, non corrugata, superfacie ventrali jugo saliente utrinque excavata. Annual or biennial, about 1 m. high; stem stout, branching ; herbage glandular-puberulent throughout and hispid, especially on the calyces and in the inflorescence; leaves petiolate, oblong to oblong-ovate, 8-15 cm. long, 8-5 cm. broad, pinnatifid, the lobes oblong, dentate, the uppermost leaves reduced and less deeply divided; inflorescence corymbose, the cymes 5-12 cm. long and erect in fruit; pedicels 2-3 mm. long in fruit; calyx lobes narrow lanceolate, 3-5 mm. long, 1-2 mm. broad, less than one-half as long as the corolla, slightly exceeding the capsule; corolla broadly campanulate, 7-8 mm. long, 8-12 mm. broad, white, the lobes entire; appendages attached a little less than 1 mm. above the base of the tube, the free portion broad; stamens and style exserted 5 mm. or more, the latter pale lavender; cap- sule ovoid, 3-4 mm. long, 2-3 mm. broad; seeds usually 2 or 4, 3-4 mm. long, very thin and pale, not corrugated, the ventral sur- face excavated on each side of a salient ridge. Type. “In full bloom along creek, Hunter Creek, altitude 1800 feet, Saline Valley, Inyo County, California,’ April 21, 1942, Annie M. Alexander and Louise Kellogg 2681 (Herbarium of the University of California no. 671871). Phacelia amabilis was first thought to be an albino form of P. crenulata with which it was growing. In his excellent, “Revision of the Phacelia crenulata group for North America” (Bull. Torrey Bot. Club 64: 81-96, 1383-144. 1937), Voss has provided the most complete modern treatment yet available for any portion of this interesting and “difficult” genus. Although interspecific differences are small in this group, most of the entities accorded specific rank by him appear to be rather sharply distinguished from one another. The present spe- cies, by possession of a broadly campanulate corolla, exserted stamens and ventrally excavated but uncorrugated seeds would appear to be most closely related to P. congesta Hook., of Texas, New Mexico, southeastern Arizona and northern Mexico. Be- sides the wide discontinuity in geographical range, P. amabilis differs from P. congesta—as indicated on the accompanying plate —in its shorter calyx, larger and differently proportioned corolla, more exserted stamens and style, and its larger, paler, thinner and more broadly margined seeds. For the past several years, the Misses Alexander and Kellogg have collected extensively in the relatively inaccessible desert mountains of the southwestern United States. The Her- barium of the University of California has been enriched by receiving their nearly 3500 numbers, as well as many valuable duplicates, which have been distributed to other leading herbaria. 58 MADRONO-~. [Vol. 7 Priate 3. Puacetia. Figs. a—-e, P. amabilis: a, habit; b, flower; c, fruiting calyx and capsule; d, expanded corolla; e, seeds. Figs. f—j, P. congesta var. rupestris: f, habit; g, flower; h, fruiting calyx and capsule; i, expanded corolla; j, seeds. (Habit drawings X 1%; flowers and capsules, x3; seeds, X 7.) 1943] KEARNEY AND PEEBLES: GILIA MULTIFLORA 59 Phacelia amabilis is but one of the rare or seldom-collected species which has been secured as a result of their indefatigable efforts. The writer would ordinarily have preferred to delay publi- cation of this species until more material is available, but in view of the curtailment of field work in the foreseeable future, it seemed advisable to describe it at this time. Department of Botany, University of California, Berkeley, November, 1942. GILIA MULTIFLORA NUTT. AND ITS NEAREST RELATIVES Tuomas H. Kearney AND Rosert H. PEEBLES Gilia multiflora Nutt., a widely distributed species of New Mexico and Arizona, is extremely variable, but is distinguished from its nearest relatives by having the corolla tube nearly always two to three times the length of the lobes, and rarely less than one and a half times the length of the calyx. Very similar to G. multi- flora in habit, foliage, and pubescence is G. polyantha Rydb. which differs, however, in having a shorter corolla tube, this approxi- mately equal in length to the lobes and the calyx. Typical G. polyantha is known apparently only from south- western Colorado. Gilia brachysiphon Woot. and Standl., of south- western New Mexico, would seem to be specifically distinct from G. polyantha were it not for the occurrence in north-central Ari- zona of a form that is intermediate in several characters. For this reason, it seems best to treat G. brachysiphon and the hitherto undescribed Arizona plant as varieties of G. polyantha. So far as present information goes, the three forms of this species are rather widely separated geographically. Another more distantly related member of this small group of perennial plants with filaments normally conspicuously exserted and declined is G. Havardi A. Gray, an apparently rare species of southwestern Texas. The corolla is more pronouncedly zygo- morphic, especially in respect to the closely grouped and parallel- declined stamens, than in G. multiflora and G. polyantha, and for this reason Brand (in Engler, Pflanzenr. 47°°: 172. 1907) re- stored this species to the genus Loeselia, where it was placed originally by Asa Gray. Gilia multiflora and G. polyantha, how- ever, also show a tendency to zygomorphy and the writers concur in Gray’s final conclusion that G. Harvardi and G. multiflora are congeners. Gilia Macombii Torr., although evidently related to G. multi- flora, is not considered here because the stamens are not exserted from the corolla tube, or project only about the length of the anthers. 60 MADRONO [Vol. 7 KEY TO THE SPECIES AND VARIETIES. Inflorescences short, not thyrsoid, the zygomorphic flowers in loose, few-flowered terminal clusters; herbage eglandular, villous, even in the _ in- florescence, with long soft white hairs; stems nu- merous from a branched caudex, short; corolla tube 6 to 9 mm. long, about twice as long as the calyx, not or only slightly longer than the lobes.. 1. G. Havardi Inflorescences elongate, thyrsoid-glomerate, the flow- ers nearly regular; herbage usually more or less glandular in the inflorescence, the long white hairs mostly confined to the lower part of the plant. Corolla tube 10 to 15 (rarely only 7) mm. long, 1.6 to 2.7 (rarely only 1.2) times as long as the calyx and 1.5 to 3.2 times as long as the corolla lobes, the lobes prevailingly oblong or nar- rowly elliptic but occasionally slightly obovate; inflorescences normally contracted and_ short- branched; pubescence various ................ 2. G. multiflora Corolla tube 4.5 to 6.5 mm. long, shorter than to 1.2 (exceptionally 1.7) times as long as the calyx and shorter than to 1.5 times as long as the corolla lobes. Lobes of the whitish, sometimes purple-dotted, corolla 3 mm. wide, oval or broadly elliptic; inflorescences relatively open and_long- branched, copiously puberulent with glandular hairs, these intermixed with short white hairs. 3. G. polyantha (typical) Lobes of the pale violet corolla 1.5 to 2.7 (excep- tionally 3.5) mm. wide. Inflorescences relatively contracted and short- branched, the glomerules dense; upper part of the plant canescent with short white hairs, the glandular hairs relatively few and mostly hidden by the others; corolla lobes prevailingly elliptic or oblong ............ 3a. G. polyantha var. brachysiphon Inflorescences relatively expanded and long- branched, the glomerules relatively loose; upper part of the plant commonly with abundant glandular hairs, these intermixed with but usually not concealed by the short white hairs; corolla lobes _ prevailingly oblanceolate or obovate <=) se oe 3b. G. polyantha var. Whitingi 1. Grrr Havarpt A. Gray, Syn. Fl. N. Amered. 2, 2°. 4 1886. Loeselia Havardi A. Gray, Proc. Amer. Acad. Arts and Sci. 19: 87. 1883. Presidio and Brewster counties, southwestern Texas, near Presidio (Havard in 1881, the type collection), vicinity of the Chinati Mountains (Nealley in 1889), vicinity of the Chisos Moun- tains (Sperry 623). The stems are less than 20 cm. long. The species is very distinct in appearance from other members of this group. The herbage is uniformly villous with long, soft, white, segmented hairs, these extending even to the calyx, whereas in 1943 | KEARNEY AND PEEBLES: GILIA MULTIFLORA 61 G. multiflora and G. polyantha such hairs are confined largely to the lower leaves. The leaves are mostly pinnatifid with 3 to 5 lobes. The ovules are “several in each cell” (Gray, ibid.). 2. GILIA MULTIFLORA Nutt., Jour. Acad. Nat. Sci. Phila. ser. 2, i 154. 1848. This species seems to be known only from the mountains of New Mexico and Arizona, where it is common and widely dis- tributed, but it probably occurs also in northern Mexico. The type, collected by Gambel on “‘sandy hills along the border of the Rio del Norte, New Mexico,” has not been seen by the writers. There is no type material in the herbarium of the Philadelphia Academy of Sciences or in the Gray Herbarium. This is a highly variable species, but departures from the norm in floral charac- ters do not appear, as a rule, to be correlated among themselves or with peculiar vegetative characters. The plants are poten- tially long-lived and some of the specimens from southern Arizona are distinctly woody toward base. The pubescence consists of two main types of hairs: (1) soft, white, flattened, segmented hairs, these often elongate near the base of the plant, especially on the leaves, but normally becoming very short in the inflores- cence, where they are usually intermixed with (2) clavate or _ stipitate glandular hairs. Occasional specimens show granular puberulence, apparently of a resinous nature. The leaves vary from entire to pinnatifid with 3 to 7 narrow lobes. The inflorescences are normally contracted and_ short- branched, but occasional specimens resemble G. polyantha Rydb. in their relatively open and long-branched inflorescences. The insertion of the filaments is normally very nearly at the same level, usually just below the rim of the throat, but sometimes deep in the tube. The anthers are at nearly the same level, or some of them are 1 to 2 mm. below the others. Counts in eight ovaries, each from a different individual plant, showed a range from 5 to 15 ovules, indicating a variation of from 2 to 5 per cell. Two outstanding variants, both found in Arizona, may prove worthy of recognition as varieties when more material is avail- able. The first, collected near Flagstaff (Rusby 729), and on “Ivy Mesa, Mogollon Mountains’! (MacDougal 644), is character- ized by numerous, slender, almost flexuous stems; inflorescences with subfarinose puberulence, obscurely if at all glandular or granular; leaves nearly all entire except near the base of the plant, very sparsely villous; corolla tube only 7 to 8.5 mm. long and only 1.6 to 2.0 times as long as the lobes, whereas in G. multi- flora the tube is usually at least 10 mm. long and more than twice as long as the lobes. The second remarkable variant was collected in Oak Creek Canyon, Coconino County (Whiting 1053/5300). It is distin- 1'The “Mogollon Mountains” of Arizona are the precipitous southern edge of the Mogollon Mesa, also often referred to as the “Tonto Rim” or “Mogollon Rim.” 62 | MADRONO [Vol. 7 guished from all forms of G. polyantha by having a corolla with tube 2.5 times as long as the lobes, but differs markedly from typical G. multiflora in the following characters: calyx 8.5 mm. long and nearly equalling the tube of the corolla, whereas in G. multiflora the calyx is normally only 5 to 7 mm. long and seldom more than three-fifths as long as the tube; corolla lobes broadly obovate, three-fourths as wide as long, whereas in G. multiflora they are commonly oblong or only slightly obovate and not more than half as wide as long; corolla throat exceptionally ample, 3.5 mm. wide at the orifice in the pressed specimen, as compared with a usual width of 2 to 8 mm. in G. multiflora. A collection in the Baboquivari Mountains, Arizona, (Gilman B124) has an equally wide corolla throat but is otherwise normal in its flower charac- ters, and the plant is decidedly woody toward base, which is not the case in Whiting’s plant. 3. Git1a PoLYANTHA Rydb., Bull. Torr. Bot. Club 31: 6384. 1904. Gilia exrserta A, Nelson, Bot. Gaz. 40: 65. 1905. Giulia multiflora var. polyantha Brand, in Engler, Pflanzenr. 4°°°: 1138. LOO7. The type of G. polyantha, which is also the type of G. easerta, was collected at Pagosa Springs, Archuleta County, southern Colorado, altitude 7,000 feet (C. F. Baker 538) and the typical form of the species seems to be known only from this locality, where it has been collected also by Bethel, Willey, and Clokey (4251). Itis taller and has more elongate, longer-branched inflo- rescences than most specimens of G. multiflora and is readily dis- tinguished from that species by the flower characters stated in the key. On the other hand, it resembles G. multiflora in habit, foliage, and pubescence. The leaves sometimes have as many as 9 lobes. Counts of the number of ovules per ovary gave 13 for the type specimen and 10 for the other. A. Nelson (ibid.) re- ported the ovules as only “‘about 2 in each cell,” in the type col- lection. The plant was stated by Nelson to be biennial, but it is almost certainly perennial, as Rydberg described it. 3a. Gitia PotyantHa Rydb. var. brachysiphon (Woot. and Standl.) comb. nov. Gilia brachysiphon Woot. and Standl., Con- trib. U.S. Nat. Herb. 16: 160. 1913. The type of Gilia brachysiphon was collected in the Organ Mountains, Dona Ana County, New Mexico (Wooton in 1894) and this variety is known also from several other localities, all in southwestern New Mexico (Sierra, Grant, and Catron counties). It differs from typical G. polyantha, and from most specimens of the following variety, in the rarity or entire absence of glandular hairs. Counts showed 18 ovules per ovary in the type and 14 in another specimen. 3b. Giza PotyantTHa Rydb. var. Whitingi var. nov. A forma typica G. polyanthae corolla violacea lobis plerumque obovatis differt; a G. polyantha var. brachysiphon inflorescentia laxiore et plerumque copiose glandulifera differt; a G. multiflora 1943] REVIEW 63 corollae tubo quam calyce et corollae lobis vix longiore distin- guitur. The type was collected at the Grand Canyon, Coconino County, Arizona (Whiting 1072/5200, U. S. Nat. Herb. 1814983). Other collections, all in Coconino County, at elevations of 6,800 to 7,200 feet are: Grand Canyon (Kggleston 15669, 15682), south rim of the Grand Canyon (Collom 1073), Grand Canyon Road (Whiting 1047/4311), near the Grand Canyon (Lemmon in 1884),? Williams (Rusby in 18838), base of the San Francisco Peaks (Wooton 489), Walnut Canyon National Monument (Beaubien 1054/5320). There is also, in the United States National Her- barium, an imperfect specimen, labeled in G. R. Vasey’s hand- writing, ““Nevada, Lt. Wheeler, 1872,” that apparently belongs here. This specimen may have been the basis for the inclusion of Nevada in the range of G. multiflora, in Tidestrom’s Flora of Utah and Nevada (Contrib. U: S. Nat. Herb. 25: 485. 1925). If it is correctly labeled as to locality, var. Whitingi has a wider range than the other forms of G. polyantha. There is a much stronger tendency than in G. multiflora, and in the other forms of G. polyantha, to have the corolla lobes widest near the apex, but this is not invariably the case. The stamens, as is usually also the case in G. multiflora, are inserted very nearly at the same level. Counts made on six individual plants, includ- ing the type, showed that the number of ovules per ovary varies from 6 to 18. Bureau of Plant Industry, United States Department of Agriculture, Washington, D. C. United States Field Station Sacaton, Arizona. June 27, 1942 REVIEW Wild Violets of North America. By Vioua Bratnerp Bairp. Pp. xv + 225, with 17 illustrations in the text and 80 plates in full color. University of California Press, Berkeley. 1942. $10.00. This attractive volume, written in non-technical language, will please both professional botanists and amateurs. The prepara- tion of a complete account of North American violets is a task for which the author is especially well equipped. Familiar with the group since childhood, Mrs. Baird assisted her father, Ezra Brainerd, in his well-known studies on the genus and during later years she has continuously added to her knowledge by field and garden studies of almost every species. The brief preamble contains an informative résumé of the distribution and probable sources of the North American species of Viola, a discussion of the variation in such characters as leaf 2Cited by Brand (Pflanzenr.) under G. multiflora var. polyantha. The collection at Cosnino, Ariz. (Jones 4043) also cited by Brand under var. polyantha, is G. multiflora. 64 MADRONO [Vol. 7 shape and flower color, and a simple explanation of the flower structure and function. There follows a key to the species groups which is both technically accurate and easily understood and utilized by the layman. Especially valuable as a scientific contri- bution are the two pages of line drawings illustrating the vari- ation in the shape of the pistil, the remarkable and almost ludi- crous development of which will interest and perhaps even amuse every reader. Since all were drawn from fresh material, persis- tent misconceptions as to pistil shape in certain species are here for the first time corrected. The classification divides the group into three sections, each of which is introduced by a brief discussion of the characters common to its members. The account of each species is very readable and contains information as to the outstanding charac- ters, distribution, habitat and flowering period of each, supple- mented by notes on any features of unusual interest such as the derivation of the name, by whom and where the species was first collected, and mention of the characters by which it may be dis- tinguished from its nearest relatives. As is fitting in a treatment which is both popular and technical, Mrs. Baird has not hesitated to group under their respective linneons minor entities which have been named as species. Thus the specific units as she conceives them are readily recognizable even by those without previous experience with taxonomic concepts. The plates, reproduced from water colors by the late F. Schuyler Matthews, are all natural size and indicate with an un- usual degree of accuracy the different shades in the flower color throughout the North American representatives of the genus. The color of practically every species was matched with great care with living material from native colonies or garden trans- plants and has been faithfully reproduced. Following the general text, in a “Key to Colors in Wild Violets,’ the species are grouped according to their principal color tones and these groups are still further subdivided as to the precise shade of flower color. Next, a list of the species is in- cluded with citations to the original descriptions and data as to the sources of the material used as a basis for the color plates. The volume closes with indices to the scientific and popular names. The book is handsomely bound in violet blue cloth with gold lettering and expertly printed in twelve point Baskerville on a highly lustrous coated paper. The edition is limited to one thou- sand signed copies.—ETHEL Crum. Nee MADRONO A West American Journal of Botany A quarterly journal devoted to important and stimulating articles dealing with plant morphology, physiology, taxonomy, and botanical history. These volumes should be a part of every botanist’s li- brary and should be made accessible to students of all universities and colleges. Volume I, 1916-1929. . . $5.00 Volume II, 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 Single numbers....... 0.75 The subscription price of MADRONO is $2.50 per year. We solicit your pat- ronage, Address all orders to: William Hiesey Bus. Mgr. Carnegie Institution of Washington Stanford University, California NUMBER 3 MADRONO A WEST AMERICAN JOURNAL OF BOTANY orewortHY Puants From IpaHo. II, Arthur Cronquist Leerrmacy or THaticrraum potycarrum Wats. Arthur Cronquist .. ann Learrimate NAMES—AND THALICTRUM PpoLycaRPUM S. Warts., C. A. Weatherby Dkk io Published at North Queen Street and McGovern Avenue Lancaster, Pennsylvania July, 1943 MADRONO A WEST AMERICAN JOURNAL OF BOTANY Board of Editors Herzert L. Mason, University of California, Berkeley, Chairman. LeRoy Asrams, Stanford University, California. Epecar Anperson, Missouri Botanical Garden, St. Louis. Lyman Benson, University of Arizona, Tucson. Hersert F. Copetanp, Sacramento Junior College, Sacramento, California. Ivan M. Jonnston, Arnold Arboretum, Jamaica Plain, Massachusetts. Miunprep E. Maruias, University of California, Berkeley. Bassetr Macume, Utah State Agricultural College, Logan. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board—ANNETTA CARTER Department of Botany, University of California, Berkeley Business Manager—Wittiam Hussey North Queen Street and McGovern Avenue, Lancaster, Pennsylvania or Carnegie Institution of Washington Stanford University, California Entered as second-class matter October 1, 1935, at the post office at Lancaster, Pa., under the act of March 3, 1879. Established 1916. Published quarterly. Subscription Price $2.50 per year. Completed volumes I to V inclusive, $25.00; each volume $5.00; single numbers $0.75. Papers up to 15 or 20 pages are acceptable. Longer contributions may be accepted if the excess costs of printing and illustration are borne by the contributor. Range extensions and similar notes will be published in con- densed form with a suitable title under the general heading “Notes and News.” Articles may be submitted to any member of the editorial board. Manuscripts may be included in the forthcoming issue provided that the contributor pay the cost of the pages added to the issue to accommodate his article. Reprints of any article are furnished at a cost of 4 pages, 50 copies $4.10; 100 copies $4.50; additional 100’s $0.85; 8 pages, 50 copies $5.95; 100 copies $6.60; additional 100’s $1.30; 16 pages, 50 copies $8.35; 100 copies $9.35; additional 100’s $2.00. Covers, 50 for $2.75; additional covers at $1.65 per hundred. Reprints should be ordered when proofs are returned. Published at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania, for the CALIFORNIA BOTANICAL SOCIETY, INC. President: H. S. Reed, University of California, Berkeley. First Vice- President: N. T. Mirov, California Forest and Range Experiment Station, Uni- versity of California, Berkeley. Second Vice-President: Jack Whitehead, Uni- versity of California Botanic Garden, Berkeley. Secretary: Clarence R. Quick, United States Department of Agriculture, 26 Giannini Hall, University of Cali- fornia, Berkeley. Treasurer: William Hiesey, Carnegie Institution of Wash- ington, Stanford University, California. Annual membership dues of the California Botanical Society are $2.50, $2.00 of which is for a year’s subscription to Madrofio. Dues should be remitted to the Treasurer. General correspondence and applications for membership should be addressed to the Secretary. 1943] MATHIAS AND CONSTANCE: TAUSCHIA 65 A NEW SPECIES OF TAUSCHIA FROM THE STATE OF WASHINGTON Mivprep EF. Maruias anp LINcoLN CONSTANCE Tauschia Hooveri sp. nov. Herba acaulescens, pseudoscapo e tubere globose, 10-14 cm. alta, omnino glabra glauca; folia in ambitu ovato-triangularia, petiolo excluso, 3-5 cm. longa, 1—6 cm. lata, pinnata vel partim bipinnata; foliola linearia, acuminata, 15-35 mm. longa, 1-2 mm. lata, integra, calloso-apiculata, mar- ginibus reflexis; petioli 2-3 cm. longi; pedunculi graciles, 2—4 cm. longi; calycis dentes obsoleti; corollae albae; antherae pur- pureae; styli breves, filiformes, recurvati; carpophorum ad medium bifidum, ramis ad apicem approximatis; fructus lineari- oblongus ad apicem versus leviter attenuatus, 5-7 mm. longus, circa 2 mm. latus, glaucus, costis evidentibus filiformibus; vittae parvae solitariae in valleculis, 2 in commissura; semini facies fere plana. Acaulescent with a pseudoscape 10-14 cm. high arising from a globose tuber; glabrous and glaucous throughout; leaves ovate- deltoid in general outline, excluding the petiole 3-5 cm. long, 4-6 cm. broad, pinnate to partially bipinnate, the leaflets linear, acuminate, 1.5—3.5 cm. long, 0.1-0.2 cm. broad, entire, callous- tipped, the margins reflexed; petioles 2-3 cm. long; peduncles slender, 2-4 cm. long; involucre and involucel wanting; fertile rays 3—7, stout, unequal, 2-10 mm. long; pedicels 1-3 mm. long; calyx teeth obsolete; flowers white, the anthers purple; styles short, filiform, recurved; carpophore 2-cleft about one-half way to the base, the halves approximate to the tip; fruit linear- oblong, tapering slightly at apex, 5-7 mm. long, about 2 mm. broad, glaucous, the ribs filiform but evident; oil tubes small, usually solitary in the intervals, 2 on the commissure; seed face nearly plane. Type. “Near Cowiche, Yakima County, Washington,” April 20, 1942, Robert F. Hoover 5689 (University of California Her- barium no. 671873). There is one other collection, “Hills south of White Swan, Yakima County,’ March 24, 1942, Hoover 5616 (Univ. Calif. Herb.). In a letter from the collector dated November 5, 1942, the following additional information is given: “In answer to your question about the habitat of T’auschia Hooveri, it grows in ‘scablands, in rather barren rocky clay, with Artemisia rigida, Sisyrinchium Douglasii, Eriogonum thymoides, Viola trinervata, etc.”’ To the best of our knowledge, this species has never been collected before, although the Yakima area has been rather in- tensively botanized previously by various collectors. It un- doubtedly escaped notice because the plants were in ripe fruit Mapronxo, Vol. 7, pp. 65-96. July 29, 1943. IAG AS AUG + 66 MADRONO [Vol. 7 d Fic. 1. TYauschia Hooveri: a, in anthesis, x1; 6, in fruit, X1; c, side view of fruit, x 5; d, cross section of fruit, X< 10. before the end of April, and because in anthesis they so greatly simulate certain of the “bulbous” species of both Lomatium and the two species of Orogenia. In a recent series of nomenclatorial transfers, the authors (Bull. Torrey Bot. Club 68: 121-124. 1941) proposed to reduce the monotypic genus Hesperogenia Coulter and Rose under Tauschia Schlecht. This step was not taken without some mis- giving inasmuch as Hesperogenia is known only from Mount Rainier, Washington, and the geographically nearest undoubted Tauschia is not known to occur north of southwestern Oregon. In addition, the suborbicular fruit of Hesperogenia is unique in 1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 67 Tauschia, although its other characters fit the latter genus very well. The discovery of YTauschia Hooveri, however, ends the supposed geographical isolation of Hesperogenia and appears to confirm the desirability of including it under T'auschia. We are glad to have the opportunity to name this remarkable species for Dr. Hoover, whose critical collections and keen ob- servations have been invaluable to us in our work with the Umbelliferae. Department of Botany, University of California, Berkeley, November, 1942. NOTES ON THE FLORA OF THE CHARLESTON MOUNTAINS, CLARK COUNTY, NEVADA V. CACTACEAE Ira W. CLOKEY This treatment of the Cactaceae continues a series devoted to a study of the flora of the Charleston Mountains in southern Nevada and published as follows: Madrofio 4: 128-130. 1937; Bull. So. Calif. Acad. Sei. 87: 1-11. 1988; l.c. 38: 1-7. 1939; and Madrofio 6: 211-222. 1942. I wish to express my thanks to Dr. L. Benson, Dr. E. U. Clover, Mr. Fred Gibson and Dr. Ira L. Wiggins for assistance in the study of the Cactaceae of the Charleston Mountains and for specimens of and information re- garding these and related species. Specimens of all of the spe- cies treated below except those of Mammillaria tetrancistra Engelm. have been widely distributed to the herbaria of the world. Types of the new species are all in the Clokey Herbarium now at the Bull. So. Calif. Acad. Sci. 87: 1-11. 1988; lc. 838: 1-7. 1939; garding these and related species. Specimens of all the spe- University of California, Berkeley. Key To THE GENERA OF CACTACEAE Stems jointed, cylindrical or flat; leaves small, subulate, de- ciduous; areoles on tubercles or on flat surfaces, with numerous, barbed glochids; glochids and flowers produced from the same areoles; ovaries with areoles and glochids; spines barbed or not barbed .......................... 1. Opuntia Stems not jointed, cylindrical, without leaves; areoles on ridges or tubercles, without glochids; flowers produced above the areoles ; ovaries with or without areoles, with- out glochids; spines not barbed. Stems ribbed; spines borne on definite ridges. Flowers borne above old spine- -bearing areoles, solitary, appearing lateral, purple or crimson; tube and ovary spiny; fruit colored, thin-skinned, spiny Re Mig e ee 2. Echinocereus Flowers borne above young areoles, appearing sub-termi- nal in a circle near the top of plant; ovary scaly; Ube reens WwATHOUG SPINES: |, Gydie aa class dod a Wath be 3. Echinocactus Stems not ribbed; spines borne on tubercles arranged in rows or scattered. 68 MADRONO [Vol. 7 Flowers lateral; tubercles not grooved on upper side; one or more of the spines hooked; fruit red; seeds black, rugose, with a thick corky base in our species ...... 4. Mammillaria Flowers terminal; tubercles grooved on upper side; none of the spines hooked; fruit green or rose, ripening slowly; seeds brown, pitted, without a thick corky | Of: |< Sane RO igi ee ooo bes, OM at lad ts um Be IU Dae. 5. Coryphantha KEY TO THE SPECIES oF OpuntTIA MILL. Joints cylindrical and tuberculate. Spines smooth, covered with loose, hyaline sheaths. Stem slender and solid, with a woody axis; tuber- cles flattened; spines solitary or absent ...... 1. O. ramosissima Stems thick and fleshy with a reticulated, cylindri- cal axis; tubercles raised; spines more than one. Tubercles two to three times as long as wide; fruit: dry, with stoutespines 2.02 9.27. 2. O. acanthocarpa Tubercles less than twice as long as wide. Plants erect, with one or few main stems; stems loosely branched; flowers yellow, tinged with red; fruit dry, with stiff spines on thesuppeér areoles!' sn 3. O. echinocarpa Plants low, bushy, without a main stem; stems compactly short branched; flowers green- ish-yellow; fruit somewhat fleshy, yellow, with a few delicate, deciduous spines .... Spines rough, without sheaths; plants prostrate .... Joints flattened, not tuberculate. Joints spineless; flowers magenta; fruit dry, spine- TOSS: | 25-2204 Se ee Be pee Ree Re ee ee 6. O. basilaris Joints spiny; flowers yellow. Areoles 2-3 cm. apart; spines white or with brown base; fruit purple, juicy, spineless. Fruit 3-3.5 cm. long, with green pulp; slopes above Griffith’s mine at an elevation of 2450 . multigeniculata . Parishii Siler eke) MEET Si ia Gt ers se A a ae eee eae 7. O. charlestonensis Fruit 4-6 cm. long, with purple pulp; at eleva- tions of 1200-1350 meters ................. 8. O. phaeacantha Areoles about 1 cm. apart; spines white (occasion- ally brownish) ; fruit dry, spiny. Spines acicular, 3-5 cm. long and bristle-like, up : too D2cm longs ce ee ee ee: 9. O. erinacea Spines all acicular, stiff, not over 2-3 cm. long .. 10. O. polyacantha 1. Opuntia RAMosissiMa Engelm. Am. Jour. Sci. ser. 2, 14: 339. 1852. Mohave and Colorado deserts, from California, Nevada and Arizona south to Sonora. Local habitat, widely scattered in sandy or gravelly soil in the Larrea Belt. Best developed in sandy soil east of Wilson’s ranch where the plants are about one- half meter tall. Cottonwood Springs, altitude 1030 meters, Clokey 8036; E-Spear ranch, altitude 1640 meters, Train 1743; south of Indian Springs, altitude 1200 meters, Clokey 8026; east of Wilson’s ranch, altitude 1100 meters, Clokey 8437. May to August. 1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 69 2. OPUNTIA ACANTHOCARPA Engelm. & Bigel. Proc. Am. Acad. 3: 308. 1856. Mohave and Colorado deserts from California to southern Utah, south to Sonora. Local habitat, in sandy or gravelly soil in the Larrea Belt. Most abundant at Wilson’s ranch. Wilson’s ranch, altitude 1150 meters, Clokey S024, 8433. June. 3. OPUNTIA ECHINOcCARPA Engelm. & Bigel. Proc. Am. Acad. 3: 305. 1856. Mohave and Colorado deserts of California, east to Utah and Arizona, south to Lower California. Local habitat, local in the upper Larrea and lower Juniper belts. Most abundant in Kyle Canyon at an elevation of about 1600 meters. Kyle Canyon, Clokey 7202, 8025. June. 4. Opuntia multigeniculata Clokey sp. nov. E Sect. Cylin- dropuntia, humilis, subadscendens, habitu conferta, plus minusve 0.5 m. alta, ad 1.5 m. lata, ramorum ligno debili, reticulato; ramis majoribus depresso-tuberculatis, ca. 1.5—2 em. crassis, geniculis lateralibus numerosis, 3-5 cm. longis, ca. 2 cm. crassis, tuberculis confertis 4-6 mm. longis, 2-3 mm. latis, 4-5 mm. altis, apice areola pallide brunneo-lanosa, 4-5 mm. longa, 2—2.5 mm. lata coronatis; aculeis ca. 12, geniculum ipsum fere occultantibus, delicatis, vaginis albidis barbulatis; aculeis centralibus 2—4, 15-18 em. longis, lateralibus gracillimis; foliis teretibus apiculatis ca. 2 mm. longis; glochidiis ex areola summa albidis, 1.5 mm. longis; floribus in apice geniculi aggregatis, ca. 2.5 cm. longis totidemque latis; ovario conferte tuberculato, areolis prominentibus, ellip- soideis, lana pallide brunnea; aculeis paucis, pergracilibus, de- ciduis; perianthii laciniis pallide viridi-luteis, spatulatis, obtusis apiculatisve, 15-18 mm. longis; stylo cum stigmatibus luteis, stig- matum lobis 6-8; staminibus luteis; fructu globoso, luteo, sub- carnoso, ca. 2 cm. longo, tuberculato, exaculeato, profunde umbilicato; hypanthii cavo pulpa incolora farcto; seminibus rotundatis, pallide luteis, haud nitidis, levibus, 2.5-8 mm. di- ametientibus, commissura brevi, lata, manifesta. A low, semi-ascending, compact Cylindropuntia, 0.5 m. or less high, up to 1.5 m. wide, with stems having a weak, reticulated, woody framework; main stems low tuberculate, about 1.5-2 cm. thick, with crowded lateral joints 3—5 cm. long, about 2 cm. thick; lateral joints with closely placed tubercles, 4-6 mm. long, 2-3 mm. wide, 4-5 mm. high, with the entire upper end of the tuber- cle occupied by an areole 4-5 mm. long, 2—2.5 mm. wide, filled with light tan-colored wool; spines about 12, almost concealing the surface of the joints, delicate, white-sheathed, barbed; 2 to 4 central spines 15-18 mm. long; lateral spines very slender; leaves terete, apiculate, about 2 mm. long; glochids white, 1.5 mm. long, from the upper end of the areoles; flowers clustered at the tip 70 MADRONO [Vol. 7 of the joints, about 2.5 cm. long and broad; ovary closely tu- berculate; areoles prominent, oval, filled with light tan-colored wool; spines few, very delicate, deciduous; perianth-segments light greenish-yellow (Ridgway’s Col. Stand. and Nom. plate 5, 25 YG-—Y, b), spatulate, obtuse or apiculate, 15-18 mm. long; style and stigmas yellow; stigma lobes 6 to 8; stamens yellow; fruit globose, yellow, with somewhat fleshy walls, about 2 cm. high, tuberculate, spineless, deeply umbilicate; seed cavity filled with colorless jelly; seeds circular, light yellow, dull, smooth, 2.5-3 mm. in diameter; commissure short, broad, distinct. Definitely known only from the type locality on an open, rocky ridge east of Wilson’s ranch, Charleston Mountains, Clark County, Nevada, along road from Blue Diamond mill to the mine, at an elevation of 1400 meters; Clokey S430 (type), S639, 8760. Flower, May; fruit, July to September. Mr. Fred Gibson of the Boyce Thompson Southwestern Arbo- retum states that plants similar to this are found near Prescott, at Congress, and along the Colorado River in Arizona. These have not been examined. Opuntia multigeniculata is intermediate between the series Thurberianae and Echinocarpae. In the Echinocarpae it is closest to O. echinocarpa Engelm. & Bigel. It differs from this species in having weaker, less woody, shorter and more crowded stems. The fruit of O. echinocarpa is dry. The somewhat fleshy fruit and the seeds resemble O. Whipplet Engelm. & Bigel. The joints, however, are thicker, shorter, more crowded and more thickly armed than O. Whipplei. In the original description Engel- mann and Bigelow (Proc. Am. Acad. 3: 807. 1856), state that the color of the flowers of O. Whipple is red. Britton and Rose (Cact. 1: 55-56. 1919) and Benson and Thornber (Cact. Ariz. 36. 1940) give the color of the flowers as yellow. Neither O. echinocarpa nor O. Whipplet grow near the type locality of O. multigeniculata. 5. OPUNTIA PaRisHm Orcutt, West. Am. Sci. 10: 81. 1896. Mohave Desert of California and southern Nevada. Local habitat, locally abundant in sand or gravelly soil in the Larrea Belt at elevations of 1000 to 1200 meters. Cottonwood Springs, Clokey 8028; south of Indian Springs, Clokey 8027; east of Wil- son’s ranch, Clokey 8434. June. Drifting sand usually partially covers the old joints. This is a characteristic plant that could not be mistaken for any other Cylindropuntia in this region. EXPLANATION OF THE Ficures, PLATE 4. Puate 4, CAcTACEAE OF THE CHARLESTON Mountains, Nevapa. Fig. A, Opuntia multigeniculata Clokey. Fig. B, Opuntia acanthocarpa Engelm. and Bigel. Fig. C, Opuntia charlestonensis Clokey. Figure A Figure B Ficure C Puate 4, CACTACEAE OF THE CHARLESTON Mountrarns, NEVADA. ' at ma ‘ al eet i = t 2 my - « ie 4y P i ma lone : " i i ion h ' ' ; * i co on : < : ‘ ae * : \ ah N ' ‘ ‘ ( ' ' ? =) ~ 2 r , ' | ‘ . ‘ ‘ . 4 2 , ) ees) 7 4 ' . rl | sy ‘ \ \ ‘ ; \ mt: , : vn 1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 71 6. Opuntia BasiLaRis Engelm. & Bigel. Proc. Am. Acad. 3: 298. 1856. Colorado and Mohave Deserts and surrounding mountains from California to Utah, south to Sonora. Local habitat, widely scattered in the Larrea Belt below 2000 meters. Clark Canyon, Clokey & Anderson 7201; southeast Indian Springs, Train 1759; Kyle Canyon, Clokey 7205, 8032; Trout Creek Canyon, Clokey & Anderson 7206. May. 7. Opuntia charlestonensis Clokey sp. nov. Humilis, patens ad 0.5 m. alta, 1.5 m. lata, ramis primo adscendentibus demum prostratis, tum geniculis 2—3 junioribus tantum suberectis; genicu- lis apicalibus vel lateralibus, ovalibus vel obovatis, saepius apice rotundatis, 10-18 cm. longis, 10—12 em. latis, viridi-lutescentibus, primo vere rubentibus; areolis rotundatis vel ovatis, 5-6 mm. latis, 2-2.5 cm. distantibus; glochidiis 3-4 mm. longis, pallide brunneis in areolae apice; aculeis 4-6 acicularibus longitudine ludentibus, longissimis ad 4.5 em., subcomplanatis tortisque, undique vertentibus, albidis vel albidis basi pallide brunneis; floribus primo laete luteis demum conferte roseis rubro-suffusis, 4—6 cm. longis, ca. 4 cm. latis; perianthii laciniis externis ovatis, acutis, mucronatis, internis ovatis, apice rotundatis saepius mucro- natis; staminibus luteis; stylo cum stigmatibus luteis vel rubes- centibus; fructu ellipsoideo 3—3.5 cm. longo, 1.5—2 cm. crasso, sordide purpurascente, cortice externo pulpaque viridibus; se- minibus applanatis, 4-5 mm. diametientibus. A low spreading plant up to 0.5 m,. high, 1.5 m. wide, the main branches at first ascending, later prostrate with only the younger two to three joints ascending; joints arising from the faces as well as from the edges of older joints, oval or obovate, mostly rounded at tip, 10-18 cm. long, 10-12 cm. wide, yellowish-green, in the spring purplish; areoles circular or oval, 5-6 mm. across, 2—2.5 cm. apart; glochids in the upper part of the areoles, 3—4 mm. long, light brown;-spines four to six, acicular, varying in length, the longest up to 4.5 cm. long, somewhat flattened and twisted, spreading in all directions, white or white with light brown base; flowers clear yellow at first, turning salmon tinged with red, 4-6 cm. long, about 4 cm. wide; outer perianth-segments ovate, acute, mucronate; inner segments ovate, rounded at tip, usually mucronate; stamens yellow; style and stigmas yellow or reddish tinged; fruit oval, 3—-3.5 cm. long, 1.5—2 em. thick, dull reddish-purple, with green rind and pulp; seeds flat, 4-5 mm. in diameter, Known only from hillsides adjacent to Griffith’s mine, associ- ated with Pinus monophylla Torr. & Frém. and P. scopulorum (Engelm.) Lemmon, at an elevation of about 2450 meters, Clokey 7208, 7592, 8029 (type), 8688, 8770. Flower, July to August; fruit, September to October. 72 MADRONO [Vol. 7 As suggested by Dr. Ira L. Wiggins, O. charlestonensis is most closely related to O. megacarpa Griffiths, a plant of the western edges of the Mohave and Colorado deserts in California. O. megacarpa has joints 20-80 cm. long, fruit 7-12 cm. long and seeds 7-8 mm. in diameter. 8. OPUNTIA PHAEACANTHA Engelm. in Gray, Mem. Am. Acad. 4:52. 1849. Texas and Chihuahua to Arizona and the Charleston Moun- tains, Clark County, Nevada. Local habitat, among Quercus, Fraxinus and Amelanchier at Wilson’s ranch and in open wash on gravelly soil at the mouth of Pine Canyon. Wilson’s ranch, alti- tude 1180 meters, Clokey 8031, 8424, 8431, 8761; mouth of Pine Canyon, altitude 1350 meters, Clokey S656, 8685. Flower, May to June; fruit, July to August. The plants of O. phaeacantha Engelm. from the Charleston Mountains do not agree entirely with the original description but, except for the size of the fruit, fall well within the range included in that highly variable species by Britton and Rose (Cact. 1: 144. 1919), and Benson and Thornber (Cacti Ariz. 58. 1940). Bois- sevain and Davidson (Colo. Cacti. 12. 1940) give the size of the fruit as 4 to 6 centimeters in length, which agrees with the fruit of our plants. At Wilson’s ranch, where the joints are up to 20 centimeters long, O. phaeacantha grows among Quercus, Fraxinus and Amelanchier. This protection may well account for the un- usual size of the joints. The red color shown in winter and spring on the joints at Pine Canyon, but not at Wilson’s ranch, is evidently due to cold. Cuttings from the Pine Canyon station grown in South Pasadena remain green throughout the winter. 9. OpuNTIA ERINACEA Engelm. Proc. Am. Acad. 8: 301. 1856. Mohave Desert of California, east to Utah and Arizona. Local habitat, most abundant on a dry ridge east of Wilson's ranch along road from Blue Diamond mill to the mine, at an ele- vation of 1200 to 1400 meters. Mountain Springs, elevation 1700 meters, Clokey $423; ridge east of Wilson’s ranch, Clokey 8033, $436. June to July. 10. OpuNTIA PoLtyAcANTHA Haworth, Suppl. Succ. 82. 1819. North Dakota to Washington, south to Texas, Nevada and Arizona. Local habitat, occasional in the Juniper Belt. Harris Springs road, elevation 1800 meters, Clokey 7591; Lee Canyon, elevation 1800 meters, Clokey 7204; Willow Springs, elevation 1200 meters, Clokey S030. Key To THE SPECIES OF ECHINOCEREUS ENGELM. Plants forming large clusters with many heads; spines few, white or gray, long and flexuous; flowers scarlet, funnelform: 2262329 ee eee 1. E. mohavensis Plants with few heads, ascending; spines white to brown, stout, curved; flowers purple, broad ................ 2. E. Engelmannii 1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 73 1. EcutNocEereus MoHAVENSIS (Engelm. & Bigel.) Rimpler, Forst. Handb. Cact. 2: 803. 1885. Cereus mohavensis Engelm. & Bigel. in Engelm. Am. Acad. Arts and Sci. Proc. 3: 281. 1856. Southeastern California to southern Utah, northern Arizona and south to Mexico. Local habitat, occasional as single plants in the Juniper Belt extending in places to the Pinyon and Yellow Pine belts. Grows in gravelly soil and on rock ledges. Cathe- dral Rock, elevation 2425 meters, Clokey 7211; Clark Canyon, elevation 1800 meters, Clokey & Anderson 7200; Charleston Park, elevation 2300 meters, Clokey 5034, elevation 8000 feet, Alexander 755; Kyle Canyon, elevation 2425 meters, Clokey 7210, elevation 1700 to 2400 meters, Clokey 8435; Lee Canyon, elevation 2670 meters, Clokey, Clokey & Baker 7597. May, June. 2. EcuiINocerEeus ENGELMANNU (Parry) Riimpler, Forst. Handb. Cact. 2: 805. 1885. Cereus Engelmannu Parry ex Engelm. Am. Jour. Sci. ser. 2, 14: 338. 1852. Mohave and Colorado deserts of California, east to Utah and Arizona, south to Mexico. Local habitat, widely scattered in the Larrea and lower Juniper belts at elevations up to 1400 meters. Kyle Canyon, Clokey 7590; Trout Creek Fan, Clokey 7207. KEY TO THE SPECIES oF Ecuinocactrus MILL. Spines stout, flattened, annulate; flowers yellow. Stems globose, aggregate; axils of scales of ovary and UNTO WO OVI ee ect pee ae a wk Fe a ds Pda dN BG a 1. H. polycephalus Stems simple, globose to cylindrical; axils of scales on ovary and fruit not woolly ....................... 2. H. acanthodes Spines subulate, not annulate; scales on ovary and fruit few, without wool in the axils; flowers amaranth purple (Ridgway) to pink ...........2.<..-2s.0.54. 3. EH. Johnsonii 1, Ecurnocacrus potycerHatus Engelm. & Bigel. Proc. Am. Acad. 8: 276. 1856. Eastern Mohave Desert of California, east to Utah and Ari- zona, south to northern Sonora. Local habitat, sparsely scat- tered in gravelly soil or on rock ledges in the Larrea Belt. South of Indian Springs, elevation 1300 meters, Clokey 7598, elevation 1250 meters, Clokey S432. July. 2. Ecuinocactus acantHopes Lemaire, Cact. Gen. Nov. Sp. 106. 18389. Ferocactus acanthodes (Lemaire) Britton & Rose, Cact. 3: 129. 1922. Southern California to southern Nevada, Arizona and Lower California. Local habitat, on gravelly hills and rock ledges in the Larrea Belt at elevations of 1100 to 1400 meters. Most abun- dant on a ridge east of Wilson’s ranch along road from Blue Th MADRONO [Vol. 7 Diamond mill to the mine. South of Indian Springs, Clokey 7212, 7593; rocky ridge east of Wilson’s ranch, Clokey 8428, 8429. 3. Ecuinocactus JoHNsonut Parry in Engelm. Bot. King’s Geol? Expl 40th Par] 5:1 7s 21871: Ferocactus Johnsoni (Parry), Britt. & Rose, Cact. 3: 141. 1922. Southwestern Utah, northwestern Arizona, southern Nevada and extreme eastern Inyo County, California. Although Echino- cactus Johnsoni so far has not been found in the Charleston Mountains it is to be expected since it occurs both to the east and to the west of the range. It occurs very locally on hot, steep, gravelly slopes in the Larrea Belt and should be looked for on the lowest foothills. East of the range: Frenchman’s mine, 7 miles east of Las Vegas, elevation 630 meters, Clokey 5900; ridge south of Logandale, Clark County, elevation 550 meters, Clokey 5901. West of the range the type locality of Hchinocactus Johnsoni Parry var. octocentrus Coult. is in the mountains east of Resting Springs, Inyo County, California. The number of central spines is rather indefinite and variable even on the same plant. The spines graduate in thickness and length from the largest centrals to the shortest laterals so that there is no satisfactory line of demarcation. MammiLuaria Haw. 1. MAMMILLARIA TETRANCISTRA Engelm, Am. Jour. Sci. ser. 2, 14:°387. 1852. Phellosperma tetrancistra (Engelm.) Britt. & Rose, Cact. 4: 60. 19238. Southern Utah to Nevada, Arizona and southeastern Cali- fornia. Infrequent and isolated. Local habitat, known in the Charleston Mountains from a single specimen collected on a gravelly hillside in the Larrea Belt south of Indian Springs at an elevation of 1250 meters, Clokey 8037 (Clokey Herbarium). Key To THE SPECIES OF CoRYPHANTHA (ENGELM.) LEMAIRE Flowers 2-3 cm. long, straw-colored with pinkish midrib to tinged with pink throughout; fruit green; seeds brown; Larrea and Jui per WelES® . oo oe ose Ae oy tn oa Ae ee 1. C. deserti Flowers 3.5-4.5 cm. long, eugenia red to old rose (Ridgway) ; fruit more or less rose-colored; seeds reddish-brown; associ- ated with Pinus monophylla Torr. & Frem., P. scopulorum (Engelm.) Lemmon and Cercocarpus ledifolius Nutt. ...... 2. C. rosea EXPLANATION OF THE FiGuRES, PLATE 5. Puate 5. CAcracear oF THE CHARLESTON Mountarns, Nevapa. Figs. a, b, Opuntia acanthocarpa Engelm. and Bigel. Figs. c, d, Opuntia echinocarpa Engelm. and Bigel. Figs. e, f, g, Opuntia multigeniculata Clokey. Fig. h, Opuntia echinocarpa Engelm. and Bigel. Fig. i, Coryphantha rosea Clokey. PLATE 5. CACTACEAE OF THE CHARLESTON Mountatins, NEvADA. 1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 15 1. CorYPHANTHA DEsERTI (Engelm.) Britt. & Rose, Cact. 4: 46. 1923. Mammillaria deserti Engelm. Bot. Calif. 2: 449. 1880. Eastern Mohave Desert of California to southern Nevada and possibly northwestern Arizona. Local habitat, widely scattered on gravelly hills and in washes in the Larrea and lower Juniper belts at elevations of 1100 to 1900 meters. It is most abundant in the wash below Rocky Gap Springs. The flowers, which last several days, are open only in the afternoons. MHarris Springs road, Clokey 7794; Kyle Canyon at mouth, Clokey 7594; Kyle Can- yon at junction of Harris Springs road, Clokey 8425, 8426; Rocky Gap Springs, Clokey 8718. Jepson (FI. Calif. 2: 550. 1936) regards C. deserti as a syno- nym of C. chlorantha (Engelm.) Britt. & Rose and reports the flower color as yellow or stramineous. C. chlorantha (Mammil- laria chlorantha Engelm. in Rothr. Bot. Wheeler 127. 1878) was originally described as having yellow or greenish-yellow flowers and white stigma lobes. Engelmann described the petals of M. deserti (Bot. Calif. 2: 449. 1880) as light straw color turning to purplish at tips. On the examination of the flowers of hundreds of plants in the Charleston Mountains and over 300 dried flowers the writer failed to find a single flower without a distinct pink tinge at least on the midrib of the perianth-segments. This pink color was not confined to the tips but extended throughout the length of the petals. The flowers are 2.5—8 cm. long; the stigma lobes are yellow or partially tinged with pink. Several collec- tions of C. deserti in 1942 by J. P. Hester from southern Clark County, Nevada, and eastern San Bernardino County, California, one of the latter from near to or possibly from the type locality, are excellent matches for the Charleston Mountains plants. F. W. Gould 1589 and Hester Utah 1, both from the Beaverdam Moun- tains west of St. George, Utah, from near the type locality, agree with the original description of C. chlorantha. The flowers are 3.5—-4 em. long, the petals are yellow or greenish-yellow without a trace of pink, the stigma lobes are yellowish. Hester reports that the flowers of C. chlorantha are fragrant while those of C. deserti are not. From the study of this material it is evident that C. deserti should not be considered even as a variety of C. chlo- rantha. 2. Coryphantha rosea Clokey sp. nov. Globosa vel ovata, 4—15 em. alta, 4-10 cm. crassa, tota aculeis dense tecta; tuberculis cylindricis, 10-15 mm. longis, supra totis sulcatis, sulcis areo- lisque tenellis albido-lanosis; aculeis 28—35 acicularibus, centrali semper angulo recto insidente, 5-10 mm. longis, saepius apice rubentibus (raro nigricantibus), centrali in lateralibus aculeis colore crassitudine intermediis sensim transeunte; floribus in axillis tuberculorum ad plantae ipsius apicem rubris vel roseis, 3.5—4.5 em. longis, squamis tubi laciniisque externis e basi angus- 16 MADRONO [Vol. 7 tatis, acuminatis, longe fimbriatis; laciniis internis similibus neque fimbriatis; filamentis luteis; stylo cum stigmatibus luteis vel roseis, stigmatis lobis linearibus; fructu ovoideo, praesucculento, perianthii reliquiis siccis persistentibus, tarde maturante, plus minusve rubro-colorato, 2—2.5 cm. longo, 1.2—1.4 cm. crasso; semine rubro-brunneo, foveolato, 2 mm. longo, 1.25 mm. crasso. Globular to oval, simple, 4-15 em. high, 4-10 em. thick ; arma- ment dense, hiding the surface of the plant; tubercles cylindric, 10-15 mm. long, grooved through the length on the upper side, with tufts of white wool in the axils and areoles of the younger tubercles; spines 28 to 35, acicular, central permanently at right angles to the surface of the plant, 5-10 mm. long, more or less tipped with reddish-brown (rarely black), the central spines grading in thickness and depth of color to the laterals, on the upper, younger tubercles the spines are nearly erect, with age the spines lose their color and radiate almost at right angles to the tubercles; flowers borne in the axils of the young spiny tubercles forming a circle at the top of the plant, eugenia red to old rose (Ridgway’s Color Stand. and Nom. plate XIII, I’ red), 3.5—4.5 em. long; scales on the tube and outer perianth-segments tapering from the base, acuminate, long fimbriate ; inner perianth segments similar but not fimbriate; filaments yellow; style and stigmas yel- low or rose; stigma-lobes linear; fruit oval, very juicy, with the withered perianth persisting, ripening slowly, more or less colored with eugenia red or old rose, 2—2.5 cm. long, 1.2—1.4 em. thick; seeds reddish-brown, pitted, 2 mm. long, 1.25 mm. wide. Widely but sparsely scattered. Cold Creek, with Juniperus utahensis (Engelm.) Lemmon at 1850 meters, Clokey 8035; ridge south of Deer Creek, with Cercocarpus ledifolius Nutt. at 2650 meters, Clokey 7595, 7596; ridge along Lee Canyon with Juniperus utahensis (Engelm.) Lemmon at 1900 meters, Clokey 7209; Kyle Canyon to Deer Creek with C. ledifolius and Pinus monophylla Torr. & Frém. at 2400 meters, Clokey 7208, 8038 (type); Kyle Canyon with Pinus scopulorum (Engelm.) Lemmon at 2200 meters, fruit ripe, Clokey 8769. Flower, June; fruit, September, October. Coryphantha rosea is quite similar to C. deserti and C. chlorantha (Engelm.) Britt. and Rose in the appearance of the armament and in the shape and ciliation of the perianth-segments. It differs from C. chlorantha in the color of the flowers and fruit and from C. deserti in the size and color of the flowers and in the color of the fruit. It grows at higher elevations than C. chlorantha and C. deserti and, so far as known, is confined to the middle eleva- tions of the Charleston Mountains. No intermediate plants have been found. C. Alversonit (Coult.) Orcutt, a plant growing at lower elevations in extreme southeastern California, has darker, heavier spines and is reported to have white stigma-lobes and green fruit. South Pasadena, California, April, 1942. 1943] CRONQUIST: IDAHO PLANTS 17 NOTEWORTHY PLANTS FROM IDAHO. II ARTHUR CRONQUIST During the summer of 1941 Mrs. Cronquist and I spent two months collecting plants in central Idaho. A complete set is de- posited at the Missouri Botanical Garden, and nearly complete sets at the Gray Herbarium and the University of Idaho, Southern Branch; partial sets are deposited at the University of Minnesota and the Utah State Agricultural College. Several collections represent plants not previously known to occur in the state; a few are apparently undescribed. In the following discussion, unless otherwise indicated, the collection numbers are my own. There are in Idaho at least two entities that have commonly been referred to Melica bulbosa Geyer ex Port. and Coult. One, the usual form of the species, is single-stemmed or loosely cespi- tose, with obvious bulbs at the bases of the culms. The other is very densely cespitose, growing in large tufts, with the bases of the culms only slightly if at all bulbous. Although the degree of bulbousness of the culms is known to be a variable character in M. bulbosa, the appearance of the cespitose form is so striking that in the field it might well be taken for a different species. The clumps are very dense, and difficult to separate into speci- mens suitable for herbarium sheets. It may be significant that in 1899 Aven Nelson considered the habit of specimens he collected so remarkable that he noted on the labels sent out with them, “in dense tufts with matted roots.” Dr. W. S. Boyle, of the Uni- versity of California, who is making a cytogenetic study of the genus, writes me that the cespitose plants I sent him do not differ sufficiently from the usual form to warrant separate treatment. Since the two forms are decidedly different in appearance, and are readily distinguishable in the field, I prefer to regard them as two varieties of one species. MELIca BULBOSA Geyer var. typicanom. nov. M. bulbosa Geyer ex Port. and Coult. Syn. Fl. Colo. 149. 1874. MEtica BULBosA Geyer var. caespitosa var. nov. Planta caespi- tosa culmis ad basim haud vel vix bulbosis, cetera similis var. typicae. Type. Open rocky granitic slope above west side of Alturas Lake, Blaine County, Idaho, altitude 7200 feet, June 22, 1941, 2603 (Missouri Botanical Garden; isotypes, Utah State Agricul- tural College, University of Idaho, Southern Branch). Cotypes. Ipano. Hillside above Mill Creek, 12 miles west of Challis, Custer County, altitude 7800 feet, July 8, 1941, 2977; granitic outcrop above Wildhorse Creek, 25 miles southwest of Chilly, Custer County, altitude 7900 feet, July 28, 1941, 33854. Wvyomine. Teton Mountains, Uintah County, August 16, 1899, Nelson and Nelson 6540. 18 MADRONO [Vol. 7 Siranion HaNseni X AGROPYRON sPicaTtuM. On a rocky granitic hillside above Bull Trout Lake, Boise County, 35 miles west of Stanley at an altitude of 7000 feet, a colony of highly variable hordeaceous grasses was observed. A few spikes of the extreme of the robust plants were taken with the other plants of the same number (3615). A mile upstream and 200 feet higher on the same hillside, specimens of Agropyron spicatum (Pursh) Scribn. and Smith were seen and collected (3635). On examining my number 3615 in the herbarium, I find that the large spikes are quite uniform and represent Sitanion Hansent (Scribn.) J. G. Smith. Other specimens of this number are so extremely vari- able as to be abnormal, transcending the differences between Sitanion and Agropyron. Spikes with all the spikelets solitary and with all the spikelets paired at the nodes occur on the same plant. One culm branches at the summit into two variable spikes. Tests made on a few rachises indicate that they disarticulate, but only on considerable urging. Individual portions of some of the spikes are strongly reminiscent of Agropyron spicatum. It is my opinion that the specimens represent a natural hybrid between Sitanion Hanseni and Agropyron spicatum. Carex obovoidea sp. nov. Planta caespitosa, culmi 20—40 cm. alti, vaginae ventraliter rubro-maculatae, ad apices rubiginosae, inferae aphyllae; spiculae plures androgynae, coarctatae in ca- pitulum 15-30 mm. longum, 8-20 mm. latum; squamae late ovatae, castaneae, marginalibus perspicuae; utriculi 8-3.5 mm. longi, 1 mm. lati, plerumque attenuatis ab basibus spongiosi- turgidus, dorsaliter paucinervosi, ventraliter enervosi; achaenium obovatum, haud vel vix stipitatum, 1.5 mm. longum, minus quam 1 mm. latum; stigmata 2. Caespitose, 20-40 em. high, aphyllopodic; sheaths red-dotted ventrally and copper tinged at the mouths; spikes small, numer- ous in an ovoid or oblong head 15-30 mm. long, 8-20 mm. wide, sometimes slightly interrupted below, androgynous; scales broadly ovate, brown, with conspicuous hyaline margins; perigynium corky-thickened at the base, yellowish brown, changing to green upwards, or somewhat green-margined, 3—3.5 mm. long, 1 mm. wide, tapering from base to apex, or a little more strongly nar- rowed near the middle, prominently few-nerved dorsally, nerve- less ventrally, margins serrulate, beak obliquely cut and dark brown at the tip; achene lenticular, obovate, broadest above the middle, scarcely stipitate, 1.5 mm. long, less than 1 mm. wide; stigmas 2. Type. Mossy rocky meadow along tiny spring above Beaver Creek, near Marsh Creek, 25 miles northwest of Stanley, Custer County, Idaho, altitude 6500 feet, July 3, 1941, 2872 (Missouri Botanical Garden; isotypes at University of Idaho, Southern Branch and Utah State Agricultural College). 1943] CRONQUIST: IDAHO PLANTS 79 This species is a member of the small section Vulpinae, as treated by Mackenzie (1). It differs from the related C. neu- rophora Mack. and C. nervina Bailey in having the perigynia ven- trally nerveless, and the sheaths ventrally red-dotted and copper- colored at the mouths. It differs from C. Cusicki Mack., of the related section Paniculatae, in its shorter stature, more contracted inflorescence, spongy-thickened perigynium-base, and tapering rather than abruptly contracted perigynium. It differs from all of these in its obovate achenes. Astragalus reclinatus sp. nov. Planta perennis prostrata, caulibus 1 mm. crassis ad 50 cm. longis plus minusve strigosis ; folia similiter strigosa foliolis 1-5 (plerumque 3) foliolo extremo 20-30 mm. longo, 8-5 mm. lato, eis lateralibus reductis vel de- fectis, stipulae virides parvulae 2—3 mm. longae, flores albidi-pur- purei 12 mm. longi alis rostrum superantibus calyce 6 mm. longo, fructus sessilis 15-17 mm. longus, 3—4 mm. latus albido-strigosus suturis prominentibus haud intrusis. Stems several from a perennial tap root, prostrate, spreading, up to 50 cm. long or more, about 1 mm. thick, green, striate, more or less pubescent with thick, short, nearly or quite basifixed hairs; leaves similarly pubescent, scattered, short-petiolate, with 1-5 (commonly 3) linear to narrowly lanceolate leaflets, the terminal one enlarged, mostly 20-30 mm. long and 38-5 mm. wide, tapering and acute at both ends, the lateral ones commonly somewhat re- duced, mostly 10-20 mm. long and 1-3 mm. wide, sometimes absent; stipules tiny, green, triangular or ovate, free, 2-3 mm. long; flowers whitish, with a trace of purple, mostly 2—10 in short racemes; banner 12 mm. long, abruptly bent nearly at right angles 5 mm. from the base, the expanded portion 9 mm. long, 8 mm. wide and suborbicular when spread out; wings about 4 mm. wide and 11 mm. long, including the 5 mm. claw; keel 9 mm. long, 4 mm. high at the end, blunt; calyx with short black and white hairs intermingled in varying proportions, the tube 3.5—4 mm. long, the teeth triangular and about 2 mm. long; pod sessile, I-celled, mostly 15-17 mm. long and 3-4 mm. wide, little-com- pressed, closely strigose with short white hairs, sutures prominent and not at all intruded. Type. Moist alkaline bottoms along roadside two miles south of Dickey, Custer County, Idaho, altitude 6300 feet, July 14, 1941, 3086 (University of Minnesota; isotypes at Missouri Bo- tanical Garden, Utah State Agricultural College and University of Idaho, Southern Branch). This species of the section Homalobus is related to A. flexuosus Dougl. ex G. Don and the polymorphic A. decumbens (Nutt. ex T. & G.) Gray. The latter is common in the general area from which the type of A. reclinatus was taken, but was not seen in the immediate vicinity, nor have I ever seen it in the same type of 80 MADRONO [Vol. 7 habitat. 4d. fleruosus, predominantly a species of the great plains, is not known to occur in Idaho. The most conspicuous difference between A. reclinatus and its close relatives is its prostrate habit. Probably more important is the type of leaf. In A. flexuosus the lateral leaflets are several to numerous and well developed. The leaves of A. decumbens are highly variable, the leaflets often being in part reduced to phyl- lodia, but the terminal leaflet is affected as much as the others, and variation in that direction connects the species with the closely related A. diversifolius Gray, a rush-like plant in which the leaves are reduced to naked or nearly naked narrow rachises. In A. reclinatus the tendency is for the terminal leaflet to be enlarged at the expense of the others. None of the leaves on my speci- mens have more than five leaflets, and some are quite simple. The flower, although superficially resembling that of the larger-flowered types of A. decumbens (“A. hylophilus’ for example), shows its own distinctive differences. In A. decumbens the wings scarcely surpass the keel; in d. reclinatus the wings obviously surpass the keel, though not so prominently as in A. flexuosus. The upward widening of the pods commonly found in A. decumbens is inconspicuous or absent in A. reclinatus. ‘The cross section of the pod is less compressed than in A. decumbens, and a little more so than in A. fleruosus. The stipules in A. recli- natus are tiny and green; in dA. fleruosus and A. decumbens they are larger, mostly 5 mm. long or more, and somewhat scarious. Although A. reclinatus is represented by only one collection, its distinguishing features, extending from technical characters of the flower to leaf type and habit, are so pronounced that I see little room for doubt as to the validity of the species. GENTIANA cCaLycosa Griseb. subsp. asepaLaA Maguire. Madrojfio 6: 151. 1942. G. idahoensis Gandoger, Bull. Bot. Soc. Fr. 65: 60. 1918. This recently described entity is represented in my collections by the following numbers: 3640, 3681, 3756, and 3757. Prior to the publication of Maguire’s paper (2), this material had been identified as G. idahoensis Gandoger, the type of which was col- lected by Evermann at Pettit Lake, Blaine County, Idaho. At that time I doubted that it was more than subspecifically distinct from G. calycosa. It is now evident that G. idahoensis should be added to the synonymy of G. calycosa subsp. asepala. Agastache Cusickii (Greenm.) Heller was described from the Steens Mountains, Oregon, and is known to Peck (3) only from the Steens Mountains. An isotype is in the University of Minne- sota herbarium. oO Doe 1943 | CRONQUIST: THALICTRUM POLYCARPUM 83 THE LEGITIMACY OF THALICTRUM POLY- CARPUM WATS. ARTHUR CRONQUIST In a recent article in Madrofio (7: 1-4. 1943), Croizat main- tains that the name Thalictrum polycarpum Loret (1859), although illegitimate and not validly published, renders 7. polycarpum Wats. (1879) illegitimate, as a later homonym. He cites article 61 of the Rules, in part, as follows: “Even if the earlier homonym is illegitimate, or is generally regarded as a synonym on taxo- nomic grounds, the later homonym must be rejected.” Croizat admits that T. polycarpum Loret was not validly pub- lished. The part of article 61 which he does not quote, states: “A name of a taxonomic group is illegitimate and must be re- jected if it is a later homonym, that is, if it duplicates a name previously and validly published for a group of the same rank based on a different type.’ (Italics mine.) The Rules are clear enough; a name must be validly published if it is to prevent the later use of the same name for a different plant. To further clinch the case, article 19 states: “A name of a taxonomic group has no status under the Rules, and no claim to recognition by botanists, unless it is validly published.”’ It seems clear that the name T’.. polycarpum Loret, since it was not validly published, cannot illegitimize T. polycarpum Wats. University of Minnesota, Minneapolis, Minnesota, February, 1943. VALID AND LEGITIMATE NAMES—AND THALICTRUM POLYCARPUM S. WATS. C. A. WEATHERBY To devise a set of rules which shall deal adequately with the almost endlessly various nomenclatural situations which exist or may arise is an exceedingly difficult task. It is therefore not sur- prising that, in the eight years since the publication of the current Rules of Botanical Nomenclature, a number of cases have been brought to light which are not directly covered by the rules or in which the application of the rules is doubtful. It is not an un- reasonable hope that, through the exercise of that faculty mis- called common sense, through carefully considered discussion of questions actually arising in the course of taxonomic investiga- tion, through decisions by a competent tribunal and cautious amendment of the rules where really necessary, these doubtful points gradually can be made clear. Their solution has not been helped by much of the theoretical discussion of nomenclature for 84 MADRONO [Vol. 7 its own sake published since 1935, especially when it raises illusory difficulties for the bewilderment of the unwary. Unhappily, Dr. Croizat’s discussion of the name of Thalictrum polycarpum S. Wats. (Madrofio 7: 1-4. 1943), must be classed with those which obscure rather than elucidate their subject. There is no real ambiguity in the use of the terms valid and legiti- mate and their opposites in the rules. A valid name is one pub- lished according to the prescriptions of articles 86 to 45 and otherwise in accordance with the rules (art. 16, note). A valid name is ipso facto legitimate; that term, however, is reserved for requirements other than publication. An illegitimate name is one which, though validly published, fails to meet other require- ments of the rules (arts. 60 to 69). A fully invalid name is like the last state of the wicked in Buddhist theology—it simply does not exist. Being, for nomenclatural purposes, non-existent, the rules pay no attention to it. Any confusion which Dr. Croizat detects arises, not from the rules, but from his own unnecessary attempt to redefine the above terms. The first sentence of article 61, the most important for the present question, has entirely eluded him. It reads: ““A name of a taxonomic group is illegitimate and must be rejected if it is a later homonym, that is if it duplicates a name previously and validly published [italics here are mine] for a group of the same rank based on a different type.” That is, the earlier homonym must have been validly published in order to bring about rejection of the later one. According to the rule voted at Amsterdam, a provisional name is not validly published; therefore it cannot be a bar to the later independent use of the same name. The rule as to provisional names was formulated in the course of debate at the Amsterdam Congress, as a substitute for a pro- posal to outlaw them originally made at the preceding Cambridge Congress and there referred to the Executive Committee (which took no action). As published in the Journal of Botany (74: 75. 1986), it carries no definition of nomina provisoria (invalid) and nomina alternativa (valid) and no adequate examples.’ Until it is perfected there may be doubt as to whether a given name is pro- visional or not. Had Dr. Croizat argued that such doubt existed in regard to Loret’s names and that it was therefore safer to employ Greene’s name for the Californian Thalictrum rather than Watson’s, he would have been on firmer, though still highly un- certain, ground and his argument would have been more to the point. For the status of Loret’s names is, in the light of article 61, the only consideration of any importance in determining the correct name for the species. If Thalictrum polycarpum Loret is provisional, then T. polycarpum S. Wats. is the legitimate name 1 The reference to the original proposal does not help; there the definition was drawn and the examples presumably selected to cover both kinds of names. 1943] ISENBERG: REDWOOD BARK 85 and must be used. If JT. polycarpum Loret is validly published, then T. polycarpum Wats. becomes a later homonym and T. ametrum Greene (or some other synonym; I do not know the tax- onomy) must be taken up. In view of the imperfect elaboration of the rule as to pro- visional names, the only safe course seems to be to admit as validly published under it only names which conform closely to the example from de Wildeman given. He had a plant which he unreservedly accepted as a new species, but he was in doubt as to the genus to which he should refer it, and called it both Cymbo- pogon Bequaerti and Andropogon Bequaerti. According to the rule, both these names are validly published. But if de Wildeman had written: “I do not feel justified in describing this as a new spe- cies; but if I were to do so, I should call it either Cymbopogon Bequaerti or Andropogon Bequaertiv’—that is, if he had not defi- nitely accepted the species*—then, as I see it, both names would be provisional and not validly published. The hypothetical quotation is very near what Loret actually did. On the above basis, then, 7’. polycarpum Loret is a pro- visional name, not validly published and incapable of functioning as an earlier homonym, and T. polycarpum S. Wats. is the correct name for the species. Gray Herbarium, Harvard University, February, 1943. THE ANATOMY OF REDWOOD BARK Irvinc H. IsenBeRG This paper is one of a series originating from the laboratories of The Institute of Paper Chemistry, Appleton, Wisconsin, and covering a fundamental study of the botanical, chemical and other characteristics of the California redwood; this work has been sponsored by The Pacific Lumber Company, Scotia, Cali- fornia. The present paper is a portion of a botanical study made by the author at the Institute in 1939. Although considerable data have appeared in the literature on the anatomy of the wood of Sequoia sempervirens (see J. N. Mitchell, Jour. Forestry 34: 988-93. 19386 for discussion and list of references) the only published work on the anatomy of redwood bark seems to be that included in the article by Abbe and Crafts, who examined the phloem of white pine and other coniferous species (L. B. Abbe and A. S. Crafts, Bot. Gaz. 100: 695-722. 1939). Their studies 2 On this point—the only definition of a provisional name we have—see Int. Bot. Congress. 1930. Nomenclature. Prop. Brit. Botanists 16 (art. 44); Rec. Synopt. 41 (art. 37 ter). 86 MADRONO [Vol. 7 were apparently confined to the inner bark and stressed matura- tion of the sieve cells. The bark of a merchantable redwood tree, Sequoia semper- virens (Lambert) Endlicher, is composed of secondary phloem, living and dead, and of periderm. The redwood is a thick-barked tree; the bark is sometimes as much as a foot thick but is usually much less. In external appearance it is reddish, deeply fur- rowed, and scaly. In transverse section the bark appears as two strikingly different colored rings—the very thin, whitish, inner one and the thicker, reddish-brown, outer one. The light-colored layer, which may have a pinkish tinge, rarely exceeds one quarter of an inch in thickness. Primary Bopy Although the primary bark tissues disappear early in the life of a redwood tree, a brief review of their structure seems to be in order before proceeding to a more detailed study of the secondary phloem. The primary body is self-sufficient and contains all the fundamental tissues and body parts. A cross section of a red- wood stem at the end of the first growing season shows (pl. 6) the pith, primary xylem, cambium, primary phloem, cortex, and epidermis, as well as the secondary xylem and phloem. An accessory protective layer, the hypodermis, is located just be- neath the epidermal layer. A few longitudinal resin canals are also evident in the cortex in plate 6. The primary tissues lying outside the cambium are pushed outward by the development of secondary tissues. The increase in circumference to which these tissues must accommodate them- selves quickly surpasses their ability to respond, with the result that, sooner or later, they are crushed or ruptured and killed by exposure, and especially by the stoppage of food and water sup- plies by the cork layers which develop within them. Obviously, the secondary phloem is very important functionally to the tree, because it soon replaces the primary phloem. SrconpaRyY PHLOEM The secondary phloem of redwood contains four types of cells—sieve cell, longitudinal parenchyma, fiber, and ray paren- ~ chyma (pl. 8). The sieve cell is the cell characteristic of secondary phloem from the viewpoint both of structure and function. In some specimens of redwood bark examined, the sieve cells comprised the majority of the cells, both in the inner living and in the outer dead bark; in others, there occurred greater amounts of phloem parenchyma. Similar to other gymnosperms, the sieve cell ele- ments are not arranged in series, end to end, forming definite con- ducting lines, but are separate and distinct. The sieve plates are scattered irregularly on the radial walls of the sieve cell ele- Puate 6. SeQUOIA SEMPERVIRENS. Cross section of one year old stem. x 73. Priate 7. SEQUOIA SEMPERVIRENS. Cross section of inner and small part of outer bark. X20.