, ce ' ‘ t ' ' to a to4 eG ‘ 4 4 ) 7 y wy ‘ j , ' “t hath bade palo Maya tditatdtage fa) - 3 serve seal annhe ye | aired state jaysildielas haired eh diag’ iypidhan he feby ae | ' 3 Aajde ote bet fae fie | . ale | : sihaii 4 . | ‘ehete haijerttatig: bite ots | | : jarbstanitations tanger’ 4 aly phyartten atte ase ae ‘ sviatan ed . 1 Ags ay Pana. 7 pi sy silete ms | i ips fee i) pire ae ; ; af OLE Nils ‘ ‘ : oy ae Y . hoe ; oh (ae : aah mite i aie whe beg ale! | ; . . Pact ; utils i he . - : vet [ Ht 1 ' it \¢ Ve ies tat oa . . apt ten Malt Mtl Sis , | . rere eres iayan . “ ' a | | | ANAT ee 4s. | Natage' , . Satrarte, ree | Bay aiteS{9oree | a ed tleere cies ' i 1 ' : wij t | . Poth petela tale ' Fat 4 a nf nod & FV foagy £59 te" ! . ty ae pi a ede tghdts ' ‘ . ‘ | $ | | | | We a 4 sna gg . ee | say lees mony ‘’ oy 1 ou ¥ oat ‘ : ra . nite Vari : 2 ah sts i pe 7 i “ Teydyaets i ioeat oto4 I ty i 4 Peer erte hs ait Ha . 4s y | «hasay, ahh 9) a care etd Np | . ee ee “is - ive iene ag 3 ; ja itiotiy ay : its ral eH rh wi dui ia foiy fen 5,4 a 7 ae: feat ica Be fagnjulsa H ag ia 4 Lune Spies ire) Vv. 4 \ MADRONO A WEST AMERICAN JOURNAL OF BOTANY VOLUME VI 1941 - - 1942 ansonian NStitys: OSS Meet Be 2 "% O ab Ls RS - National Mus 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 Herszert L. Mason, University of California, Berkeley, Chairman. LeRoy Asrams, Stanford University, California. Epcar ANDERSON, Missouri Botanical Garden, St. Louis. Lyman Benson, University of Arizona, Tucson. Herpert F. Corpetann, Sacramento Junior College, Sacramento, California. Ivan M. Jounston, Arnold Arboretum, Jamaica Plain, Massachusetts. Mivprep E. Marnias, University of California, Berkeley. Bassett Macuire, Utah State Agricultural College, Logan. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board EruHe.t Crum, Department of Botany, University of California, Berkeley. Business Manager WittiAm Hiesey, Carnegie Institution of Washington, Stanford University, California. (09% ‘d 908) ADTIVWVY SIONVUT ‘ONOUAV] “9 DUINTO A 0} dddICsTQUOA ST CONTENTS Some Details of the Reproductive Structures of Sarcodes. Bernice HE. Doyel and L. Marguerite Goss A Study of Isoetes in San Diego County, California | Lowis C. Erickson A New Frasera from Oregon .. Morton EH. Peck and Hlmer I. Applegate New Plants from ‘Oregon... .:. ... i... 00.524 .. Morton EH. Peck Field Characters Distinguishing Pinus ponderosa and Pinus Jeffreyi. Kenneth E. Bradshaw A New Species of Astragalus from Arizona .............. C. L. Porter MiberbulvaGGini SO Weetsenns oa. ..6 4c 8 cuaw Waa dal LeRoy E. Detling A New Species of Paronychia from Mexico ............ .... Harl L. Core Great Basin Plants—III. Caryophyllaceae......._.. .. Bassett Maguire A New Limnanthes from Oregon ........................ LeRoy Abrams RUCVAC WiSHee eee, a tht Gan os tte athe 4 Gab Racha aoknd 28, 58, 90, 142, 206, IN@EES BP aAndeyNG Wii.) 2 kite ae eos sas bo ale Ue ee eb eed 30, 63, 94, 144, 173, Marlarionyimeyucca Whipplel 22.0. 20.5 cus 8S eee Sheek tae Lee Haines Certain North and South American Distributions in | Scirpus. Alan A. Beetle The Problem of Life Zones on Mount Shasta, California. William Bridge Cooke A New Species of Lotus from the Mount Hamilton Range, California. Helen Kk. Sharsmith Proceedings of the California Botanical Society ....... 64, 96, 175, 208, Alpine Flora of San Francisco Mountain, Arizona .... Hlbert L. Little, Jr. Mary Fisk Spencer ..... Frederick Grover New Species of Vascular Plants from the Northw est Coast. George Neville Jones Noteworthy, Plants from Idaho... ....... 62.684 64.5 8.65 Arthur Cronquist MiaskarCedaranm California . oo... 5. ec eee eh te ee Herbert L. Mason Further Studies on Monotropoideae................ Herbert F. Copeland Podocarpus ¢gracilior in Cultivation .. ............4...%. John T. Buchholz The Taxonomic Status of Microsteris Greene ......... Herbert L. Mason The Holacanthoid Plants of North America . _. . Cornelius H. Muller Combinations Proposed in “The Higher Plants of Oregon.” i Morton EH. Peck An Undescribed Species of Stipa from California. G. L. Stebbins, Jr., and R. M. Love A Synopsis of the American Species of Cicuta. Mildred EH. Mathias and Lincoln Constance Great Basin Plants—VI. Notes on Gentiana.......... Bassett Maguire mMrmicauimonlaska and YUKON of. eda ad Palen cee is Bassett Maguire A Comparison of the Embryogeny of Picea and Abies John T. Buchholz Anatomy and Ecology of Ammophila arenaria Link |... . Edith A. Purer An Undescribed Species of Ceanothus from California. Howard E. McMinn Development of the Female Gametophyte in Erythronium helenae and Erythronium tuolumnense .. : Marion S. Cave Some Chemical Properties of Eucalyptus in Relation to Their Evolutionary SLU SMR nite Aare Ath ape MWR Ns puhaid tothcne ng 6 James B. McNair Grassland and Related Vegetation in Northern Mexico Forrest Shreve (CIE OZ SRT Ma ED et POA a eR I’, P. Cronemiller Notes on Polemoniaceae . ._.................. Herbert L. Mason Eburophyton Heller: A Valid Genus of the Orchidaceae. Louis O. Williams : The Significance of Certain Plant Names — .. Carl Sumner Knopf ‘ Notes on the Flora of the Charleston Mountains, Clark County, Nevada. WV Pe SUAS AUST ee oh Se hake ako! Seale sells. Ira W. Clokey : The Type Locality of Polystichum Lemmoni -Underw ood ~=6@ Harold St. John ‘ i 265 239 33 239 Kar Western Noveltiessin, Salixv. net la}, -« Weanotnus PTOStTatus .....6. 6. cst ee eee ewe Ceanothus velutinus Cercocarpus ledifolius Chamaebatiaria millefolium Chrysothamnus Bloomeri var. angustatus Chrysothamnus nauseosus var. occidentalis Cornus californica Cornus Nuttallii Juniperus communis var. montana ............... Kalmia polifolia Libocedrus decurrens Lutkea pectinata eo ee we eiecewisugs. Berle? é2ue: ‘i 16 ‘ellie minimums 255.2) sckeee eee H Roly sonum, Parnyl 4 1, V4 We ¢ WN ff iat) Be Ne ff ~~ Gsfrarar ary =~ - ~~ é NAR i] sf ~~ 2 a ey ened Lee pues aD get gas, sarees TS MADRONO A WEST AMERICAN JOURNAL OF BOTANY Contents URTHER STUDIES ON MonotropoweEakE, Herbert F. Copeland ............. 97 DOCARPUS GRACILIOR IN CULTIVATION, John T. Buchholz .................. 119 Tue Taxonomic Status or Microstreris GREENE, Herbert L. Mason ..... ™. 122 its HoracantHow Puants or NortuH America, Cornelius H. Muller ..... 128 Pret CR? Oe RAM nese te etn, 2 be eta tatiana Baaame yas oe tak pee, "133 R. Mm. JSG EIS SOO SOA eR Noe SAE CAR ML Ht LOSME eS Anup in CAT) ois Lees 137 "B Rueelaid): ; Fortunato L. Flemera sa de la Flora del Cuzeo (J. F. Macbride) ; Ivar Tidestrom and Sister Teresita Kittell, 4. Flora of Arizona and New Mexico (Herbert L. Mason) ..... LIRA Nga aA 142 " Published at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania October, 1941 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. Jounston, Arnold Arboretum, Jamaica Plain, Massachusetts. Miuxprep E. Maruias, University of California, Berkeley. Bassett Macume, Utah State Agricultural College, Logan. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board—EtTuHeEt Crum Department of Botany, University of California, Berkeley Business Manager—Wiuti1aAm 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. Volume I, Numbers 1 to 17, complete, $5.00. Volume II, Numbers 1 to 17, complete, $5.00. Volume an Numbers 1 to 8, complete, $5.00. Single num- bers $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 $3.70; 100 copies $4.10; additional 100’s .75¢; 8 pages, 50 copies $5.40; 100 copies $6.00, additional 100’s $1.20. Covers, 50 for $2.50; additional covers at $1.50 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: Ernest B. Babcock, University of California, Berkeley. First Vice-President: Roxana S. Ferris, Stanford University, California. Second Vice-President: Palmer Stockwell, Institute of Forest Genetics, Placerville, California. Treasurer: Wiiliam Hiesey, Carnegie Institution of Washington, Stanford University, California. Secretary: Lincoln Constance, Department of Botany, 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. 1941] COPELAND: MONOTROPOIDEAE 97 FURTHER STUDIES ON MONOTROPOIDEAE Hersert F. CopeLtanp This paper continues a series (5, 6, 7, 8, 9), of which the account of Sarcodes by Doyel and Goss (12) is to be considered a unit. It records observations on Pterospora, Hypopitys, and Monotropa, and adds to the previous account of Hemitomes (New- berrya). : It has been a pleasure to acknowledge from time to time the unstinted cooperation of various institutions and individual cor- respondents. The following have facilitated the preparation of the present contribution: the Herbarium of the University of California; the Dudley Herbarium of Stanford University; the Herbarium of the California Academy of Sciences; Dr. W. L. Jepson; Mr. Willman Spawn; Dr. W. H. Camp; Dr. P. L. Zimmer- man; and the Oregon Biological Supply Company. PTEROSPORA ANDROMEDEA Nuttall Pterospora andromedea was described by Nuttall (25) as col- lected “In Upper Canada, near the Falls of Niagara. Mr. C. Whitlow.” There has been essentially no nomenclatorial confu- sion as to this plant; no segregation of species or varieties has been proposed; Small (28) cites a single obscure synonym. The plant is common in the mountain ranges of western North Amer- ica; rare eastward to the region of the type locality. The material studied was collected from time to time at Jones- ville, Butte County, California, at an altitude of about five thou- sand feet. There the plant shares the habitat of Sarcodes and Pleuricospora, in forests of fir (Abies concolor). As compared with its congeners, Pterospora emerges from the ground and flowers noticeably later in the season; it is usually in full anthesis late in July. The shoots come up from globular masses of roots. It is noticeable that whereas they emerge in the neighborhood of dead shoots of the previous year, they are not in immediate contact with such dead shoots, nor at a distance to be measured in centi- meters, but a meter or more away. Jepson (19) believes that the plant is monocarpic, and it is probable that he is correct. There is, however, the possibility that the scattered shoots come up from long roots which have emerged from the masses. This is sug- gested by the behavior of the generality of monotropoid plants, in which the genetic individuals are polycarpic, forming shoots year after year as adventitious buds on the same root system. A positive determination of this matter could be attained only by determining the course of individual roots in a mass of humus, and I have not undertaken it. ManroXo, Vol. 6, pp. 97-144. October 15, 1941. a ‘eo a aus 98 MADRONO [Vol.6 The standard accounts of the gross structure of the shoot, as by Jepson and Small, leave essentially nothing to add. The tall and slender stems, densely glandular, are of a light purplish red color. The greater part of each shoot is a rather lax bracted raceme, usually but not always exhibiting orthodox phyllotaxy. The recurved glandular pedicels bear no bractlets. The five glandular sepals are separate. The five petals form a glabrous sympetalous urceolate corolla, yellowish in color, becoming char- taceous in age. Each of the ten stamens bears two horns on the back of the anther. The globular ovary is belted at the base by a nectary from which ten evenly spaced lobes project between the bases of the filaments. Internally, the ovary is five-chambered below, one-chambered above, filled by massive placentae bearing numerous ovules. The stigma is obscurely five-lobed, the lobes opposite the petals, that is, at the ends of the carpels. MacDougal has described the anatomy of the vegetative parts. The stem includes a cylinder of bundles so closely packed as almost to be continuous; around this there is a well developed continuous sheath of fibers. The vascular supply of the flower (pl. 8, fig. 1) enters the re- ceptacle as acylinder. From this there depart radially, first, five sepal bundles, and then, alternating with them, five petal bundles. There are no gaps above these bundles. Five stamen bundles arise as branches from the upper sides of the petal bundles. Above the departure of the petal bundles the stele breaks up into a ring of ten bundles. Of these, the five which are opposite the sepals bend outward and fork periclinally; the outer branches constitute the supply to the five remaining stamens; the inner ascend the ovary wall in the planes of the septa, and are to be interpreted as fused pairs of lateral bundles of adjacent carpels. The five remaining bundles enter the ovary in the planes of the locules, being the planes of the petals and carpels; each of them is a fused pair of ventral bundles of a single carpel. They branch out into the placentae and disappear. In several features the vascular system just described is peculiar. No other monotro- poid plant is known to be without carpel dorsals (though they are not well differentiated from carpel laterals in Sarcodes) ; in none do the placental bundles lie in the planes of the petals. Here it is as though the proper placental bundles had swung outward into the ovary wall, and the carpel dorsals inward, into the placentae. In most plants of the group, the style is supplied by the carpel dorsals, which ascend in the thin bands of tissue between the ridges projecting into the style channel. Sarcodes is exceptional in that the placental bundles supply the style, ascending within the ridges. In Pterospora, lacking carpel dorsals and having the placentals in a peculiar position, the style is without vascular tissue. As is known, the anther (pl. 8, figs. 2-7) projects horizontally toward the style from the summit of the filament, and bears two 1941] COPELAND: MONOTROPOIDEAE 99 horns, inserted respectively on the two sides of the insertion of the filament and projecting toward its base. A vascular bundle runs from the summit of the filament to the inward end of the anther; by this it is known that the inward end is distal, the end bearing the horns proximal, the upper side dorsal, and the lower side ventral. There are four parallel horizontal pollen sacs; the dorsal ones are larger than the ventral. A cross section shows the epidermal cells extended into conical points on the dorsal and ventral, but not the lateral, surfaces. Between the epidermis and the tapetum there are some two or three layers of wall cells; near the horns a few of these cells develop reticulate lignified thicken- ings, as in the endothecium of a typical flowering plant. A similar structure has been reported in Sarcodes, and will be re- ported below in Monotropa; it is less extensive in Pterospora than in these other genera, being apparently merely a relic structure. The cells of the tapetum become binucleate. The pollen grains are four-grooved. Dehiscence of the anther begins by two vertical slits, each of which crosses the proximal ends of the two pollen sacs of one lobe of the anther. Formation of these slits amounts to the same thing as the dehiscence of the anthers of Sarcodes or Monotropsis, being, as I take it, the typical dehiscence of the anthers of Eri- cales. Subsequently two secondary slits develop from the middles and at right angles to the primary ones, that is, in the plane between the two pollen sacs of each lobe of the anther. By a general shrivelling of the walls of the anther, accompanied by growth of the connective, these secondary slits gape widely; it is through them that most of the pollen escapes. Drude’s (138) figure of the fully dehisced anther is accurate, though it gives a false impression that the connective becomes torn in a horizontal plane from proximal to distal. The internal surface of the ovary wall is covered by two layers of well developed fiber-like cells. A complete series of stages of the development of the seed has not been seen, and there is nothing to add to my earlier notes (4). The embryogeny appears to be altogether typical of the group. The integument is of two layers of cells. The wing on the seed is of two layers of cells. It commences to form, by proliferation of the epidermis of the integument at the chalazal end, before the embryo sac is fully developed. Hyporitys Monortropa Crantz Only two species of monotropoid plants were known to Lin- naeus (23). One, known in English as the pine-sap, occurs on all continents of the north temperate zone; the other, the Indian pipe, occurs in North America and eastern Asia, but not in western Asia and Europe. The oldest Latin designation of the pine-sap seems to be Orobanche quae hypopithis dici potest, of Bauhin (1671). It is of course no Orobanche; Tournefort (1706) 100 MADRONO [Vol. 6 gave it as name the adjective Orobanchoides ; Dillenius (1719) took up the substantive designation recorded by Bauhin, as Hypopitys. The Indian pipe, first recorded by Plukenet (1671) under Oro- banche, was named Monotropa by Gronovius (about 1740). Lin- naeus included both species in one genus, for which he used the name Monotropa, the species becoming respectively M. Hypopithys . and M. uniflora. The breach of priority as to the generic name was immediately protested by Hill (16): “Linnaeus takes away its received name hypopitys and calls it monotropa.” Because he was the first after 1753 to use Hypopitys as the name of a genus, Hill is cited as authority for it; this in spite of the facts, that he did not originate it, and that he did not use binomials and cannot be cited as authority for any of the species. Binomials under Hypopitys' were first made by Crantz (10), the pine-sap becom- ing H. Monotropa and the Indian pipe H. uniflora. Inasmuch as the first post-Linnaean authors gave to Hypopitys exactly the extent which Linnaeus had given to Monotropa, it might be held that the two names are exact synonyms, and that Hypopitys is not available as the name of any genus. On the other hand, the pre-Linnaean history shows that the two names are based on different types, and that if the pine-sap is placed in a different genus from the Indian pipe its name is Hypopitys. Such in effect was the conclusion of the pre-type-system authors Nuttall (25) and Bentham and Hooker (1); such was the conclu- sion of Small (27), who made the combination Hypopithys Hypo- pithys. Repeating binomials being excluded by our rules, we must accept the first specific epithet published after 1753; this yields, as has been shown, the combination Hypopitys Monotropa. We might be glad to reject this combination as a matter of taste; it has gone almost completely unaccepted since its original pub- lication; but the rules are designed to spare us the responsibility for a choice. I follow Kamienski (21) and Domin (11; this work has been a most valuable guide to the history and literature) in recogniz- ing only one species of Hypopitys. There are considerable varia- tions, and Small (28) has recognized five species in North America alone; but these variations seem so inconstant as to make the recognition even of varieties a critical matter. The available material preserved in liquid has included sev- eral roots, stems, and flowers collected by Dr. W. H. Camp, in Oregon, Ohio, and Tennessee, and two shoots with nearly ripe fruit collected by Mr. Willman Spawn in Rock Creek Park, Wash- ington, D. C., in July of 19388. 1'Variations in the spelling will be noted. I have not consulted the pre- Linnaean publications. Linnaeus wrote Hypopithys, Hill and Crantz Hypo- pitys. These are mere variant spellings; but we are forbidden by rule to meddle with them. In using the word as a specific epithet, we must follow Linnaeus; in using it as a generic name, we must follow Hill. 1941] COPELAND: MONOTROPOIDEAE 101 Nothing is here added to knowledge of the gross structure. The shoots originate endogenously in roots, a mass of which con- stitutes the permanent organ of the plant. Domin, after a long discussion of the literature, concludes that the underground struc- tures are not true roots, but an axis not differentiated as root or stem, to be called the Prokaulom (anglicizable as procaulon). I have not examined these structures, but the conclusion is inescap- able, from Kamienski’s (20, 21) description and from what is known of the other monotropoid plants, that Domin is mistaken. As Christoph (3) has shown, the occasional more or less com- plete suppression of some of the characters of roots—the cap and endogenous branching—depends on the presence of mycorrhiza. The same effacement of character appears in the roots of other plants when they are beset with mycorrhiza. The shoot is usually yellowish and more or less pubescent. The upper part of it constitutes a bracted raceme. When it emerges from the ground, the raceme is bent to one side, and the buds or flowers are crowded and more or less pendant; later the axis and pedicels become erect and the flowers or fruits become separated. At its maximum the shoot is rarely twenty-five centi- meters tall. The lateral flowers (pl. 8, fig. 8) stand in the axils of bracts whose margins vary from entire to lacerate. There are no bract- lets; four sepals, a lower pair placed laterally and an upper pair placed dorsally and ventrally with regard to the flower; four separate petals with saccate bases, alternating with the sepals; eight stamens; a nectary with eight horn-like lobes arranged in pairs which clasp the bases of the petalad stamens; a pistil, the ovary four-chambered below, one-chambered above, the stigma obscurely lobed, the lobes opposite the petals. Variations in the proportions of ovary and style, as well as in color, pubescence, and the indentation of the margins of the bracts, have been util- ized for subdivision of the species. Older accounts definitely described the terminal flower as pentamerous. I have not found sucha flower. I believe that in the few shoots preserved in liquid which I have examined the ter- minal flower has been suppressed. The highest flower has been tetramerous, the pedicel embraced by two bracts instead of one (evidently as a result of shortening of the last internode, the one above the insertion of the flower), the lower pair of sepals some- what removed from the flower. Domin cites many authors who have disagreed as to whether the outer envelope of the flower is really a calyx, and its segments sepals: Eichler is the chief authority in the affirmative, Baillon in the negative. He quotes observations of Irmisch (17) and Wydler (30) and gives his own, to the effect (a) that the upper pair of leaves of this envelope are, one or both of them, often suppressed; (b) that the lower pair are often somewhat with- drawn from the base of the flower; (c) that these leaves, espe- 102 MADRONO [Vol.6 cially the lower pair, often have buds in their axils. He con- cludes that they are not true sepals, but elevated bractlets in proc- ess of conversion into sepals. It seems to me unreasonable to recognize a calyx in course of coming into existence in any group as advanced as Ericales: rather, any irregularities are to be inter- preted as matters of degeneration. Since Allotropa is the only positively asepalous genus of the monotropoid group; since only Monotropsis and sometimes Allotropa have definite bractlets; we may interpret the facts assembled by Domin in some such fashion as this: the structures of Hypopitys now under consideration are positively sepals; they are affected by a tendency to degenera- tion; it is possible that a tendency to produce bractlets, almost completely extinct in this genus, retains enough strength to affect the course of the degeneration. Kamienski has accounted in full for the anatomy of the vege- tative parts; I have here only to describe the vascular supply to the flower. One bundle from the circle in the stem turns out- ward. It becomes flattened tangentially and presently splits into three branches, of which the middle one supplies the bract while the two on the sides swing together and unite as a cylinder ascend- ing the pedicel. All this is quite the same as in the genera previ- ously studied. The cylinder of vascular tissue ascending the pedicel becomes compressed in the dorso-ventral plane, so as to approximate a four-sided prism (pl. 9, fig. 10). From each of its faces there departs a sepal trace, leaving a small gap or none. The traces to the lateral sepals depart at a much lower level than those to the dorsal and ventral sepals; this is in harmony with Domin’s observations as to the relative positions of the sepals. The petal bundles emerge as broad bands from the angles of the prism; each promptly forks into three, of which the one in the middle is the smallest and descends under the sac of the petal, while the larger lateral ones ascend past the sac. The petalad stamen bundles are not fused with the petal bundles, but are closely associated with them; each originates as a pair of bundles at the sides of a petal bundle, the pair drawing together and unit- ing above the petal bundle. The sepalad stamen bundles emerge at the edges of the siphonostele as it breaks up. The carpel dor- sals, well developed in the ovary walls and style, are only with difficulty traced to their origin; the feebly developed provascular strands that lead into them seem to originate typically as paired strands beside, above, and resembling the petalad stamen bundles. Foster (15) has recently quoted Gregoire to the effect that floral EXPLANATION OF THE Ficures. Puate 8. Piate 8. PTEeROSPORA ANDROMEDEA. Fic. 1. Model of the vascular system in the receptacle X 50. Ca, sepal bundles; Co, petal bundles; St, stamen bundles; Cl, carpel laterals; Pl, placental bundles. Fics. 2, 3. Juvenile stamens, x 10. Fic. 4. Longitudinal section of juvenile stamen, x50. Fic. 5. Cross section of juvenile stamen at the plane marked zw in fig. 4, X50. Fic. 6. Area marked x in fig. 5, X400. Fic. 7. Cross section of dehisced stamen, X50. Hypoprirys MONOTROPA. Fic. 8. Old flower in which the fruit is nearly ripe, x 5. 1941] COPELAND: MONOTROPOIDEAE 103 P{ Piate 8. PTEROSPORA ANDROMEDEA; Hyporirys MONOTROPA 104 MADRONO [Vol. 6 leaves are distinguished from vegetative leaves by acropetal devel- opment of the vascular supply. The carpel dorsals of Hypopitys (and likewise, as will be seen, of Monotropa and Hemitomes) con- stitute an exception to this principle. The breaking up of the siphonostele finally yields one bundle to each lateral placenta, and — two each to the dorsal and ventral placentae. This seems to be an outcome of the bilateral character of the whole vascular sys- tem. It is as though the ventral bundles of adjacent carpels were fused at the sides of the flower, but not at the front and back. It is regretted that no anatomical study has been made of young stamens. The anther opens by two vertical slits at the outer, presumably proximal, end; these slits meet above, separat- ing a small outer valve from a large inner one; and soon the valves liberate the pollen by swinging widely apart. Young stages and old ones (pl. 9, figs. 11, 12) respectively agree exactly with corresponding stages of Pityopus as illustrated by Eastwood (14). It is particularly regretted that the position of the pollen sacs was not ascertained; though it may be presumed that there are four in each anther, lying horizontally, each slit of the anther crossing the ends of two of them. The pollen grains are two- grooved. The fruit is a capsule, its inner surface covered by a single layer of elongate cells not distinguished by staining reactions. The development of the seed has been described in detail by Koch (22). I have seen only one stage, conforming well to Koch’s account, and so beautifully clear that I could not refrain from drawing it (pl. 9, fig. 9). MonoTROPA UNIFLORA L. Of the name of this genus and species enough has been said above. The genus is apparently monotypic; no variation even of varietal rank is recognized as occurring within the United States. Monotropa coccinea Zuccarini, of Mexico and Central America, and M. australis Andres, of Colombia, were treated as varieties by Domin. The available material preserved in liquid has included six collections, as follows: (1) Three shoots purchased some years ago from the New York Biological Supply Company, as a museum specimen, without collection data. (2) Three shoots presented by the Oregon Biological Supply Company; collected by R. E. Griffin, Bullrun, Oregon, July 11, 1989. The fluid preserves the white color of the plants, and makes specimens excellent for mu- seum use rather than for sectioning. (3) A number of shoots presented by the Herbarium of the University of California, with- out collection data. (4) One shoot, with roots, furnished by Mr. Willman Spawn; collected in Rock Creek Park, Washington, D. C., July, 19388. (5) One shoot, with roots, furnished by Dr. P. W. Zimmerman; collected in the arboretum of the Boyce Thompson Institute for Plant Research, New York, summer, 1939. 1941] COPELAND: MONOTROPOIDEAE 105 Placentals Ge WN Pirate 9. Hyporrrys Monotropa. Fic. 9. Longitudinal section of nearly ripe seed X 400. The endosperm, at a certain early stage, is four-celled; the Roman numerals I-IV indicate the derivatives of these cells. Fic. 10. Model of the vascular system in the receptacle X 12.5: Ca, sepal bundles; Cod, petal dorsals; Col, petal laterals; St, stamen bundles; Cd, carpel dorsals. Fes. 11, 12. Dehisced anthers X10. Monorropa unirtora. Fic. 13. Flower X 2.5. Fic. 14. Flower with perianth removed <5. Fic. 15. Pistil x5. Fic. 16. Stigma X5. Figs. 17-20. Anthers X 10. 106 MADRONO [ Vol. 6 (6) Roots, shoots, flowers and fruits furnished by Dr. W. H. Camp; collected in central New York State at various times. As is well known, the erect shoots are in life white, like paraf- fin (the austral races mentioned above are distinguished by red color) ; dried, or placed in most preservatives, they turn black. The stem is clad with spiral scales; study of a single example showed the spiral to be orthodox. The solitary flower (pl. 9, fig. 13) is terminal on the recurved summit of the stem. Domin has quoted many conflicting authorities as to whether or not sepals are present. I find that a varying number of scales may project past the base of the flower. Of these, sometimes none and some- times one is inserted so immediately below the petals that it can be regarded as a sepal. It is not particularly different from the leaves, and I have not found more than one. The five separate petals overlap at the margins so as to form a campanulate corolla about fifteen millimeters long. Each petal is saccate at the base and truncate at the apex, sometimes with an apiculation. There are ten stamens, manifestly in two whorls, the lower opposite the petals. The densely pubescent filaments are curved inward and bear the subglobular anthers pressed against the lower side of the stigma (pl. 9, fig. 14). The ovary is belted at the base by a ten-lobed nectary; the lobes are cylindrical; they are obscurely but perceptibly paired, clasping the bases of the petalad stamens. The ovary is ovoid, marked by five deep longitudinal grooves op- posite the petals, that is, in the median planes of the carpels; and by five shallow grooves between them where the carpels meet (pl. 9, fig. 15). A shallow circular depression at the summit of the ovary is filled by the base of the short, stout, obconical style. The stigma is five-lobed, the lobes opposite the petals and sur- rounding an unusually large crater-like depression which leads into the style passage. The depression is lined by five masses of tissue which stand above the grooves, not the ridges, in the style passage. The surfaces of the five masses of tissue are more or less wrinkled, and the grooves in the style passage are obscurely continued upward upon them to some distance (pl. 9, fig. 16). The stem resembles in its anatomy that of Pterospora rather than those of Monotropsis and Hypopitys. It shows in cross sec- tion a ring of separate strands of xylem and phloem (in most specimens seriously shattered by shrinkage, apparently during fixation, a difficulty commonly encountered also in Sarcodes and Pterospora). Around the cylinder of bundles there is a continu- ous sheath of pericyclic cells with thinly lignified walls, being imperfectly developed fibers. At the summit of the stem the sheath disappears. Exactly ten bundles enter the receptacle (pl. 10, fig. 25). Each of the five lying in the planes of the petals breaks up, typically, into five. Of these, the middle one is the petal bundle; it forks further into three, a small petal dorsal which follows the contour of the sac, and two larger petal laterals which ascend past it. The behavior 1941] COPELAND: MONOTROPOIDEAE 107 ( St “\\ A pr Pa Pirate 10. Mownorropa uniIFtoraA. Fic. 21. Longitudinal section of anther X25. Fic. 22. Cross section of anther at the plane indicated by w in fig. 21 X25, Fic. 23. Area marked X in fig. 22 x 400. Fic. 24. Pollen grain Xx. 900. Fic. 25. Model of half of the vascular system in the receptacle X 12.5: Cod, petal dorsals; Col, petal laterals; St, stamen bundles; Cd, carpel dorsals; Pl, placental bundles. Hemiromes concrstum. Fic. 26. Cross section of primary flower with three secondary flowers in the axils of the sepals x 10. 108 | MADRONO [Vol. 6 of these petal bundles is quite the same as in Monotropsis, Hypo- pitys, and Pityopus. The two bundles adjacent to the petal bundle unite above it to form a petalad stamen bundle, and the two marginal branches unite to form a carpel dorsal which is traced with difficulty here at its origin. As to the five bundles which enter the receptacle between the planes of the petals, each of these forks into three: the middle one supplies a stamen of the upper whorl; the lateral ones enter the placentae, being ventral bundles of adjacent carpels. The figure shows minor deviations from the assumed typical structure as just described: one of the petal lateral bundles does not originate from a proper petal bundle; one of the placental bundles is suppressed. The carpel dorsals enter the style and ascend it for some distance, but they are there poorly developed. If the anther (pl. 9, figs. 17-20; pl. 10, figs. 21-23) is correctly understood, the outer end is proximal, the inner distal, the upper side dorsal and the lower ventral. At anthesis it includes a single chamber. Apparently (these points were regrettably not estab- lished) there are in the juvenile anther four horizontal pollen sacs, of which the ventral pair are much the larger. Dehiscence is through two curving slits at the proximal end; in dried material, and doubtless in life, these slits flare open as gaping pores. Along the margins of the slits the anther walls, elsewhere of one layer of collapsed cells, are of two layers, the walls of the inner layer bearing reticulate lignified thickenings as in a normal endo- thecium (pl. 10, fig. 23). Jepson states that the slits eventually meet, so as to convert the anther wall into two valves. This is not true in such fruiting material as I have seen; on the contrary, the pores retain their individuality until the plant dies and decays. The pollen grains (pl. 10, fig. 24) are three-grooved. The tube nucleus stains poorly and becomes distorted; the generative nucleus remains globular and deeply staining, and is surrounded by a clear space (not shown in the figure), the generative cell. The inner surface of the ovary wall is covered by a single layer of somewhat elongate cells not distinguished by staining reactions. Of the stages in the development of seed, we know only the structure of the mature ovule. It was described and figured long since by Campbell (2) quite as in Hypopitys.? 2 Since the above was written Dr. Zimmerman has had the kindness to send a beautiful collection of fruiting material made in the Arboretum of the Boyce Thompson Institute in late summer of 1940. The seeds are much as in Allo- tropa; they are elongate, having a tail at each end; the: embryo, borne on a suspensor, is usually of two cells; the endosperm has a haustorium at each end. It is now known that haustoria are produced on the endosperm in Sarcodes, Allotropa, Monotropa, and probably (to reinterpret a former observation) in Monotropsis; and that they are not produced in Hypopitys and Pleuricospora. I suppose that the absence of haustoria in these two genera is a result of par- allel evolution and that the classification given at the end of this paper, in which Hypopitys falls near Monotropa and Monotropsis and far from Pleuriscospora, may stand. 1941] COPELAND: MONOTROPOIDEAE 109 HEMITOMES CONGESTUM Gray Under the name of Newberrya, I have given a partial descrip- tion of this rather uncommon plant of the Pacific Coast of North America, and have quoted from Jepson; but Jepson’s description and mine require extension and amendment. The original generic name Hemitomes was rejected by Torrey (29) as inappropriate; and the rejection was maintained by Small (28) on account of the priority of Hemitomus L’Her. Torrey’s ob- jection is of no standing in modern nomenclature; and Small’s is disposed of by the rule (26) ““When the difference between two generic names lies in the termination, these names must be re- garded as distinct, even though differing by one letter only.” Small recognized five species. Jepson has reduced three of these, whose type localities are in California, to synonymy with the type species. With this I fully agree, and I add the one which Jepson omitted as outside his area. JI am glad to remark that Professor John Davidson has tended to support this action, in a paper read at Seattle in June, 1986; and that Dr. W. H. Camp has done so in private correspondence. The synonymy, then, is as follows: Hemitomes congestum A. Gray, Pacif. Rail. Rep. 6: 80. 1857. Newberrya congesta Torr. in Gray, Bot. Calif. 1: 464. 1876. WN. spicata A. Gray, Proc. Am. Acad. 15: 44. 1879. Hemitomes pumilum Greene, Erythea 2: 121. 1894. Newberrya subterranea Eastw., Proc. Calif. Acad. Sci., ser. 3, 1: 80. 1897. Hemitomes spicatum Heller, Cat. No. Amer. Pl. 5. 1898. H. subterraneum Heller, op. cit. Newberrya longiloba Small, No. Amer. Fl. 29: 18. 1914. N. pumila Small, op. cit. The Herbarium of the University of California has a photo- graph of the type of Newberrya spicata, and a photograph and a duplicate of Suksdorf 2168, the type of N. longiloba. I have been particularly helped by Dr. Jepson, who loaned a specimen col- lected by W. G. Wright, the type of H. pumilum. Two collections preserved in liquid have been available: (1) One shoot without roots, furnished by Dr. L. R. Abrams, who collected it in the California State Redwood Park, Santa Cruz County, June 14, 1934. (2) Several shoots and roots collected by Dr. W. H. Camp at Sol Due Hot Springs, in the Olympic Peninsula, Washington, August 5, 1982. Shoots, arising endogenously from roots, vary in height, the flowers being borne approximately at ground level; they expand in ascending, when well developed exceeding two centimeters in diameter at the base of the inflorescence. The inflorescence is essentially a bracted spike, often so brief and compact as to be accounted a head; in Abrams’ specimen the phyllotaxy, both of the leaves below the inflorescence and of the flowers, is orthodox, the apparent divergence being 3/8 or 5/18. Depauperate shoots may bear a single flower; on the other hand, vigorous shoots may bear axillary branches with one or more flowers, and may develop 110 MADRONO [Vol.6 secondary flowers in the axils of the sepals of the primary flowers. This behavior was noted on dissection of the type of H. pumilum; of Abrams’ specimen; and of a specimen by C. A. Reed in the Herbarium of the California Academy of Sciences. Jepson’s words, “Inflorescence . . . composed of short 2- to 5-flowered spikelets” imply that it is normal; but I find it only in a minority of the specimens; I do not find it in Camp’s preserved material. The flowers (pl. 11, fig. 27) are practically always tetramerous; the four sepals are oriented as in Hypopitys and Pleuricospora, dor- sally, ventrally, and laterally, the dorsal and ventral sepals being not infrequently suppressed. The sympetalous corolla is vari- able in length, from less than one centimeter to nearly two centi- meters long. The four lobes alternate with the sepals, and are separate to a depth of more or less than one-third the total length of the corolla. The dried corolla is very fragile, and it is hard to be certain as to how deep the sinuses extend. Newberrya longi- loba was distinguished by particularly deep sinuses, but I do not find these in our duplicate of the type. Indentation of the corolla lobes is variable; the retuse apex mentioned in my former account was merely the character of an individual. Stamens are normally eight, anthers dehiscing by two lengthwise slits on the outside; lobes of the nectary moderately prominent, evenly spaced; car- pels eight, alternating with the perianth parts and stamens, so that the parietal placentae, and the lobes of the stigma, standing above them, are opposite the stamens and perianth parts. Secondary flowers are usually in the axils of the lateral sepals, but may appear also in the axils of dorsal and ventral sepals (pl. 10, fig. 26). In the cluster figured, there is a second pair of scales, like sepals, above the lateral secondary flowers, as if the scales subtending these were bractlets: Hemitomes would have afforded Domin even better evidence than did Hypopitys, that the outer floral envelope is no true calyx. But what is the scale in whose axil is the dorsal secondary flower? And what is the scale on the ventral side of the primary flower, if not a sepal? I re- gard the scales subtending the lateral secondary flowers as sepals; and the scales above these flowers as a secondary pair of sepals, anomalously developed in connection with the anomaly of buds in the axils of the primary pair. The secondary flowers ordinarily have parts in smaller numbers than the primary ones, as two sepals, seven or six stamens, seven, six, or five carpels. They are later in development than the primary flowers. The vascular system in the stem, as reconstructed from sec- tions of Abrams’ specimen (pl. 11, fig. 80), is a cylinder inter- rupted by very large leaf gaps; as there are no other gaps than these, and as the vascular tissue runs in broad oblique bands rather than in slender vertical bundles, it may be regarded as a siphonostele. The leaf traces, emerging at the bases of the gaps, fork into three as they enter the scales. Flower traces are not at all united with the traces to the subtending scales; the supply to 1941] COPELAND: MONOTROPOIDEAE 111 Puate 11. Hemiromes concestum. Fic. 27. Flower x5. Fic. 28. Cross section of juvenile stamen X50. Fic. 29. Pollen grain x 900. Fic. 30. Model of the vascular system in the stem x5. Fic. 31. Model of the vascular system of the cluster of flowers shown in fig. 26 X 15: Br, bundles to the bract; Ca, sepal bundles; Fl:, supplies to secondary flowers; Caz, secondary sepal bundles; Co, petal bundles; St, stamen bundles; Cd, carpel dorsals; Pl, placental bundles. 112 MADRONO [Vol. 6 each flower consists of two bundles springing from the sides of the gap above one of the upper leaves. The figure of the vascular system of a primary flower with its attached secondary flowers (pl. 11, fig. 81) was constructed from a series of sections of Abrams’ specimen, one of which yielded the diagram (pl. 10, fig. 26). The two bundles of the flower trace approach one another, become flattened, and unite as a siphono- stele. Before this union is complete, four sepal traces emerge as single bundles above which there are large gaps. It seems that a secondary flower is typically supplied, like a primary flower, by two bundles springing from the sides of the gap left by the supply to the subtending leaf, which is in this case a sepal. Actually, we find that the secondary flower placed on the left in the figures is supplied as just described; the one on the right is partially so supplied, and partially from the sepal trace; the third, the dorsal secondary flower, is supplied entirely from the trace to the sub- tending sepal. The single bundles to the secondary sepals of the primary flower arise next to two of the secondary flower bundles. Each petal is supplied by two bundles which leave a single gap in the stele. Beyond these the stele emits three whorls each of eight bundles: stamen bundles, one opposite each perianth seg- ment, all apparently in one whorl, not at all. associated with the petal bundles; carpel dorsals, alternating with the stamen bundles, very poorly developed at the base, so that formerly I failed to discover their origin; and placental bundles, opposite the stamen bundles. Juvenile stamens were found in the secondary flowers. The anther is borne vertically on the summit of the filament and in- cludes four parallel vertical pollen sacs, of which the two on the dorsal side are much the larger. (I was formerly mistaken in making the ventral sacs the larger.) Dehiscence is through ver- tical slits on the dorsal sides of the dorsal sacs; the wall between the two pollen sacs of each lobe breaks down, and the whole outer wall swings around to the ventral side of the anther as a broad valve. The cells of the inner surface of the ovary wall are not dif- ferentiated except perhaps by smaller size. No new observations have been made upon the development of seeds; all that is known is that the ovules are essentially like those of Hypopitys and other members of the group. Discussion In concluding the previous paper of this series, I put forward a tentative scheme of classification of the plants construed as con- stituting the subfamily Monotropoideae of family Ericaceae: Tribe Pterosporeae: Pterospora, Sarcodes, Allotropa. Tribe Monotropeae: Monotropsis, Hypopitys, Pityopus, Mono- tropa, Monotropastrum, Wirtgenia? Tribe Pleuricosporeae: Pleuricospora, Newberrya, Cheilotheca? 1941] COPELAND: MONOTROPOIDEAE 113 Observations have now been extended to all of the above- named genera which are native in North America. Not a few of my previous statements have been found to require correction. A summary of the observations constitutes a formidably exten- sive table (Table 1). The data there assembled enable one to construct a tentative phylogenetic tree (text fig. 1) and to recon- sider the proposed scheme of classification. Hemitomes ~ * we Pityopus SS Monotropa Pleuricospora P / Monotropsis | / Allotropa Sarcodes 7 Peerospora rs 7 a ~—S: ue / — Bp eS ie ! Arbutoideae Fic. 1. Apparent phylogeny of the Monotropoideae. Pterospora, Sarcodes, and Allotropa continue to appear more primitive than the other genera. The primitive features are tall and erect habit; bright coloration; axile placentation in the lower part of the ovary; evenly spaced lobes of the nectary; corolla sympetalous, urceolate (in Pterospora and Sarcodes) ; bractlets present (in some specimens of Allotropa only); anthers horned (in Pterospora only) ; anthers not permanently erect. The only single characters which separate these genera as a group from the others are the habit and the coloration; but the com- bination of characters, lobes of the nectary not paired and anthers not erect, is also distinctive. In all this, there is nothing to raise doubt as to the status of the tribe Pterosporeae as a natural group; often a primitive group can be distinguished from several derived groups taken together only by a combination of negative charac- ters. Doubt appears when, on the other hand, we compare these genera with the supposed ancestral group Arbutoideae. The primitive characters of the Pterosporeae are in large part merely the characters of Arbutoideae; only saprophytism and the habit ~ MADRONO 114 SUITES PUB JPA2T PuNnois yyeoys yno ‘sjeqod ‘sjedas snoje}odurAs quid ye ZSullamoy -YYIM Jopul[éAd | pozemusloyipun éA110q a}ytsoddo [eqorsed Ssno1oUl-P -MOT[PA afed | ‘yoo190 ‘[[eurs sauojiwa Fy [P42] punois yyeoys yno syedas sno[e}edt10yo ye SULIOMOY -Y}YIM JopurpAd | poyerusleyipun AL19q opytsoddo [eyorsed Sno.JoUl-p PIM ‘yooi19 ‘Tews | v1odsoai4una]d yyeoys ST[90 07e3u0TO sunod uoyM poylusty yy poylusi[-uou o[nsdvo sjeyod dA0qe [eyoI1ed | snoyezedt.10yo poulpop ‘10ys so[punq jo sult jo 1aAv] GUO [e@PlT[No0] o}1soddo ‘MOTOq JIIXe Sno.19UI-G OPIUM Ioyyel S1opus[s}|'* ndowjouopy yyeoys no sno[e}edr10yo pode ‘.10Ys -Y}IM Jopulpéo | poyerpuotayipun é AL19q dIndsqo [eqorszed SnNO1OUI-f OPIUM Joyyel Stapusys|* sndohzid ST[90 93e.3uU0 TO sunod usyM yyeoys ou poylust[-uou ainsdeo syTeyod dA0Q” [eJoI1ed | snoTeyodi40yo MOTO pourpep ‘104s ‘sgjpunq jo sult) jo 1aAe[ ou0 auouiere| a}1soddo ‘MO]Oq I[IXe Sno1oUl-f 0} OPIUM Loyyed Slapusy[s shjdodiA S]T[90 97e8uU0Ta sunod uoyM Yyyeoys ou poylust[-uou sno[eyoduds poulpop ‘y210ys ‘satpunq jo dura; jo 1aAe[ ou0 A113q d1ndsqo [eqorred Ssno.ouwl-G é oyyM Jaye ‘tapus[s|* sisdo1zouo sno[edase yyeoys S][99 OAT]-19qy glnsdeo syTeyod dA0qge [eyorted | snoreyedts0yod | sodi4js pot YY Jepurpséo jo tov] VUO avoumere| d}1soddo ‘MOTO I[IXe SNOAIUI-G YM oy pods ‘Te 9401401] P doUsoSIyop jo oul, 1eou a[Ays Jo aseq yyeoys Jno UOI}eIPUIAIIp |ynoqe surusdo steyod sA0qe [eyorsed | snoreyoduiAs [[2} A[oze19pow -YJIM Jopurpso yuRos a[usdeo ayisoddo ‘MO[Aq J[IXB SNOJ9UI-G pot *yOa.19 *yn04s SAPOILDY yyeoys syeyod POYluUsly] YIM S][99 BYI[-1Iqy. o[usdeo oyisoddo = jaaoqve [ejor11ed | snoreyodudAs pou sa[punq jo sul. JO S1dAVT OM} [BplorpnooT, ‘dAndsqo ‘MOTOq JIXe SNO19UI-G ystdand aed pata Tle} pLodso1ajd ALBAO JO BULSI}S W19}8 UI [91S aoe fins 1guUy | yA | jo Surqor] UOI}e}UVNe[ IIMO[ | B Ke) (0) | yqeH | snuer) AVAGIOdOULONO|, NVOIWAWY HLUON dO SUMLOVUVHD ‘[ ATAVY, 115 MONOTROPOIDEAE COPELAND 1941] v éSIQOT Jo spuo [eurxoid SaqoT Jo spuo [eurxoid Saqol jo spuo [eurxoid Saqo] Jo spuo [eurxoid Soqol JO Sopis Jo as{MyysueT (Z) ‘SIqO] JO spua [eurxoid (T) é LIS UOT SoeS STVIJUIA ‘[eyUoZzI.LoYy [enboqns sovs [eizUuIA pue [esiop * po}19AUl oinjeu ‘sunoA usy MA [B}UOZILOY jenboqns Soes [B1JUIA puv [esiop ‘91N}CUL Udt{ pe}oaut ‘Sunod Udy M JooI19 IISUO] I1IV sovs [es..0op ‘ pd} AIAUI IISUO| 918 soes [es1op : [eJUOZL.LOY shijzdodhk sisdo1qouo pdo1407)F sapoowwy DLOdsoLayd oo sulei3 usjod UO S9AOOIS jo toquinN S1oyzue jo soUddSTIYIq soevs ualjod pure sisqquy sjedaeo qyusoelpe JO STe1qUuoA pesnjun 10 sa[punq 10430 s[es.op posnj ¢ spedas sTesiop jo 9o1J JaputpAo Jedaeo jo souvyd utr [edaeo ‘s[LOyM OMY poue}}ey sjedaivo yusoe pe Jo sTe1jUsA posn} so[punq 10y}0 | Sajpunq posnj s[es.op SI yore :stedas s[es.iop JO oJ Ayjetqazed pedaeo jo soue|d ur Jedaeo ‘s[LOUM OM} p ynoqe sjedieo yusoe (pe jo do1J s[ei}ueA pasn) IJUUT ‘soypuNnq s[es.op SI Yove S$ syedas s[esiop [ejod 0} posny Jedaeo jo soue|d ut Jedaeo [4oyA 19}n0 JopurpAéo sjedieo sor_punq yusoelpe Jo s[es1op [Tem ALVAO OF sTeijusA posnj Jedaeo I9UUT ‘saTpuNng ST yore ‘stedas d}yUyep ou [eyed 03 posnj so[punq g sTeyusoe|d jo soured ur ‘sno.oumMu [4oyM Ia}no | ynoQge jo suls S[V19}2[ Jedaieo sures Jo Jedaed 0} STe1jUIA posnj Iauul ‘soTpunqg SI yore ‘steyod S[e194e| [eyed 0} pasng ouou jo soueyd utr Jedaeo [LOW 193n0 Japutpso dpoeydooo1 so[pun so[pun so[pun Sse SANS (a wea ciane ade cae eae GHANILNOD ‘[ FIAVY, snusxy [Vol. 6 MADRONO Igpeoiq SUITIL}S YIM | So_puNq 19430 soes [BS1op SUI}eULI][e JO vo1J Soes [eS.1op Jo ‘[edT}AIA s[es.op SUITILYS JO syTes.op ‘ [LOYM 9UO JapurpAo 6 SopIs [esiop ATvUo1edde [edieo souv|d ut ‘g spedivo g Ajyuoredde UadxO1g sauLojiua FY sjedievo qusoe (pe jo do1J Saqol [enboqns STe1jU9A posny IOUUT ‘soTpuNnqg JO SapIs Jo soes [v1]UIA pue s[Tes.op SI ore ‘stedas syTes.op [eyod 03 posnj v ISIMY} SUI [esiop ‘[eoIz19A Jedavo jo sour|d ul Jedaveo [404M 19ynNo JapurtpAo p1odsorrina) d sjedaeo yusoe (pe Bm ero (0) | JO ST@1JUDA solpunq 19430 éSIQOT JO SoVs [V1JUDA s[Tes.op posnjun ‘stedas s[es.op JO 901} so[pung g spuo [euixoid ‘[e}UOZI.LOY Jedavo jo soueyd ul [edaeo ‘S[LOUM OMY ud} Jo dura | ° vdowouopy suoule}s YIM | sotpunq [Iystd é19du0| SUI}PeUII}e 0} popou éS9qQoT Jo soes [P1qUdA s[Tes.iop suowey}s JO s[Tes.op -u0d IOUUI be spuo [eurxoid ‘[eqUOZI.10Y Jedavo sour|d ut ‘g jedivo g ‘s[1oyM OMY JapurpAéo sndofijid suleis usljod gporydoavo1 uo Bie Reeuyae Jo Semone LS ae sTeyusoR[ g Sib as tBont as ee snuexy yo raquinn doUsDSTYIq pue sisyqquy s[A1S | [Tes A1eAQ UIdUI2}S ur 312 S 116 GdIGNTIONOD ‘T ATAV, 1941] COPELAND: MONOTROPOIDEAE 117 which is presumably a result of it are distinctive. Furthermore, the Pterosporeae are notably heterogeneous in anatomy of stem and receptacle, and in corolla, anther, fruit, and seed. I would allow the tribe to stand as a tentative group, and would expect studies of the Arbutoideae to afford evidence as to whether or not it is a natural group, having one origin among Arbutoideae. The Monotropeae, though differing among themselves in vari- ous features, yet exhibit marks of unity; the saccate bases of the petals and the paired lobes of the nectary are associated with a peculiar and definite structure of the vascular supply to the petal. This is safely a natural group. In the corolla, Monotropsis is the most primitive of the genera; but in the ovary, Hypopitys and Monotropa are more primitive, and the feebly developed sheath of fibers is a primitive character in the stem of Monotropa. We may take it that in most respects Monotropsis represents a direct ances- tor of Hypopitys, and Pityopus an only slightly modified descen- dant; Monotropa stands apart from all three, differing not only in the well known solitary flower and in the anatomy of the stem, but also in the number of grooves on the pollen grain. We may safely maintain Hypopitys and Monotropa as separate genera. The tribe Pleuricosporeae has ceased to be tenable as consti- tuted. The common characters of Pleuricospora and Hemitomes, in habit, floral diagram, placentation, and ovules, are features all of which are apparently readily reached by parallel change and some of which are widely distributed. They are outweighed by the differences in pubescence, vascular anatomy of the receptacle, and structure of the anther. Pleuricospora is strictly glabrous. The petal bundles, not forking into two large bundles and one small one, are those of Pterosporeae rather than of Monotropeae. The anther is unique; the only thing elsewhere in the group that suggests it is that of Pterospora. The four-grooved pollen grains again suggest Ptero- sporeae rather than Monotropeae. But we cannot connect this genus with any one genus of Pterosporeae. Hemitomes has the floral diagram of Hypopitys, and, in more extreme form, the irregularity of inflorescence observed in that genus. We may account for the unique petal trace of two bundles by supposing the small middle bundle of the petal traces of the Monotropeae to have been suppressed, this suppression being associated with loss of saccate bases to the petals and pairing of the lobes of the nectary. The slits along the dorsal sides of the dorsal sacs of the anthers seem to amount to a modification of the pores of normal Ericales, which are still present in Monotropsis and Monotropa. The two-grooved pollen grains are those of most Monotropeae. It would be possible to dispose of these genera by assigning one to Pterosporeae, the other to Monotropeae, distinguishing these tribes by glabrous stamens and pistils in the former, pubes- cent stamens and pistils in most of the latter. Such a scheme, 118 MADRONO [Vol. 6 however, would increase the heterogeneity of the Pterosporeae and break the unity of the Monotropeae. I think it best to place each of these genera in a separate tribe, making altogether four tribes to be distinguished as follows (the oriental genus Cheilo- theca remains unplaced; it may perhaps constitute a fifth tribe) : aXe ee a 10. Stamens and pistils glabrous; lobes of the nectary not paired; no gaps above the petal bundles, to which the outer whorl of stamen bundles are more or less fused; grooves on the pollen grain 4 (3 in Allotropa) ; young stems not nodding. 1. Stems erect above ground; red pigment present; anther bent inward; placentation axile below: Tribe Prrro- SPOREAE: Pterospora, Sarcodes, Allotropa. 2. Inflorescence at ground level; red pigment absent; anthers erect; placentation parietal: Tribe Pxirurico- SPOREAE: Pleuricospora. . Stamens and pistil often pubescent; red pigment usually absent (present in varieties of Monotropa, a trace of it in Hemitomes) ; gaps present above the petal bundles; grooves on the pollen grain 2 (3 in Monotropa). 1. Bases of petals saccate; lobes of the nectary in pairs; petals with small dorsal bundles and large lateral bundles; young stems nodding (unknown in Pityopus) : Tribe MonotropeaE: Monotropsis, Hypopitys, Pityopus, Monotropa, Monotropastrum, Wirtgenia? 2. Base of petals not saccate; lobes of the nectary evenly distributed; dorsal bundles of the petals suppressed; inflorescence at ground level: Tribe Hemrromear: Hemi- tomes. | Sacramento Junior College, Sacramento, California, August, 1940. LiTerRATuRE CiTED Bentuam, G., and J. D. Hooxer. Genera plantarum. 3 vols. London, 1862-1883. Campsett, D. H. Monotropa uniflora as a subject for demonstrating the embryo sac. Bot. Gaz. 14: 83. 1889. CuristropH, H. Untersuchungen iiber die mykotrophen Verhialtnisse der “Ericales” und die Keimung der Pyrolaceae. Bot. Centralbl. Beih. 38°: 115-157. 1921. Corrtann, H. F. The development of seeds in certain Ericales. Amer. Jour. Bot. 20: 513-517. 1933. . The structure of the flower of Newberrya. Madrofo 2: 137-142. 1934. On the genus Pityopus. Madrofio 3: 154-168. 1935. . The reproductive structures of Pleuricospora. Madrofo 4: 1-16. 1937. The structure of Allotropa. Madrono 4: 137-153. 1938. The structure of Monotropsis and the classification of the Monotropoideae. Madrono 5: 105-119. 1939. Crantz, H. I. N. Institutiones Rei Herbariae.... 2 vols. Vienna. 1766. 1941] BUCHHOLZ: PODOCARPUS GRACILIOR 119 11. Domin, K. Vergleichende Studien iiber den Fichtenspargel. Sitzungsber. kgl. bohm. Gess. Wiss., II KI., I Stiick: 1-111. 1915. 12. Doyen, B. E., and L. M. Goss. Some details of the structure of Sarcodes. Madrono 6: 1-7. 1941. 13. Drupe, O. Pirolaceae. in A. Engler and K. Prantl. Die natiirlichen Pflanzenfamilien 4°: 3-11. 1889. 14. Eastwoop, Auice. Some new species of Californian plants. Bull. Torrey Bot. Club 29: 75-82. 1902. 15. Fostrr, A. S. Problems of structure, growth and evolution in the shoot apex of seed plants. Bot. Rev. 5: 454-470. 1939. 16. Hitt, Joun. The British herbal. ... London. 1756. 17. Irmiscu, T. Einige Bemerkungen iiber die einheimischen Pyrola-Arten. Bot. Zeit. 14: 585-591, 601-602. 1856. 18. Jepson, W. L. Manual of the flowering plants of California. Berkeley. 1925. 19. ———————.. A flora of California, vol. 3, part 1. Berkeley. 1939. 20. Kamiensk1, F. Die Vegetationsorgane der Monotropa hypopitys L. Bot. Zeit. 39: 457-461. 1881. 21. —_—_————_. Les organes végétatifs du Monotopa Hypopitys L. Mém. Soc. Nat. Sci. Nat. Cherbourg 24: 5-40. 1882. 22. Kocu, L. Das Entwicklung des Samens bei Monotropa Hypopitys L. Jahrb. wiss. Bot. 13: 202-252. 1882. 23. Linnaeus, C. Species Plantarum. 1753. 24. Mac Doveat, D. T. Symbiotic saprophytism. Annals Bot. 13: 1-47. 1899. 25. Nurratt, T. The genera of North American plants. ... Philadelphia. 1818. 26. Renpite, A. B., and others, editors. International rules of botanical nomenclature ... dritte Ausgabe. Jena. 1935. 27. Sma, J. K., and A. M. Var. Report of the botanical exploration of south- western Virginia during the season of 1892. Mem. Torrey Bot. Club 4: 93-201. 1893-1894. 28. Smatt, J. K. [Monotropaceae] in North American Flora 29': 11-18. 1914. 29. Torrey, J. On Ammobroma, a new genus of plants. ... Ann. Lyc. Nat. Hist. New York 8: 51-56. 1864. 30. Wypier, H. Kleinere Beitrage zur Kenntnis einheimischer Gewachse. Flora N. R. 18: 613-617. 1860. PODOCARPUS GRACILIOR IN CULTIVATION Joun T. BucHyHouz > The “African fern-pine,” in recent years popular as a decora- tive tub plant and also planted in the open as an ornamental coni- fer, may become a tree of considerable size. Since mature speci- mens growing in California have produced pollen cones during the past winter, it is now possible to identify the species as Podocarpus gracilior Pilger. The plant has been regarded as a conifer of South African origin and has usually passed under the nursery trade name of “Podocarpus elongata,” which is unquestionably an error. The latter is the legitimate name of a plant of South Africa where there are two narrow leaved species that have been confused and have at one time or another passed under this botanical name. The California exotic has narrow leaves that are somewhat similar to those of Podocarpus elongatus L’Herit. (the earliest described 120 MADRONO [Vol. 6 species in this genus), but the pollen cones and seeds differ con- siderably. It resembles P. falcatus (Thunb.) R. Br. (P. elongata Carr.) more closely than P. elongatus L’Herit., but has green in- stead of brown twigs, leaves that become longer and wider, and pollen cones that are more than twice as long; also the tips of individual microsporophylls (connectives) of the pollen cones are more pointed. In both P. gracilior and P. falcatus the pollen cones are axillary, borne singly or in fascicles of two or three. The seeds of Podocarpus gracilior were brought over from East Africa in 1911 by Mrs. Stewart Edward White. Franceschi, of Santa Barbara, germinated some, if not all, of these. The state- ment is usually current that these seeds were introduced from South Africa (an error which has appeared in print), but the material itself does not agree with the descriptions of either of the narrow-leaved South African podocarps. It fits very closely the description of Podocarpus gracilior. This entire question has been clarified recently by a letter received from Colonel Stewart Edward White in which he states that the tree from which these seeds were collected was located in Kenya, British East Africa. This region of Africa is included in the range given for Podocarpus gracilior Pilger. Specimens of this species at the United States National Herbarium were collected in Kenya by Edgar A. Mearns of the Smithsonian African Expedition which was led by Theo- dore Roosevelt in 1909-1910. The male flowering material col- lected in California in January and February, 1941, in three © widely separated localities agrees in all essential details with the Mearns specimens. Podocarpus gracilior is dioecious, as are nearly all species of Podocarpus. Except for the limited number of original seedlings, the trees in California have been propagated from cuttings. The stock tree used for propagation since about 1922 at the Coolidge Rare Plant Garden Nursery in Pasadena is a male tree. During the past winter Mr. J. J. Mulvihill has kindly sent me a number of reproductive twigs. Thousands of plants have been grown as cuttings from this tree over a period of years and furnished to the nursery trade under the names “African Fern-Pine” and ‘Podo- carpus elongata.” They do very well as tub plants and when these long-suffering decoratives become too old they may be trans- planted to parks and gardens. The writer has seen this conifer used as a street tree in Los Angeles and Pasadena. Apparently they do not become reproductive until they are mature specimens of large size. The largest known specimen, about fifty feet high, is growing at Alameda Plaza in Santa Barbara where two other mature specimens may be seen, none of which had been observed in reproductive condition. However, during July, 1941, Mr. Van Rensselaer of Santa Barbara Botanic Garden found one of these (located in the east section of Alameda Plaza) bearing seeds. Many of the seeds were abortive when less than half grown; some 1941] BUCHHOLZ: PODOCARPUS GRACILIOR 121 Fic. 1. Twigs of Podocarpus gracilior Pilger bearing nearly full grown pollen cones, January, 1941. had enlarged to full size, but all of those which were examined by the writer, were without embryos, with the endosperms shriveled in the manner usual for unpollinated parthenocarpic seeds. With this one exception, all reproductive specimens of P. gracilior thus far observed have proved to be male, although pre- sumably the seeds would produce equal numbers of male and fe- male seedlings. The location of many of the original seedlings is not known. Two specimens growing on the estate of Colonel Stewart Edward White near Burlingame have not been observed in reproductive condition at any time. The three large trees at Alameda Plaza are probably seedlings, and likewise any other old specimens growing in Santa Barbara. It is likely that the tree belonging to Mrs. E. N. Hazard, mentioned by Van Rensselaer (Trees of Santa Barbara 1940, p. 84) is also one of the original seedlings. In 1936 the writer found five or six large specimens on the Dwight Murphy estate at Montecito, Santa Barbara County and these may be seedlings. They have not been observed in reproductive condition. The late Miss Kate Sessions of San Diego informed the writer in 1936 that she had obtained three of the original seedlings from Franceschi. The location of two of them was given. One is growing in the gardens of Julius Wangenhein, 148 West Juniper Street, San Diego. Another is the large tree at the Rosecroft Gardens in San Diego. The latter plant was not reproductive in 122 MADRONO [Vol. 6 1936, but was found with pollen cones during the past winter. Mr. A. D. Robertson furnished the writer with male reproductive specimens from this plant in January and February, 1941. A male tree growing in the Botanical Garden of the University of California, Berkeley, was observed to be in reproductive condi- tion in January, 1941, by Mr. Donald G. Nelson of that institution. The origin of this plant is not known to the writer. Aside from the dozen plants enumerated here, there are probably a number. of other specimens in cultivation on private estates that represent original seedlings, which were distributed in the days before it was discovered that these plants are easily propagated. University of Illinois, Urbana, July 21, 1941. THE TAXONOMIC STATUS OF MICROSTERIS GREENE Hersert L. Mason Perhaps no member of the Polemoniaceae has been so greatly misunderstood as the very polymorphic aggregate species, Phlox gracilis (Dougl.) Greene. It has been variously treated as a mem- ber of the following genera: Gilia, Collomia, Phlox, Navarretia, Polemonium, and is the type species of the genus, Muicrosteris Greene; it has been divided and subdivided into species, sub- species, varieties, subvarieties and forms within these genera ac- cording to the particular whim of the author treating it. The plant ranges from the Pacific Coast to the Rocky Mountains and from temperate Alaska south to Mexico, and recurs in the South- ern Hemisphere in Bolivia, Chile, and Argentina. Essentially an early spring annual, it occurs from the coastal bluffs to timberline. The intent of the present paper is to deal only with the generic position of the aggregate species and not to be concerned with the status and disposition of the smaller taxonomic units. Therefore, the entire group of variants will be treated, for the present at least, as one large, polymorphic species. The species was first collected by Douglas on the banks of the Spokane River [Washington] and given the manuscript name, Collomia gracilis; it was first described by Hooker (6) in 1829 under the name Gilia gracilis with Collomia gracilis Douglas cited as a synonym. In 1887 Greene (4) referred the species to the genus Phlox with the statement: “This interesting plant came to the knowledge of botanists some years in advance of Phlox Drum- mondii Hook. and its allies. It was at first a thing of dubious aspect, not at home either in Gilia or Collomia. But since the dis- covery of the Texan group of annual species of Phlox with peculiar habit, it must have been the mere force of custom which has kept men from seeing that it is an absolutely perfect congener of Phlox Drummondi.” In 1891 (7, p. 433) O. Kuntze, recognizing the page priority of Navarretia over Gilia, made a purely nomencla- 1941] MASON: MICROSTERIS 123 torial shift in the combination Navarretia gracilis (Dougl.) Kuntze. In 1898 Greene (5) erected the genus Microsteris recognizing as species seven segregates of Phlox gracilis. In so doing Greene stated: ““At present I am disposed to adopt it as a principle that species with mucilaginous seeds are nowhere, in this family, to be placed as congeneric with such as have seeds devoid of the gum- miferous coating. This implies the removal of my Phlox gracilis from the genus Phlox.” In his description of Microsteris he states “Calyx, corolla, stamens and capsule wholly as in Phlox.” Thus Greene’s Microsteris hangs by the single character “mucilaginous seeds.” Inthe same year O. Kuntze (8, p. 203) referred Collomia gracilis Dougl. to Polemonium by the simple statement: “P. More- nonsis OK (Collomia gracilis Dgl. non Polemonium gracile).”’ His reasons are forever hidden in parenthetical synonymy. Wecan dismiss without further comment the references to Navarretia and Polemonium. The reference by Douglas to Collomia is understand- able. It was based upon superficial resemblance; furthermore, at that time the genus Collomia had not been clearly circumscribed in the light of the family as a whole. Our problem resolves itself into determining whether Phlox gracilis shall be retained in Gilia as interpreted by Hooker, be retained in Phlox as interpreted by Greene in 1887 or be placed in Microsteris following Greene’s later interpretation. Of subsequent authors most have preferred to follow Asa Gray’s adaptation of Hooker’s treatment in a broad concept of the genus Gilia while only a few have used either Phlox or Microsteris when referring to this species. Brand (2) in his monograph of the Polemoniaceae with its highly elaborated sys- tem of “pigeon holes” chose to place Phlox gracilis in the genus Gilia, subgenus Benthamiophila, section Phlogastrum and _ pro- ceeded to divide the species into fourteen entities in various sub- specific categories. With respect to the generic position of the species I quote from Brand, “‘Species sic intermedia inter genera Phlox et Gilia, ut vix discernere possis, cui generi eam attribuas; a Collomia tamen, quacum plurimi autores junxerunt, calyce, ut cl. Greene docuit, valde diversa.” Although he cited Microsteris as a synonym it is clear from the above quotation that he did not re- gard Microsteris as offering any problem. He was concerned with differentiating Gilia from Phlox. Here again we find but a single character utilized to place the species in Gilia, namely, the fact that the seeds develop mucilage when wetted. Other characters which it possesses that align it with Phloz are treated by Brand as exceptions in Giulia. The most recent treatment that bears on this problem is that of Wherry (9) from whom we quote, “Microsteris. A few diminu- tive western annuals constitute this genus, which has been by vari- ous authors referred to Collomia, Gilia and Phlox. It shows little relationship with the first two genera, and in view of the differ- ence in seeds can scarcely be congeneric with the last, although it may well be a derivative.” Wherry, it will be seen, dismisses 124 MADRONO 7 [Vol. 6 Gilia and Collomia from consideration but parries between Phlox and Microsteris. He finally eliminates Phlox on the basis of ““COROLLA-LIMB small; seeds becoming sticky when moistened,” but he does at least suggest the responsibility of Phlox for the offspring. In an effort to validate the genus Microsteris an- other very insignificant character is added to the one previously utilized, namely the small size of the corolla limb. It is perhaps a reasonable mode of escape when a group of plants does not fit comfortably in any of the related genera to erect a genus for it. However, this procedure should not be adopted until all of the evidence is carefully weighed to determine the precise nature of the differences that seem to make it neces- sary. As pointed out above Microsteris was erected by Greene who listed for it a single character difference from Phlow. Wherry’s additional character of a small corolla limb adds scarcely anything of generic significance. The following tabular arrangement presents the facts pertaining to the development or non-development of mucilage or spiracles in the seed coats of most of the more widely accepted genera or Polemoniaceae. Bonplandia: all species develop mucilage. Cantua: a few species develop spiracles, the rest do not. Cobaea: some species produce spiracles, and other species mucilage. Gilia: very diverse, some species produce mucilage, others do not. The section Ipomopsis, recognized as a distinct genus by Wherry, is about equally divided in this respect. Hugelia: some species produce mucilage, others do not. Langloisia: all species produce mucilage. Leptodactylon: in species examined none produce mucilage. Linanthus: most species produce mucilage, some do not. Loeselia: some produce mucilage, others do not. Navarretia: some produce mucilage, others do not. Phlox: as interpreted by Greene and by Wherry, does not pro- duce mucilage, but if Microsteris is included, will be on the same basis as the other large genera. Polemonium: some species produce mucilage, others do not. The remaining few genera are each very small, and I have not as yet investigated them. But from the above data it would ap- pear that the development of mucilage by the seed coat cannot be relied upon as of primary generic significance. All we can say of Phlox is that in the majority of species the seeds are immutable when wetted. This leaves as a character for the segregation of Microsteris only the small corolla limb. The magnitude of dif- ference here, however, is no greater than the variational limits of corolla size in several other genera of Polemoniaceae, such as Col- lomia, Navarretia and Linanthus. This evidence, it seems, is just cause for denying generic status to Microsteris. When we consider the characters that serve to keep the Phlozr gracilis aggregate out of Gilia we turn from the flower and seed to 1941] MASON: MICROSTERIS 125 Puate 12. Comparison oF PHLOx GRACILIS AND PHLox DRUMMONDII VAR. TENUIS. Fic. 1. Phlox gracilis. Fic. 2. Phlox Drummondii var. tenuis. Fic. 3. Phlox gracilis, flower. Fic. 4. Phlox gracilis, opened corolla. Fic. 5. Phlox gracilis, capsule showing disarticulation of valves. Fic. 6. Collomia, capsule showing valves with margins reflexed. Fic. 7. Campanulate type of capsule found in many species of Linanthus and Gilia. 126 MADRONO [Vol. 6 other parts of the plant. Of the authors who have referred the group to Gilia we find some who regard most of the small genera (Linanthus, Hugelia, Gymnosteris, Loeselia, Collomia, Leptodactylon) as belonging to this genus; others who recognize the small genera mentioned above but who have followed precedent in the disposi- tion of Phlox gracilis. When we exclude from Gilia these small genera there still remains a polymorphic but closely related group of species. The leaves of this remaining group are normally alternate (occasionally through shortening of the internodes they may appear subopposite), and frequently pinnately toothed, lobed or dissected; the corolla lobes are normally entire; the stamens are usually, but not always, equally inserted and equal in length; the capsule valves do not disarticulate on dehiscence but remain united at the base, and although the valves may spread cam- panulately or sometimes reflex on the midvein the capsule falls as a whole (pl. 12, fig. 7); the locules of the ovary are usually more than one-seeded, but occasionally are one-seeded; the seeds are usually small and angular. Phlox gracilis does not conform with Gilia as the following summary of its characters demonstrates: the leaves are pre- dominately opposite (pl. 12, fig. 1), at least below, and are always linear, or oblong and entire; the corolla is salverform, the limb rotate, the lobes frequently emarginate (pl. 12, fig. 3); the sta- mens are unequally inserted and unequal in length (pl. 12, fig. 4); the capsule valves are rigid and disarticulate completely on dehiscence; the locules are one-seeded, the seeds large (pl. 12, fig. 5). Greene was quite correct when he said in his diagnosis of Microsteris, ““Calyx, corolla, stamens and capsule wholly as in Phlox.” And of course Wherry accepts for this group a close relationship to Phlox. It seems that the presence of such typical Phlox characters as the rigid, disarticulating capsule valves and the solitary large seeds in the locules, together with several minor characters which are usual in Phlox and occasional or abnormal in Gilia, throw the weight of the argument to Phloz, not to Gilia. Another line of evidence supporting a relationship with Phlox rather than with Gilia is found in cytological studies; the basic chromosome number in Gilia appears to be n=9 while the basic chromosome number in Phloz is n=7. In Phlow gracilis 2n=14, the count being made from root tip cells. However, in a group with such wide climatic tolerance and such great morphological diversity we may anticipate some polyploidy. Botanists familiar with the genus Phlox only in western North America may be pardoned for hesitating to place P. gracilis, a plant so different from P. Douglasii and P. adsurgens, in the same genus. It is, as Greene points out, only when we take into con- sideration the range of variation of the entire genus that we can hope for a true picture of relationship. In this case the Phlox Drummondii complex of Texas offers a key to the relationship. A collection of Phlox Drummondii var. tenuis Gray from Texas 1941] : MASON: MICROSTERIS 127 (Lindheimer 468) is an excellent example of a connecting type between Phlox gracilis and other members of the genus. A com- parison of figures 1 and 2 (pl. 12) will at once show the great similarity in aspect between the two. Figure 1 represents a plant of Phlox gracilis collected at Tuolumne Meadows, Yosemite National Park, California (Mason 4869). It was especially se- lected for this comparison but is representative of a large segment of the “Microsteris” variants. The evidence of a general similar- ity of aspect substantiated by indisputable Phlox characters up- holds Greene’s first opinion of the generic position of this group. The fact that this western group of plants is related to an eastern group by way of a southern bond is not inconsistent with the growing body of information now being accumulated relative to the history of vegetation in the southwest. Among other genera with related species showing a similar distribution pattern are Juglans, Cercis, Forestiera and Frazinus. This group of trees and shrubs are all represented in fossil floras of Middle Tertiary time and today occur in savanna like floras where Phlox gracilis is a common associate. It would seem that these relationships go back at least as far as the Miocene, if not the Oligocene, in the Sierra Madrean flora of Axelrod (1). Perhaps this region has been the center of origin and differentiation of the entire Pole- moniaceae. Certainly not all Phlox species have had their origin in Keewatin Land as postulated by Wherry (10). If this were true it would be reasonable to expect a higher development of the genus in the old world than is now evident, since migration routes through Beringia would have been available. The occurrence of Phlox in this northern region during the Pleistocene, north and west of the Keewatin center of glaciation is attested by fossil fruits reported and figured by Chaney and Mason (3 p. 17, figs. 34,36). These specimens are strikingly similar to P. sibirica L., a species occurring in the Alaska region today, and ranging west- ward into Siberia. Department of Botany, University of California, Berkeley, August 27, 1941. LIrERATURE CITED — AxeEtrop, D. A Miocene flora from the western border of the Mohave Desert. Carneg. Inst. Wash. Publ. 516. 1939. 2. Brann, A. [Polemoniaceae] in Engler, Pflanzenreich 4°°: 1-203. 1907. 3. CuHaney, R. W., and H. L. Mason. A Pleistocene Flora from Fairbanks, Alaska. Amer. Mus. Nov. no. 887: 1-17. 1936. 4. GREENE, E. L. New or noteworthy species. Pittonia 1: 139-143. 1887. 5. ————————.. Some western Polemoniaceae. Pittonia 3: 299-305. 1888. 6. Hooxer, W. T. Bot. Mag. 56: t. 2924. 1829. 7. Kuntze, O. Rev. Gen. PI. 1. 432-434. 1891. 8. ———_—_—_—_—_.. Rev. Gen. Pl. 3”: 202-203. 1898. 9. Wuerry, Epcar T. A provisional key to the Polemoniaceae. Bartonia 20: 14-17. 1940. 10. ————_—__., Geographic relations in the genus Phlox. Op. cit. 20: 11-14. 1940. 128 MADRONO [Vol. 6 THE HOLACANTHOID PLANTS OF NORTH AMERICA Cornetius H. Mutyer One of the striking botanical phenomena of the southwestern United States and adjacent Mexico is the occurrence of several genera of different families with vegetative characters so similar as to make them confusing. Perhaps the most outstanding such instance is that of the usually leafless green spiny shrubs which resemble Holacantha Emoryi A. Gray, a member of the Simarub- aceae. The similar forms are Koeberlinia spinosa Zucc. in the Koeberliniaceae, Canotia holacantha Torr. in the Celastraceae, and Thamnosma montana Torr. & Frem. in the Rutaceae. These four species are characterized by leaves reduced to scales and early caducous, and spinescent stems persistently green to carry on the photosynthetic process. The lack of leaves and similarity of the spines make these plants rather difficult to dis- tinguish without flowers or fruit. Yet, a series of vegetative characters may be recognized by which the species are readily distinguished. There are other spinescent plants which might be confused with the group here treated. The others, however, are charac- terized by shorter and more slender spines and usually more prominent leaves. For instance, Adolphia in the Rhamnaceae and Forsellesia in the Celastraceae contain species highly similar in habit to the Holacantha-like plants, but in these the spines do not exceed an average of 2.5 millimeters in thickness, and the leaves are prominent throughout the greater part of the growing season. Furthermore, Adolphia is characterized by opposite leaves and branches, while those of the plants here treated are all alternate. Two variants of Koeberlinia spinosa have been distinguished under varietal names. Both of these are characterized by very slender spines which might make them difficult to distinguish from Adolphia except for their alternate habit of branching. Leafless individuals of the leguminous shrubs Cercidium and Cassia armata might become confusing when they lack flowers and fruit. The highly artificial group comprised of the Holacantha-like plants is characterized by the alternate-branching, spinescent stems, the spines usually quite coarse, branches green and photo- synthetically functional for several years, leaves much reduced and early caducous. Johnston (Journ. Arn. Arb. 21: 356-368. 1940) has pointed out the significance of this group of unrelated plants of similar habit as indicative of a relationship between the North American deserts and those of South America where the same habit is common. He showed the habit to be much more general in the South American deserts than in North America. The highly endemic character of all our species except Koeber- linia spinosa would indicate that the plants are relicts of a time when the habit was more common in North American deserts. 1941] MULLER: HOLACANTHOID PLANTS 129 Key Basep oN VEGETATIVE CHARACTERS Leaf scars and branching alternate, leaves inconspicuous and early caducous. Plants glabrous, stem either densely glandular or with minute longitudinal lines of white waxy flakes. Stems densely glandular or warty with translucent glands; buds and spine tips tan; branches with a pair of extra- axillary buds at the bases (one on each side).......... Thamnosma Stems not glandular, with minute longitudinal lines of white waxy flakes; buds dark brown or black, spine tips brown; branches with no extra-axillary buds at the bases (though the buds along the length of the branch may descend to aapoint wear the WaASe) fis dof nar eees eo ps Rane i in aes Canotia Plants with young stems pubescent, neither glandular nor with waxy exudations. Younger stems yellow-green, minutely puberulent with short spreading hairs or these reduced to pustules; branches and spines with a pair of buds at the bases, these extra- axillary. (one on each side) ..............0.. 0c eens Koeberlinia Younger stems gray green with densely matted appressed silky hairs; branches and spines with a single axillary bud and no lateral or extra-axillary buds at the bases .. Holacantha Leaf scars and branching opposite or, if alternate, the leaves conspicuous and persistent ... (genera not treated in this article). THAMNosMA MoNTANUM Torr. & Frem. in Frem., Rep. Exped. Rocky Mts. 813. 18465. Central Arizona north to southwestern Utah, southern Ne- vada, and southern California; most common in western Arizona. A shrub usually 3 to 6 or 7 decimeters tall. (fig. 1). CanoTia HOLACANTHA Torr., U. S. Rep. Survey Railroad Miss. Pac. 4: 68. 1856. Southeastern and central Arizona to northwestern Arizona and doubtfully in the Providence Mountains of southeastern Cali- -fornia. A shrub 8 to 5 or even 6 meters tall. (fig. 2.) KoEBERLINIA SPINOSA Zucc., Abh. Akad. Muench. 1: 859. 18382. Southern Arizona and New Mexico, western Texas, Baja Cali- fornia, northern Sonora and Chihuahua, south through Coahuila and Nuevo Leon to Puebla and Oaxaca. A shrub 0.5 to 2 meters or even a small tree to 5 meters tall. (fig. 1.) KoEBERLINIA SPINOSA Var. TENUISPINA Kearney & Peebles, Journ. Wash. Acad. Sci. 29: 486. 1939. Yuma County, Arizona, and Sonora, Mexico. Differs from the species in its elongate slender spines, blue-green color, and usually longer sepals, petals, and filaments. Typical K. spinosa apparently occurs nowhere west of Tucson, Arizona. This variety is not distinguished from the species on the distribution map. A second variety (K. spinosa var. verniflora Bogusch, Torreya 31: 74. 1931) now scarcely seems worthy of distinction. Al- though this form differs strikingly from the species by its slender spines (and generally reduced size of the organs) and its early flowering (in March and April), its differences are not sufficiently profound nor constant to warrant formal recognition. Nor is [ Vol. 6 ~ MADRONO 130 WOANVLNOW VWSONWVHL X VSONIdS VINITHAGIOy @ es ILYVYM3LG VHLNVOOTOL] O \AMOW VHLNVOOTOL) xX VHLNVOOIOH VILONVS @ Ga el 1941] MULLER: HOLACANTHOID PLANTS 131 there any geographical segregation from the species. The mor- phological basis of this form has been observed by the author in various parts of western Texas and adjacent Chihuahua and Coahuila. In a letter to the author under the date October 7, 1940, Bogusch discusses the variety as follows: “ . observa- tions made in the field upon specimens of Koeberlinia spinosa Zuce. convince me that the variety verniflora described by me is prob- ably a normal reaction of the plant to wound stimulus or repre- sents a form of juvenile growth. From material seen both in the type locality of the Rio Grande Valley of Texas and in the region west of Uvalde, I have seen that new growth which follows ex- tensive injury to the plant often results in being more attenu- ated, both in branches and the spines. The peculiarities asso- ciated with the time of flowering—not necessarily a taxonomic character—may be best explained that in the Rio Grande Valley the vegetation in many respects matures earlier and comes into flower sooner than elsewhere in the state.” Horacantua Emoryr A. Gray, Mem. Am. Acad. ser. 2, 5 (PI. Nov. Thurb.): 310. t.8. 1855. Central and southwestern Arizona to southern California. A shrub 1 meter tall to a small tree reaching 3.5 meters (fig. 2). An undescribed species of Holacantha was recently discovered in northen Mexico. Holacantha Stewartii sp. nov. Frutex procumbens vel ad- scendens 1.5-3(6) dm. altus duplo latior; spinae plusminusve appresso- vel patenti-pubescentes glabratae, papillis minutis exceptis; fructus acutus margine ventrali obtuso-costatus. Low shrub, 1.5—3 dm., rarely 6 dm. tall, usually 2—4 times as broad, procumbent or somewhat ascending, soon leafless, coarsely spiny; stems and spines divaricate; spines 2.5-—6(12) cm. long, (1.5)2.5-3 mm. thick, tips subulate, brown, 3-4 mm. long, branches terete, the immature ones becoming sulcate in drying, glabrate or somewhat spreading-pubescent or appressed-seri- ceous, hairs short, bases persisting as minute papillae; buds in- conspicuous in axils of spines and sparsely scattered along their length, surrounded by small tufts of coarse appressed hairs; leaves quickly deciduous, oblong, acute at both ends, sessile, 5—8 mm. long, 2—2.5 mm. broad, red at vernation (as are the young spines for a time), densely white- or rose-hirsute, becoming green and sparsely hirsute ; flowers dioecious; staminate calyx of 6 ovate, acute, pubescent sepals about 1 mm. long; corolla of 6 fleshy, dorsally pubescent, deeply concave petals with narrow, thin mar- gins, 4 mm. long, 1.75 mm. broad (not flattened), stamens about 12, filaments 1.5—-2 mm. long, broadened basally, strongly hirsute, apex subulate, glabrous; pistillate calyx similar to the staminate; pistillate corolla not seen; staminodes similar to the functional filaments; carpels distinct, apically connivent, stigmas sessile, fused; fruit persistent 1-2 years, 6-carpellate (or carpels fewer 132 MADRONO [Vol. 6 by abortion), carpels distinct, divaricate, lenticular-ovate, acute, ventral margins obtusely ridged, superior, 8-9 mm. long, 5-6 mm. broad, 3—4 mm. thick, glabrous, red or green, surface nearly smooth, lacquered. (fig. 2.) Range: Mexico; western Coahuila and northern Zacatecas. Holacantha Stewartiti is named in honor of Mr. Robert M. Stewart of Santa Elena, Coahuila, whose superior hospitality and whose company on several side trips contributed markedly to the pleasure and success of the several weeks’ work in the vicinity. This species is the second described in this rare and striking genus, the first being Holacantha Emoryi which is confined to Ari- zona and California. From that species H. Stewarti differs in its low sprawling habit, the sparse pubescence of its stems (com- pared with the densely short-tomentose stems of H. Emoryz), the persistence of papillae-like hair-bases, and its usually markedly acute and ventrally ridged carpels. The great discrepancy in the ranges of these two endemics further attests their distinctness. Specimens examined. Coanuiua: Sierra de las Cruces, gulch in limestone hills 0.5 mile north of Santa Elena, August 13, 1940, I. M. Johnston & C. H. Muller 215 (United States National Arbo- retum, USNA, type, sheet no. 96733; Arnold Arboretum, AA) ; southeast base of Sierra de las Cruces, 3 miles northeast of San José, September 5, 1940, 7. M. Johnston & C. H. Muller 1003 (AA, USNA); north base of Sierra de las Cruces, at San Rafael, Sep- tember 8, 1940, I. M. Johnston & C. H. Muller 1084 (AA, USNA); northwest base of Sierra de las Cruces, at San Vicente, September 8, 1940, I. M. Johnston & C. H. Muller 1065 (AA, USNA); 8 to 5 miles south of Laguna de Jaco, September 10, 1940, J. M. Johnston & C. H. Muller 1104 (AA, USNA); north end of Bolson de los Lipanes between E] Almagre and Cerros de Leja, September 12, 1940,I.M. Johnston & C. H. Muller 1239 (AA, USNA). Zacatecas: banks of arroyos in foothills, Hacienda de Cedros, 1908, F. E. Lloyd 191 (United States National Herbarium ). With the exception of the type collection and Lloyd’s from Zacatecas all the plants collected or observed grew in deep, heavy silt flats, usually associated with Koeberlinia spinosa Zuce. Al- though the most luxuriant growth and fruiting occurs in rocky arroyo sites, the species is obviously more at home in the former habitat, as is evidenced by its more frequent occurrence there. The plant is often the only one (or one of a few) on otherwise bare silt. Its procumbent habit serves to bind the soil and forms hillocks down the sides of which the branches sprawl. In two of the seven collections studied fasciated stems were noted. These are flattened and falcate with two ranks of simple normal spines issuing from their edges. It is odd that two such cases of identical abnormality were encountered in so rare a plant, about fifty individuals being observed in the course of a wide and painstaking search. Bureau of Plant Industry, Washington, D. C., February, 1941. 1941] PECK: NEW COMBINATIONS 133 COMBINATIONS PROPOSED IN “THE HIGHER PLANTS : OF OREGON” Morton EK. Peck There has been some question as to the validity of certain new combinations proposed in the ‘“Manual of the Higher Plants of Oregon,’ recently published by the writer. The citations re- quired if one is to adhere strictly to the International Rules (Ch. 3, Sec. 6, Art. 44), are, therefore, published herewith. Numbers in parentheses refer to pages of the “Manual” on which the new combination was made. ToFIELDIA OCCIDENTALIS Wats. var. intermedia (Rydb.) comb. nov. I’. intermedia Rydb., Bull. Torr. Club 27: 528. 1900. (p. 189) BropiaEA Dovuerasu var. Howellii (Wats.) comb. nov. B. Howellii Wats., Proc. Am. Acad. Sci. 14: 801. 1879. (By error as B. grandiflora var. Howellii, p. 199) HaAseEnaRIA ELEGANS var. multiflora (Rydb.) comb. nov. Piperia multiflora Rydb., Bull. Torr. Club 28: 688. 1901. (p. 219) ERIOGONUM OVALIFOLIUM var. ochroleucum (Small) comb. nov. E. ochroleucum Small, Mem. N. Y. Bot. Gard. 1: 123. 1900. (p. 256) ATRIPLEX PATULA var. obtusa (Cham.) comb. nov. A. angusti- folia var. obtusa Cham., Linnaea 6: 569. 1881. (p. 267) Lewista CotyLepon var. Howellii (Wats.) comb. nov. Calan- drinia Howellii Wats., Proc. Am. Acad. Sci. 23: 262. 1888. (p. 281) ARENARIA ACULEATA var. uintahensis (Nels.) comb. nov. A. uintahensis Nels., Bull. Torr. Club 28: 7. 1899. (p. 286) SILENE orEGANA Wats. var. filisecta comb. nov. S. filisecta Peck, Proc. Biol. Soc. Wash. 47: 186. 19384. (p. 296) ANEMONE GLososa var. lithophila (Rydb.) comb. nov. A. lith- ophila Rydb., Bull. Torr. Club 29: 152. 1902. (p. 809) 'AQUILEGIA FoRMosa var. flavescens (Wats.) comb. nov. A. fla- vescens Wats., Bot. King Geol. Expl. 40th. Par. 5: 10. 1871. (p. 312) Dicentra CucuLiaria var. occidentalis (Rydb.) comb. nov. Bicuculla occidentalis Rydb., Bull. Torr. Club 29: 160. 1902. (p. 323) Draza rEPTANS var. Stellifera (Schulz.) comb. nov. D. caro- liniana f. stellifera Schulz., in Engler, Pflanzenr. 4°°: 333. 1927. (p. 326) Descurainta RicHaRDSONI var. viscosa (Rydb.) comb. nov. Sophia viscosa Rydb., Bull. Torr. Club 29: 238. 1902. (p. 840) Descurainia pinnata var. filipes (Gray) comb. nov. Sisym- brium incisum var. filipes Gray, Pl. Fendl. 8. 1849. (p. 340) DescuRAINIA PINNATA var. halictorum (Cockerell) comb. nov. Sophia halictorum Cockerell, Bull. Torr. Club 25: 460. 1898. (p. 340) 134 MADRONO [Vol.6 DEScURAINIA PINNATA var. paradisa (Nels. & Kenn.) comb. nov. Sophia paradisa Nels. & Kenn., Proc. Biol. Soc. Wash. 19: 155. 1906. (p. 340) Descurarnia PInNATA var. Nelsonii (Rydb.) comb. nov. Sophia Nelsoniti Rydb., Bull. Torr. Club 34: 436. 1907. (p. 341) PaRRYA CHEIRANTHOIDES var. lanuginosa (Wats.) comb. nov. P. Menziesii var. lanuginosa Wats. in Gray Syn. Fl. 1: 152. 1895. (p. 352) Streptanthus hastatus (Wats.) comb. nov. Caulanthus_ has- tatus Wats., Bot. King Geol. Expl. 40th. Par. 5:28. pl. 3. 1871. (p. 3538) Sedum glanduliferum (Hend.) comb. nov. Cotyledon glanduli- ferum Hend., Rhodora 32: 26. 19380. (p. 861) Sedum oregonense (Wats.) comb. nov. Cotyledon oregonensis Wats:, ProcAm. Acad, Sci, 18:2 373.5 18625 (paso) PorENTILLA GRAcILIs Dougl. var. dichroa (Rydb.) comb. nov. P. dichroa Rydb., N. Am. Fl. 22: 819. 1908. (p. 893) HorkE.ia Fusca var. capitata (Lindl.) comb. nov. H. capitata Lindl., Bot. Reg. 23, sub. pl. 1997. 1837. (p. 398) HorkELIia Fusca var. pseudocapitata (Rydb.) comb. nov. dH. pseudocapitata Rydb. in Howell, Fl. N. W. Am. 180. 1898. (p. 398) Horkeia Fusca var. parviflora (Nutt.) comb. nov. H. parwi- flora Nutt. ex. Hook. & Arn. Bot. Beechey Voy. suppl. 338. 18388. (p. 399) Horke.ia Fusca var. filicoides (Crum) comb. nov. Potentilla Douglasii var. filicoides Crum. Leafl. West. Bot. 1: 100. 1984. (p. 399) HorkELIA CoNGESTA var. nemorosa (Keck) comb. nov. H. con- gesta subsp. nemorosa Keck, Lloydia 1: 108. 1988. (p. 399) Geum gracilipes (Piper) comb. nov. Potentilla gracilipes Piper, Bull. Torr. Club 27: 392. 1900. (By error as Geum filipes, p. 402) Rosa pisocarPa var. ultramontana (Wats.) comb. nov. R. cali- fornica var. ultramontana Wats. in Brew. & Wats. Bot. Calif. 1: 87. 1876. (By error as R. pisocarpa var. transmontana, p. 404) CERCOCARPUS LEDIFOLIUs var. hypoleucus (Rydb.) comb. nov. C. hypoleucus Rydb., N. Am. Fl, 22: 424. 19138. (p. 407) Astragalus Laurentii (Rydb.) comb. nov. Homalobus Lau- rentii Rydb., Bull. Torr. Club 51:15. 1924. (p. 443) Astragalus subglaber (Rydb.) comb. nov. Homalobus sub- glaber Rydb., Bull. Torr. Club 51:17. 1924. (p. 444) ASTRAGALUS REVENTUs var. Hoodianus (Howell) comb. nov. A. Hoodianus Howell, Erythea, 1: 111. 1898. (p. 444) Astragalus lagopinus (Rydb.) comb. nov. Xylophacos lago- pinus Rydb., Bull. Torr. Club 52: 3872. 1925. (p. 445) AstracaLus Hooxerianus var. siskiyouensis (Rydb.) comb. nov. Phaca siskiyouensis Rydb., N. Am. Fl. 24: 340. 1929. (p. 446) 1941] PECK: NEW COMBINATIONS 135 AstraGaLus Forwoopu var. wallowensis (Rydb.) comb. nov. Atelophragma wallowense Rydb., Bull. Torr. Club 55: 122. 1928. (p. 447) 3 ASTRAGALUS coNJuNcTUs var. Sheldoni (Rydb.) comb. nov. Tium Sheldoni Rydb., N. Am. Fl. 24: 3938. 1929. (p. 448) ASTRAGALUS LENTIGINoOsuUS var. cornutus (Rydb.) comb. nov. Cystium cornutum Rydb., N. Am. FI. 24: 412. 1929. (p. 449) ASTRAGALUS LENTIGINOSUS var. platyphyllidium (Rydb.) comb. nov. Cystium platyphyllidium Rydb., N. Am. Fl. 24: 410. 1929. (p. 449) VIOLA PRAEMORSA var. major (Hook.) comb. nov. V. Nuttalli var. major Hook., Fl. Bor. Am. 1:79. 18380. (p. 486) VIOLA PRAEMoRSA var. linguaefolia (Nutt.) comb. nov. VJ. linguaefolia Nutt., Torr. & Gray, Fl. 1: 141. 1888. (p. 486) VIOLA PURPUREA var. atriplicifolia (Greene) comb. nov. V/V. atriplicifolia Greene, Pitt. 3: 38. 1896. (p. 486) EPILOBIUM GLANDULOSUM var. Cinerascens (Piper) comb. nov. E. cinerascens Piper, Proc. Biol. Soc. Wash. 31: 75. 1918. (p. 494) EPILOBIUM PANICULATUM var. Hammondi (Howell) comb. nov. E. Hammondi Howell, Fl. N. W. Am. 224. 1903. (p.497) MENZIESIA FERRUGINEA var. glabella (Gray) comb. nov. M. gla- bella Gray, Syn. Fl. N. Am. 2: 39. 1878. (p. 542) PuHLox speciosa var. occidentalis (Dur.) comb. nov. P. divari- cata var. occidentalis Dur., Proc. Acad. Nat. Sci. Phila. 3: 97. 1855. (p. 570) Putiox speciosa var. lanceolata (E. Nels.) comb. nov. P. lan- ceolata E. Nels., Rev. W. Am. Phlox 29. 1899. (p. 570) PHLOX LONGIFOLIA var. compacta (Brand) comb. nov. P. Stans- buryt var. compacta Brand in Engler, Pflanzenr. 4°°°: 67. 1907. (p. 571) PHtox tonairo.ia var. longipes (Jones) comb. nov. P. lineari- folia var. longipes Jones, Contrib. West. Bot. 12: 538. 1908. (p. 571) PHLox Lonairouia var. humilis (Dougl.) comb. nov. P. humilis Dougl. ex Hook. Fl. Bor. Am. 2: 72. 1838. (p. 571) PHLox LoNGIFOLIA var. Calva (Wherry) comb. nov. P. longi- folia subsp. calva Wherry, Proc. Acad. Nat. Sci. Phila. 90: 136. 1938. (p. 571) Puiox Doverasnu var. rigida (Benth.) comb. nov. P. rigida Benth. in DC., Prodr. 9: 806. 1845. (p. 571) Puiox Dovetasu var. Hendersonii (E. Nels.) comb. nov. P. condensata var. Hendersonii E. Nels., Rev. W. N. Am. Phlox 14. 1899. (p. 571) PuHtox pirFrusa var. longistylis (Wherry) comb. nov. P. dif- fusa subsp. longistylis Wherry, Proc. Acad. Nat. Sci. Phila. 90: 139. 1938. (p. 572) PHLox pirFusa var. subcarinata (Wherry) comb. nov. P. dif- 136 MADRONO [Vol. 6 fusa subsp. subcarinata Wherry, Journ. Wash. Acad. Sci. 29: 517. 1989. (p. 572) Puiox pirrusa var. scleranthifolia (Rydb.) comb. nov. P. scleranthifolia Rydb., Mem., N. Y. Bot. Gard. 1: 313. 1900. (p.: 572) Putiox Hoop var. canescens (Torr. & Gray) comb. nov. P. canescens Torr. & Gray, Pac. R. Rept. 27: 122. 1855. (p. 572) APLOPAPPUS CARTHAMOIDES var. rigidus (Rydb.) comb. nov. Pyrrocoma rigida Rydb., Bull. Torr. Club 27: 624. 1900. (p. 712) ApLopaPpPpus RAcEMosus var. halophilus (Greene) comb. nov. Pyrrocoma halophila Greene, Leafl. Bot. Obs. & Crit. 2: 16. 1909. (By error as A. racemosus var. halophiloides, p. 712) APLOPAPPUS RACEMOsUs var. duriusculus (Greene) comb. nov. Pyrrocoma duriuscula Greene, Leafl. Bot. Obs. & Crit. 2:16. 1909. — Go 712) APLOPAPPUS RACEMosuUs var. glomeratus (Nutt.) comb. nov. Homapappus glomeratus Nutt., Trans. Am. Phila. Soc. n. ser. 7: 331. 1840. (p. 712) APLOPAPPUS RACEMosus var. brachycephalus (Nels.) comb. nov. Pyrrocoma brachycephala Nels., Bot. Gaz. 37: 265. 1904. (p.— 713) APLOPAPPUS RACEMOSUS var. congestus (Greene) comb. nov. Pyrrocoma congesta Greene, Pitt. 3:23. 1896. (p. 713) APLOPAPPUS LANCEOLATUs var. tenuicaulis (D. C. Eat.) comb. nov. A. tenuicaulis D. C. Eat., Bot. King Geol. Expl. 40th. Par. 5: 160. 1871, (p) 718) APLOPAPPUS HIRTUS var. sonchifolius (Greene) comb. nov. Pyrrocoma sonchifolia Greene, Leafl. Bot. Obs. & Crit. 2:18. 1909. (p.7 18) 7 APLOPAPPUs HiRTUs var. lanulosus (Greene) comb. nov. Pyrro- coma lanulosa Greene, Leafl. Bot. Obs. & Crit. 2: 16. 1909. (p. 713) APLOPAPPUS UNIFLORUs var. Howellii (Gray) comb. nov. A. Howelliti Gray, Syn. Fl. 1°, suppl.: 446. 1886. (p. 713) ANTENNARIA LUZULOIDEs var. oblanceolata (Rydb.) comb. nov. A. oblanceolata Rydb., Mem. N. Y. Bot. Gard. 1: 409. 1900. (p. 741) ARTEMISIA CAMPESTRIS var. Spithamaea (Pursh) comb. nov. 4. spithamaea Pursh, Fl. Am. Sept. 522. 1814. (p. 768) ARTEMISIA CAMPESTRIS var. pycnocephala (Less.) comb. nov. Oligosporus pycnocephalus Less., Linnaea 6: 524. 1881. (p. 768) ARTEMISIA CAMPESTRIS var. pacifica (Nutt.) comb. nov. A. pacifica Nutt., Trans. Am. Phila. Soc. ser. 2,7: 401. 1841. (pp. 768) SENECIO TRIANGULARIs var. trigonophyllus (Greene) comb. nov. S. trigonophyllus Greene, Pitt. 3: 106. 1896. (p. 780) CREPIS RUNCINATA var. imbricata (Babe. & Steb.) comb. nov. C. runcinata subsp. imbricata Babc. & Steb., Carneg. Inst. Wash. Publ. 504: 102. 1938. (p. 804) 1941] STEBBINS AND LOVE: STIPA 137 CREPIS MODOCENSIS var. Subacaulis (Kellogg) comb. nov. C. occidentalis var. subacaulis Kellogg, Proc. Calif. Acad. Sci. 5: 50. 1873. (p. 805) CREPIS ATRIBARBA Var. Originalis (Babe. & Steb.) comb. nov. C. atribarba subsp. originalis Babe. & Steb., Carneg. Inst. Wash. Publ. 504: 162. 1938. (p. 806) CREPIS OCCIDENTALIs var. pumila (Rydb.) comb. nov. C. pumila Rydb., Mem. N. Y. Bot. Gard. 1: 462. 1900. (p. 806) Crepis Baxeri var. Cusickii (Eastw.) comb. nov. C. Cusicki Eastw., Bull. Torr. Club 80: 502. 1903. (p. 806) Willamette University, Salem, Oregon, August 9, 1941. AN UNDESCRIBED SPECIES OF STIPA FROM CALIFORNIA G. L. STepsins, Jr., AND R. M. Love Due to the growing interest in the native forage plants of Cali- fornia, particular attention is now being given by a number of workers to Stipa pulchra Hitchc., one of the most common and val- uable of the native perennial forage grasses in the valley and foothill regions of this state. Several people, including the pres- ent writers, have noticed that this species as recognized in the current manuals actually consists of two distinct types. One, with deep green foliage, relatively broad leaves, stiffer panicle branches, large glumes, thick, fusiform lemmas, and stout, stiff awns, is predominant in the outer Coast Ranges and the wooded parts of the Sierra Nevada foothills. This is typical S. pulchra, of which the type came from Healdsburg, Sonoma County. The other form, with somewhat glaucous foliage, narrower leaves, flexuous, often nodding panicle branches, smaller, narrower glumes, slender lemmas, and slender, often flexuous awns, occurs chiefly in the treeless parts of the inner Coast Ranges, the San Joaquin Valley (in scattered areas undisturbed by cultivation), the valleys of southern California, and the edges of the deserts. The two types have been given different common names, typical S. pulchra being known as purple needle grass, and the slender, interior type as nodding needle grass. The writers have observed these two needle grasses carefully during two seasons of collect- ing in the field, have grown and compiled extensive morphological data on several cultures of each, and have examined their chromo- somes at both mitosis and meiosis. From these observations enough evidence has accumulated to warrant the recognition of the interior type as a distinct species. It may be described as follows. Stipa cernua sp. nov. Folia glauca angusta; panicula ampla, ramis tenuis, flexuosis, cernuis; glumae inaequales, 12-19 mm. longae, 1-1.4 mm. latae, pallidae vel roseo-purpureae, semper 3-nervatae; lemma angusta, 5—10.5 mm. longa, cum fructu 0.6—1 138 MADRONO [Vol. 6 LEGEND @ @ S. PULCHRA O S. CERNUA @® O © S. PULCHRA & S. CERNUA TOGETHER O 3, O [) ® o® © SAN FRANCISCO 'N CALIFORNIA @ Fic. 1. Distribution of Stipa pulchra and S. cernua. mm. crassa, ad basim et supra nervis pubescens, parte superiore glabra; arista 6-11 cm. longa, flexuosa, scabra vel ad basim pubescens. | Mostly in large clumps, the basal leaves numerous, narrow, usually glaucous; culms several, mostly 60 to 90 cm. tall, middle culm leaves 1.2—2.4 mm. broad. Panicle ample, the basal portion often within the lowermost leaf sheath; panicle branches slender and flexuous or cernous; glumes scarious, pale or reddish purple, long acuminate, the lower 12-19 mm. long, the upper somewhat shorter and broader, 1—-1.6 mm. broad, both strictly 3-nerved; lemma 5—10.5 mm. long, 2.2—2.9 mm. broad when unrolled, with caryopsis 0.6—-1 mm. thick at maturity, 5- or sometimes 7-nerved, the callus acute, pubescence dense on the callus and on the lower 1941] STEBBINS AND LOVE: STIPA 139 Piatt 13. INFLORESCENCE AND Fiorets 1n Stipa. Fic. 1. Stipa pulchra Hitche., from Berkeley (Stebbins 2670), panicle branch X .33. Fie. 2. 8S. cernua, from Santa Barbara County (Stebbins 2875), panicle branch x 33. Fic. 3. S. pulchra, from Berkeley, glumes and lemmaX3.3. Fic. 4. S. cernua, type collection, glumes and lemma x 3.3. 140 MADRONO [Vol. 6 TABLE 1 Comparison of the Principal Characteristics of Stipa pulchra and Stipa cernua Width of next to uppermost culmecleat, (tear ee Colorsot leaves tac03 3 eee Character of panicle branches Length lower glume ........ Width upper glume ........ Number of nerves on upper SMMC oes kere Length of lemma’.......... Width of lemma (unrolled) . Number of nerves on lemma Thickness of mature cary- opsis including lemma .... Pubescence of lemma ....... S. pulchra 2.4-6 mm. deep green spreading, or slightly cernuous 15-26 mm. 1.4-2.2 mm. 3-5 75-13 mm. 3.3-4.8 mm. 5-9 1.0-1.4 mm. throughout, or at base and = on nerves to mid- dle or summit S. cernua 1.2-2.4 mm. glaucous slender, flexuous, or cernuous 12-19 mm. 1.0-1.6 mm. 3 5.0-10.5 mm. 2.2-2.9 mm. 5-7 0.6-1.0 mm. at base and on nerves’ one-third to four-fifths of way to summit Lenethoor sawn) 2 oe 60-90 mm. 60-110 mm. Ratio of awn to lemma ..... 5.9-9.8 8.7-14.5 Pubescence of awn _ below nrst-bendée.. Ws. 5 ae eae pubescent scabrous or slightly pubescent | one-fifth to one-fourth of the body of the lemma, the upper part of the lemma scabrous and with the nerves pubescent one-third to four-fifths of the distance to the summit, lemma usually pale straw- or buff-colored, occasionally purplish, brownish, or black- ish; awn 6-11 ecm. long, 8.7—14.5 times the length of the lemma, twice bent at maturity, slender and flexuous beyond the second bend, scabrous or short-pubescent below the first bend. California, from Tehama County south to San Diego County east to the Sierra Nevada foothills, reaching the coast from Mon terey southward, and extending up to about 1360 meters (450¢ feet) altitude in southern California. 7 Type. West side of Cedar Mountain Ridge, southeast of Liver- more, Alameda County, altitude 400 meters (1300 feet), G. L. Stebbins, Jr. 2732, (Herb. Univ. Calif. 641631). The follow- ing specimens in the University of California Herbarium are also typical: west of Orland, Glenn County, Heller 11434; Nacimiento River, Monterey County, Davy 7688; Santa Barbara, Elmer 3874; Mount Pinos, Ventura County, Hall 6426; Mint Can- yon, east of Saugus, Los Angeles County, Munz 6794; near Rialto, San Bernardino County, S. B. §& W. F. Parish 2038 ; San Bernardino Valley, Parish 6204, 11257; near Winchester, Riverside County, Hall 2921; Box Springs Mountain, Riverside County, Hall 2977; San Diego, Brandegee 833. As mentioned previously, Stipa cernua is most closely related to S. pulchra. The differences between the two species are sum- 1941] STEBBINS AND LOVE: STIPA 141 marized in Table 1, and partly illustrated in plate 13. The follow- ing may be used as convenient key characters: ) Leaves green, the middle culm leaves 2.4-6 mm. broad; lemmas fusiform, with caryopsis 1-1.4 mm. thick at maturity; awn stout and stiff, mostly 7-9 times as long as the lemma ....... S. pulchra Leaves somewhat glaucous, the middle culm leaves 1.2-2.4 mm. broad; lemmas slender, with caryopsis 0.6-1 mm. thick at maturity; awn slender, flexuous beyond the second bend, mostly 9-12 times the length of the lemma ......................... S. cernua The distribution of the two species, as determined by the col- lections and observations of the writers as well as by the speci- mens in the University of California Herbarium, is shown in figure 1. It will be seen that they occur together throughout a rather large area, and can often be found growing side by side. In these places they are usually quite distinct, but forms inter- mediate between them do occur. The writers studied particularly these intermediate plants as found in two localities near San Benito, San Benito County, and found them to be completely sterile or nearly so. Although the surrounding plants of S. pulchra and S. cernua were producing good seed in abundance, not a single fertile grain was found on any of the intermediates. The latter showed considerable hybrid vigor, often forming clumps much larger than those of the parent species. In the University of Cali- fornia Herbarium there are three specimens of such sterile inter- mediates: north base of Mount Hamilton, Santa Clara County, Sharsmith 664A; Pacific Grove, Monterey County, Elmer 3507; Las Flores Canyon, Santa Monica Mountains, Los Angeles County, Epling in 1930. Only the latter specimen had anthers with pollen for examination; in it 5 per cent of the pollen grains were large and well filled with cytoplasm, as contrasted with 90— 96 per cent in typical S. pulchra and S. cernua. The chromosome numbers of the two species are reported else- where as 2n = 64, n= 82 for S. pulchra, and 2n=70, n=8385 for S. cernua (Stebbins and Love, Am. Jour. Bot. 28: 371-382. 1941). Three of the sterile intermediate plants from San Benito were dug up and transplanted to pots in Berkeley, and their somatic chro- mosome number was determined as 2n = 67, indicating that they are actually first generation hybrids between S. cernua and S. pulchra. Their meiosis will be studied during the coming season. Except for S. pulchra, S. cernua has no close relative among the North American species of Stipa. It resembles S. comata in its glaucous leaves and long awns, but that species has large, thick lemmas as in S. pulchra, and lacks the collar at the apex of the lemma which is found in both S. pulchra and S. cernua. It is pos- sible that S. cernua is related to some of the numerous South American species of Stipa, but the present writers have not seen adequate material of any of them to judge their relationships. College of Agriculture, University of California, Berkeley, January, 1941 142 MADRONO [Vol. 6 REVIEWS The Evolution of Land Plants | Embryophyta|. By Doveuas Hoventon CampBeLyt. Pp. 1-731 with 351 text figures. Stan- ford University Press. 1940. This large volume has two main elements. It is a résumé of the author’s half century of morphological, chiefly embryological, study; and it is a condensed summary of the work of other writers on the relationships of the higher plants. Of these, the former is the more valuable, which is natural since Dr. Campbell has him- self been the foremost contributor to our understanding of the broader lines of the evolution of the land plants. It is as to these broad lines, where the work of the past fifty years has brought confidence in some things and doubt as to others, that this book registers well the present state of science and can serve as a mile post. The larger part of the book is taken up by the finer classifica- tion, to orders, families, in some places to genera. Here the author depends more upon the views of others, and the presenta- tion is distinctly less authoritative. For example, among the ferns, the Eusporangiatae are well presented, but the treatment of the higher Leptosporangiatae is comparatively weak. In accord with custom, the evidence of paleophytology is treated with respect. This science has of course made progress in various respects. But the reviewer would still recall a remark of Dr. Joseph Hooker. “Amongst the many collections of fossil plants that I have examined, there is hardly a specimen, belonging to any epoch, sufficiently perfect to warrant the assumption that the species to which it belonged can be recognized.” Yet, specific characters may petrify better than those of classes. Consider spermatozoids and the embryo-sac. A discussion of the nature and value of evidence would be a valuable introduction to a book of this kind. Next to its completeness, the most marked characteristic of the book is its lack of dogmatism. Correlated with caution is a tendency to entertain the idea of multiple origin of apparently natural groups, and to admit question, even where affinity seems best established. Two examples: the probable central position of some such plant as Anthoceros in the ancestry of Embryophyta is perfectly presented, and the group “‘Anthocerotes” is made a class, coordinate with Hepaticae and Musci. The latter are treated as probable derived groups, but the derivation seems to be pictured as exceedingly ancient, from primeval Anthocerotes, of which “Of course, the sporophyte . . . must have been much simpler than in any living forms—perhaps comparable to that of such liverworts as Riccia or Sphaerocarpus.” To the reviewer, the stoma, common to Anthoceros and many mosses, provides positive proof that their common ancestor, if not exactly Anthoceros itself, had at any rate a sporophyte independent enough in its nutrition to have evolved this structure. 1941] . REVIEWS 143 As to the angiosperms, polyphylesis is explicitly advocated. The characteristic structure of this group is not seed, nor flower, nor pollen tube, nor trachea; it is the embryo-sac. If the hypoth- esis of multiple origins means that this structure has been evolved several times independently, it is hard to accept. Even more than the stoma, it requires good evidence of repeated evolution before it is questioned as proof of real affinity. There are 351 numbered figures, most of which are composed of a considerable number of drawings, largely original and well reproduced. They add materially to the value of the book. The text is a remarkable mine of detailed information. How much there is of this ntay be shown by the index, which occupies 37 pages of fine print, two columns to the page, and is still incom- plete; thus, under “stoma” there is no reference to the text, and under “‘embryo-sac”’ there is only one.—E. B. Copreitanp, Depart- ment of Botany, University of California, Berkeley. Sinopsis de la Flora del Cuzco. Fortunato L. Herrera. Tomo I. Parte Sistematica. Pag. 1-528. Publicado bajo los auspicios del supremo gobierno. Lima, Peru, 4 de Julio, 1941. This check list of the plants of the Department of Cuzco by the distinguished Peruvian botanist is by far the most complete of several similar works by the same author, the first of which ap- peared in 1919. It lists 2166 species (with a few varieties) 588 of which are cryptogams, about 250 of these being ferns. and fern allies. Even so, the author suggests that probably only about one-half of the species growing within the area have been re- corded. The predominant families are Compositae, Gramineae, and Leguminosae. The names are accompanied, at least for the phaenerogams, by source of publication and citation of specimens. The latter are given in detail, usually including altitude, information which will be invaluable in any study of the flora; habitats, however, are rarely indicated. Often the range of the plant outside of Cuzco, if known, is mentioned; there are some economic notes. An appendix contains descriptions of new species based on the author’s collections. There is also a list of native names and their scientific equivalents, and an index to the genera. In a work of this nature, based of necessity on the literature available—of which there is a good bibliography—there are of course always omissions; on the other hand there are a few addi- tions to the flora of Cuzco. In supplements, which it is to be hoped will be issued from time to time, it would be well to give the source of determination and to indicate where the collections may be consulted in order that identifications may be checked when desired. The work would be more consistently useful, too, if publication citations were always given (which is obviously the intent but they are not infrequently omitted). Most of the typo- graphical errors will easily be corrected; only one mistake in the 144 MADRONO ; [Vol. 6 presentation of the material has been noted, namely the including of the composite Orthopappus, on page 321, in the Melasto- mataceae. The author in preparing the work and the Peruvian Govern- ment in publishing it have made a meritorious contribution to the scientific study of the rich and useful flora of Cuzco; may there be many more similar endeavors based increasingly on the activities of Peruvian students.—J. F. Macsripre, Field Museum of Natural History. A Flora of Arizona and New Mexico. By Ivar Tipestrom and Sister TeresiTa KiTTELL. Pp. xxvi+ 897 with frontispiece. The Sn University of America Press, Washington, D. C., 1941. 5.00. The flora of Arizona and New Mexico, listing 898 genera and 3975 species, is arranged according to the systems of DeCandolle and Bentham and Hooker with some slight emendations, chief of which is the arrangement of the orders and families in a descend- ing numerical sequence as to the number of cotyledons. Hence the Coniferae with many cotyledons come first and the ferns and fern allies with none appear at the close of the work. The keys are brief and to the point and brief descriptions aid materially in amplifying the keys. There is a general citation of habitat and range accompanying each entity. The work is ambitious and as such is worthy but one cannot read it without a feeling of regret. Much of the advance in botany of the past fifteen years is ignored. Many monographs which have appeared during this time are not alluded to either as to the species accepted or in the synonymy. The frontispiece is a map of Arizona and New Mexico showing the major rivers and the two thousand foot contour intervals. There are fifteen circles indicating localities but no evident refer- ence to these in the text. On the other hand the table of contents refers to the map as showing the “belts of vegetation.” The reason for this confusion is not clear to the reviewer. The volume is lithoprinted and would have been materially improved by either a little more space between the species or by underlining the species names. As it stands the pages appear crowded and the typography does not invite the reader’s atten- tion. In many instances the craftsmanship of both typist and printer is definitely at fault—Hersert L. Mason. NOTES AND NEWS On May 24, 1941, the University of California conferred the honorary degree of doctor of laws on Dr. Willis Linn Jepson, Pro- fessor of Botany Emeritus of that institution. The honor is in recognition of Dr. Jepson’s contribution to our knowledge of the California flora and his long and successful promotion of forest conservation in the state. 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 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 VI NUMBER 5 MADRONO A WEST AMERICAN JOURNAL OF BOTANY gi eT NAEP. mn, nes ‘AQONIAN INS7/FE2S wo iv aN Contents _ A Synopsis or THE AMERICAN Spectres oF Cicuta, Mildred E. Mathias and UIC ONGLADICON hi ween tere nt ete emu O NEU a ye Oa Na Wow hai 145 Great Basin Prants—VI. Nores on GENTIANA, Bassett Maguire ........ 151 Arnica In ALASKA AND YUKON, Bassett Maguire .................0000005 153 _ A Comparison or THE Empryoceny or Picea anp Asixs, J. T. Buchholz .... 156 _ ANATOMY AnD EcoLocy oF AMMOPHILA ARENARIA LINK, Edith A. Purer ... 167 An UNDEscrizep SPeciIes or CEANOTHUS FROM CaLirorniA, Howard EH. Mc- ee coli PE a TRS ECT ES Ee Sea Ca a gaan a etn A ae aE: Gn 171 - Norss anp News: Range Extensions in Species of Western North America, Boykinia Jamesii var. heucheriformis and Saxifraga eriophora (Bassett Maguire), Eriodictyon capitatum (John Tucker); News ............. 173 _ PRocEEDINGs oF THE CALIFORNIA BOTANICAL SOCIETY ..................-. 175 Published at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania January, 1942 MADRONO A WEST AMERICAN JOURNAL OF BOTANY Board of Editors Hersert L. Mason, University of California, Berkeley, Chairman. LeRoy Asrams, Stanford University, California. Epcar ANnpERsSON, Missouri Botanical Garden, St. Louis. Lyman Benson, University of Arizona, Tucson. Hersert F. Copetanp, Sacramento Junior College, Sacramento, California. Ivan M. Jounston, Arnold Arboretum, Jamaica Plain, Massachusetts. Mirprep E. Maruias, University of California, Berkeley. Bassett Macurre, Utah State Agricultural College, Logan. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board—EtuHet Crum Department of Botany, University of California, Berkeley Business Manager—Wiut1i1am 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. Volume I, Numbers 1 to 17, complete, $5.00. Volume II, Numbers 1 to 17, complete, $5.00. Volume III, Numbers 1 to 8, complete, $5.00. Single num- bers $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 $3.70; 100 copies $4.10; additional 100’s .75¢; 8 pages, 50 copies $5.40; 100 copies $6.00, additional 100’s $1.20. Covers, 50 for $2.50; additional covers at $1.50 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: Ernest B. Babcock, University of California, Berkeley. First Vice-President: Roxana S. Ferris, Stanford University, California. Second Vice-President: Palmer Stockwell, Institute of Forest Genetics, Placerville, California. Treasurer: Wiiliam Hiesey, Carnegie Institution of Washington, Stanford University, California. Secretary: Lincoln Constance, Department of Botany, 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. 1942] MATHIAS AND CONSTANCE: CICUTA 145 A SYNOPSIS OF THE AMERICAN SPECIES OF CICUTA Mixtprepo EF. Maruias anp LIncoLtn CONSTANCE The water hemlocks are of especial interest because of their poisonous properties, which make them a menace both to livestock and human beings. In a recent paper, Bomhard (1) remarks that, “The extremely toxic character of the underground parts ranks this genus as one of the most virulently poisonous groups of flowering plants native to the North Temperate Zone.” Hence, a great deal of space has been devoted to them in agricultural bulletins, some of which have directed considerable attention to the specific differences allegedly discernible in the subterranean portions of the plants. It is the purpose of the present paper to * evaluate the characters which have been used to delimit species in the genus, and to propose a somewhat revised taxonomic treat- ment. Specific descriptions for Cicuta will appear in a forthcom- ing number of “North American Flora.” In their pioneer revision of North American Umbelliferae in 1888, Coulter and Rose (2) referred the known members of the genus in America to three species, the European Cicuta virosa L. (with two varieties), C. Bolanderi Gray, a plant endemic to the California salt marshes, and C. bulbifera L. Cicuta bulbifera, easily separable from the other species by virtue of the usually asexual reproduction through the agency of bulblets in the axils of the upper foliage leaves, will not enter into the following discussion. This rather simple picture of the genus was upset the following year by Greene (4), who excluded C. virosa from America, and described three additional species from the western United States, C. occidentalis, C. purpurata, and C. vagans. At the same time, he enunciated his faith in the value of vegetative characters in dis- tinguishing the species of water hemlocks in the following words: “In the few families of plants which are, like the Umbelliferae, preéminently natural, the anthological and carpological charac- ters, whether of genera or of species, are apt to be very slight. But here Nature comes usually to the rescue of the despairing car- pological systematist, and gives him good characters for his genera, or for his species, in the vegetative organs. Only by re- garding these latter can a man set good limits to species in such a genus as Cicuta.” Again, in November of the same year, he re-emphasized (5) and elaborated this view, as follows: “The fact is well recognized, or should be, by descriptive botanists, that in herbaceous plants of all kinds, characters of the roots or other subterranean organs are of the very best for specific distinctions. Those of pubes- cence, foliage, and to some extent, of the flower also, are less con- stant within specific limits than are the peculiarities of the root, when the root happens to have peculiarities, which is however by Manproxo, Vol. 6, pp. 145-176. January 17, 1942. 146 MADRONO [ Vol. 6 no means the rule in nature. Most commonly the roots, rhizomes, tubers and other such organs will be much the same throughout. the whole group—a long series of species, or even an entire genus. In an order so extremely natural as that of the Umbelliferae, in which the fruits are so similar that plants of the same carpology are sometimes placed in different genera in deference to merely vegetative differences, it would seem altogether unphilosophical to require of the fruit that it furnish specific characters; or, to assume that unless the supposed members if a genus can be dis- tinguished carpologically the species is but one.” Coulter and Rose, in their later treatment of the Umbelliferae (3), accepted two of the three species Greene had described, but judged C. purpurata to be conspecific with C. Douglasi (DC.) C. & R. They added to the list of species C. Curtissii, conspecific with C. mexicana which they had already described earlier in the same year. Although they accepted Greene’s species, they were not entirely convinced of the validity of the characters he had origin- ally employed in distinguishing them: “We reproduce the above key in the hope that it may be further tested in the field, for with the material at our command we have not been able to follow it fully. While we recognize in Professor Greene’s typical mate- rial the differences suggested, we do not find them constant. The fleshy thickening of the rootstocks and their direction, as well as the thickness and elongation of the roots, seem to vary with the nature of the substratum, as might be expected.” The synoptical key offered by Greene, modified by the inclusion of the species he later described follows. * Root axis very short, nearly or quite erect, not enlarged, its partitions crowded. + Roots all alike, slender fibrous. C. virosa. ++ Main roots coarse, elongated, fleshy fibrous. C. Bolanderi, C. occidentalis, C. purpurata, C. frondosa, C. arguta, C. Sonne. +++ Main roots oval or oblong, fleshy tuberiform. C. maculata, C. bulbifera, C. subfalcata. ** Rhizomatous species; the root axis greatly enlarged, hori- zontal, only partly or not at all subterranean, emitting fibrous roots from beneath. C. vagans, C. californica, C. grandifolia, C. dakotica. Incapable of placement in this scheme are the three following species, also described by Greene: C. valida, “stature of the plant and its underground parts not known’; C. fimbriata, described only from the foliage; C. ampla, “known to me only in the fruiting summit of a single plant.” EXPLANATION OF THE Figures. PLatTe 14. ° Pirate 14. Frurrs or Cicura. Longitudinal view (x 6.5) and cross section (x 9.5). Fras. 1, 2, C. Bolanderi; 3, 4, C. virosa; 5, 6, C. mexicana; 7, 8, C. Doug- lasii; 9, 10, C. Victorinii; 11, 12, C. mackenzieana; 13, 14, C. maculata. MATHIAS AND CONSTANCE: CICUTA 147 1942] Puatr 14. Fruits or Cicuta. 148 MADRONO [Vol.6 Illustrating their scepticism of the very foundation of Greene’s classification, Coulter and Rose proceeded to base their own key on their observation that the “species seem to be best grouped primarily by their oblong or orbicular fruits, further separation being made upon differences in the fruit ribs and in the foliage.” Greene, however, seems not to have been at all discouraged and described C. grandifolia in 1907, and C. frondosa, C. subfalcata, C. dakotica, C. arguta, C. valida, C. Sonnei, C. fimbriata and C. ampla in 1912. Cicuta cinicola A. Nels. was added from Idaho, also in 1912, and recently C. mackenzieana Raup has been described from northern Canada and C. Victorini Fernald from the mouth of the St. Lawrence River. Authors of manuals and floras have not been prone to accept Greene’s second deluge of segregates, but have admitted C. occidentalis or C. vagans or both to their treat- ments of the appropriate areas. In preparing a revision of this genus for our account of the Umbelliferae in the “North American Flora,’ we have had occa- sion to re-examine the material on which Greene based his con- clusions. We are in agreement with him that Cicuta virosa is to be excluded from North America, although its relationship to C. mackenzieana appears to be rather close. We question the whole thesis that the underground parts of the water hemlocks afford specific characters, at least of the nature noted by Greene, and believe that they are so susceptible of modification in relation to soil composition and fertility, and especially to soil aération, as to be essentially useless in this connection. Furthermore, we are unable to attach sufficient importance to the degree of leaf-divi- sion to warrant the retention of C. californica as distinct from C. Douglasii, and find that Coulter and Rose’s distinction between “oblong” and “‘orbicular” fruits sometimes breaks down on the same plant. The diagnostic value of leaf-venation as a means of distinguishing Cicuta from other umbelliferous genera, and also for separating individual species in the former genus, has been stressed recently by Bomhard (1). She distinguishes C. cali- fornica from the other species by the fact that the secondary veins of the leaflets are directed at the marginal teeth, rather than at the sinuses between them. Our studies, however, have not enabled us to find any marked difference in this entity, which would warrant its continued retention as a species. On the other hand, characters of the fruit, especially its com- pression, the proportion between the dorsal and lateral ribs, their relation to the intervals, and the nature of the oil tubes and the seed do seem to enable us to distinguish natural species in the genus. Our treatment, admits the following species for North America: C. Bolanderi, C. meaicana, C. Douglasi, C. mackenzieana, C. maculata, C. Victorinit and C. bulbifera. All of the species pro- posed by Greene will thus be relegated to the synonymy which they appear richly to deserve. 1942] MATHIAS AND CONSTANCE: CICUTA ; 149 Taxonomic TREATMENT Grcuta L. Sp. Pl. 12 255. 1753. Creutaria Lam. FI]. France 3: 445. 1778. Keraskomion Raf. New. Fl. Amer. 4: 21. 18386. Type species: Cicuta virosa L. Axils of the leaves not bulbiferous. Fruit constricted at the commissure; lateral ribs about equalling the dorsals in surface display. Oil tubes large (pl. 14, figs. 2, 6); seed oily, evidently channeled under the tubes. Ribs narrower than the oil tubes: plants of Pacific GoOase Gall MALESHES! fies 8 Sana e ees os 1. OC. Bolanderi. Ribs broader than the oil tubes; plants of coastal eastern and southeastern United States and ECASLEIM, WLCXICO!. . 1g.) eo Sear he eS 2. C. mexicana. Oil tubes small (pl. 14, figs. 8, 12); seed less oily, un- channeled or only slightly channeled under the tubes. Fruit oval to orbicular, at least as long as broad, 2-4 mm. long, 2-3 mm. broad; rays 12-20, 2-6 cm. long; pedicels 3-8 mm. long .............. 3. C. Douglasit. Fruit elliptical, conspicuously broader than long, 1.5-2.2 mm. long, 2-3 mm. broad; rays 7-14, 7-8 cm. long; pedicels 7-12 mm. long ........ 4. C. mackenzieana. Fruit not constricted at the commissure; lateral ribs much broader than the dorsal in surface display. Leaflets coarsely serrate to incised; fruit oval to orbicular, rounded at apex and base; lateral ribs prominent, about equalling the intervals ........ 5. C. maculata. Leaflets finely serrate; fruit ovate, narrowed toward apex, cordate at base; lateral ribs obscure, much narrower than the intervals ................... 6. C. Victorinii. Axils of the leaves bulbiferous ......................0. 7. C. bulbifera. 1. Crcuta BoLtanperi Wats. Proc. Amer. Acad. 11: 189. 1876. Type locality. Suisun, Solano County, California, in salt marshes, Bolander. Distribution. Salt marshes of central and southern Cali- fornia. Representatives. Davy 4106, 6668, 6789, 6877; Heller 7541. 2. CicuTA MEXICANA Coult. & Rose, Proc. Wash. Acad. 1: 145. (January) 1900. C. maculata L. sensu Hemsl. Biol. Centr.-Amer. Bot. 1:566. 1879-81, not C. maculata L. 17538. C. Curtissii Coult. & Rose, Contr. U. S. Nat. Herb. 7: 97. (December) 1900. C. maculata var. Curtissii Fern. Rhodora 41: 439. 19389. Type locality. Coatzacoaleos, Isthmus of Tehuantepec, Vera Cruz, Chas. L. Smith 1161, Distribution. New Jersey to Florida, south and west to Tamaulipas, Nuevo Leon and Vera Cruz. Representatives. Curtiss 6845; Heller 1165; Palmer 445; Pringle 10,804. 38. Cicuta Doverasi (DC.) Coult. & Rose, Contr. U. S. Nat. Herb. 7: 95. 1900. Sium Douglasti DC. Prodr. 4: 125. 18830. Cicuta maculata L. sensu Hook. & Arn. Bot. Beechey Voy. 142. esaiy> not L. 1753. C. californica Gray, Proc. Amer. Acad. 7: 150 MADRONO [Vol. 6 344, 1867. C. crassifolia Nutt. Rept. Wilkes Exped. 17: 3816. 1874. C. virosa var. californica Coult. & Rose, Rev. N. Amer. Umbel. 1380. 1888. C. occidentalis Greene, Pittonia 2:7. 1889. C. occidentalis f. frondosa Greene, op. cit. p. 7. C.purpurata Greene, op. cit. p.8. C.vagans Greene, op. cit.p.9. C.grandifolia Greene, Leafl. Bot. Obs. 2: 124. 1909. C. Douglasii var. occidentalis Jones, Bull. Univ. Mont. Biol. ser. 15, 42. 1910. C. Sonnei Greene, Leafl. Bot. Obs. 2: 239. 1912. C. subfalcata Greene, op. cit. p. 237. C.frondosa Greene, op. cit. p. 236. C. valida Greene, op. cit. p. 238. C. fimbriata Greene, op. cit. p. 240. C. cinicola A. Nels. Bot. Gaz. 54: 141, fig. 1. 1912. C. occidentalis f. californica Wolff ex Engl. Pflanzenr. 4°78: 90: 82. 1927. C. occidentalis f. oregonensi-idahoensis Wolff, op. cit. p. 82. C. occidentalis f. arizo- nensis Wolff, op. cit. p. 82. C. occidentalis f. wyomingensis Wolff, Op City. (82; Type locality. “In America boreali-occid.,’’ Douglas. Distribution. Alberta and Montana to western Alaska, south to California, New Mexico, Arizona and Chihuahua. Representatives. Baker 655; Brown 501; Cusick 2556, 2779; Heller 7174; Nelson & Macbride 1815; Townsend & Barber 57. 4. CICUTA MACKENZIEANA Raup, Journ. Arn. Arb. 17: 279, pl. 197. 1986. Type locality. Sandy margin of a lagoon near the south shore of Lake Athabaska about 1.5 miles west of Ennuyeuse Creek, Canada, August 25, 1935, Raup 6976. Distribution. Hudson Bay to the Mackenzie Basin. Representatives. Macoun 79,261; Raup 6764, 6964. 5. Crcura macutaTa L. Sp. Pl. 1: 256. 1758. C. maculata Lam. Encycl. 2: 2. 1786. C. virosa var. maculata Coult. & Rose, Rev. N. Amer. Umbel. 1380. 1888. C. dakotica Greene, Leafl. Bot. Obs. 2: 237. 1912. C. arguta Greene, op. cit. p. 2388. G: ampla Greene, op. cit. p. 241. C. dakotica var. pseudovirosa Lun- nell, Amer. Midl. Nat. 4: 486. 1916. C. dakotica var. pseudo- maculata Lunell, op. cit. p. 486. Type locality. Virginia, Kalm. Distribution. Prince Edward Island and Quebec to North Carolina and Tennessee, west to North Dakota and Texas. Representatives. Fernald & St. John 1141; M. L. Grant 3238; Heller 1002; Lindheimer 615. 6. Cicuta Victrorinu Fern. Rhodora 41: 441, pl. 561, figs. 1-2. 1939. Type locality. Tidal flats of the St. Lawrence River, Quebec: “greves intercotidales, Cap Rouge pres du Pont de Quebec, 9 aout 1922,” Victorin 15,479. Distribution. Known only from the estuary of the St. Lawrence River, Quebec. Representatives. Fernald & Long 24,249; Victorin 15,480. 1942] ' MAGUIRE: GREAT BASIN PLANTS 151 7, CIcUTA BULBIFERA L. Sp. Pl. 1: 255. 1753. Cuicutaria bulbi- fera Lam. Encycl. 2:3. 1786. Keraskomion bulbiferum Raf. New. Pl4:-21. 1886. Type locality. Virginia, Canada, Clayton. Distribution. Newfoundland and Quebec to Delaware and Pennsylvania, west to British Columbia and Oregon. Representatives. Cusick 2966; Heller §& Heller 551; Sandberg, MacDougal 6 Heller 789. Department of Botany, University of California, Berkeley, May, 1941. LirerAtTure C1rep 1. Bomuarp, Mira1am L. Journ. Wash. Acad. Sci. 26: 102-107. 1936. 2. Couuter, Joun M. & J. N. Rose. Revision of North American Umbelliferae. 129-130. 1888. 3. —. Contr. U.S. Nat. Herb. 7: 93-100. 1900. 4. GREENE, E. L. Pittonia 1: 271-272. 1889. 5, —————_—_—.. Pittonia 2: 1-1l. 1889. GREAT BASIN PLANTS—VI. NOTES ON GENTIANA Bassett MaAGuIRE Continuing the series of minor papers discussing plants of the Great Basin, these notes are concerned with the delineation of a newly recognized geographical population of Gentiana calycosa, and the confirmation of a range extension of G. barbellata. All specimens herein cited are on deposit at the Intermountain Herbarium. The United States Forest Herbarium, Washington, D. C., is designated by the symbol USFH. GENTIANA BARBELLATA Engelm. The following collection con- firms the queried inclusion of the species within the Utah range by Tidestrom. Urau. Sanpete County: frequent, stony, steep, west-facing slopes, summit of Horse Shoe Mountain, South Peak, 12,000 feet, Manti National Forest, August 11, 1940, Maguire 20059. GENTIANA CALycosa Griseb. subsp. typica nom. nov., G. calycosa Griseb. in Hook. Fl. Bor. Amer. 2: 58. 1838. Dasystephana ob- tus.loba Rydb., Bull. Torr. Bot. Club 40: 464. 1918. D. monticola Rydb. l.c. GENTIANA caLycosa Griseb. subsp. asepala subsp. nov. Herbae perennes parvulae; caulibus decumbentibus, 5-10(12) cm. longis; foliis ovatis vel ovato-ellipticalibus 1.0-1.5(1.7) cm. longis; flori- bus solitariis; corollis 3-4 em. longis; calicibus membranaceis, graviter incisis, lobis obsoletis vel inconspicuis subulatisque, rare excedentibus 1-1.5 mm. Low perennial herbs; stems decumbent, 5—10(12) cm. long; leaves ovate to ovate elliptical 1.0-1.5(1.7) cm. long; flowers solitary, corolla 8-4 em. long; calyx membranaceous, character- 152 MADRONO [Vol.6 istically deeply incised on two sides, lobes obsolete, or incon- spicuous and subulate, rarely more than 1—1.5 mm. long. Type. Meadows about seepage areas, southeast slopes, saddle west of Mount Agassiz, 11,500 feet elevation, Uinta Moun- tains, Duchesne County, Utah, August 15, 1933, B. Maguire, Ruth Maguire, & A. G. Richards, 4225. Specimens examined. IpauHo. Custer County: meadow, Steward Canyon, 8000 feet, Lemhi National Forest, July 17, 1981, S. L. Jacobs 90 (USFH, 65577). Idaho County: Floyd Meadow, 6000 feet, Idaho National Forest, August 8, 1930, C. Gray CG57 (USFH, 67318). Lemhi County: Allen Lake Meadow, 8000 feet, Salmon National Forest, September, 1930, A. H. Wheeler 70 (USFH, 84421). Valley County: Nameless Meadow, 6500 feet, Fayette National Forest, September 1, 1930, L. N. Wellman 18 (USFH, 67628) mixed with G. affinis. Blaine County: alpine slopes, base Devil’s Bedstead, Sawtooth Range, 8000 feet, July 28, 1936, J. W. Thompson 13549. Custer County: damp, springy soil, Toxaway Lake, 10 miles west-southwest of Obsidian, Sawtooth Mountains, 8500 feet, August 8-11, 1937, C. L. Hitchcock & J. S. Martin 5749; Mount Hyndman, August 11, 1939, Ray J. Davis 1704. Nevapa. Elko County: meadow, Ruby Ranger Station, 6000 feet, Humboldt National Forest, June 28, 1930, L. E. Mc- Kenzie 28 (USFH, 64353). Uran. Summit County: meadows, Lily Lake, 10,500 feet, 3 miles west of Bald Mountain, Uinta Mountains, August 14, 1983, Maguire et al. 4224; bogs, west shore Henry’s Fork Lake, 10,850 feet, Uinta Mountains, August 4, 1936, Maguire et al. 14381. Uintah County: wet place under spruce, south base of Liedy Peak, 10,000 feet, Uinta Mountains, August 21, 1939, Maguire 17678. Specimens from Oregon intermediate to subsp. typica are: Baker County: Antony Lakes Region, Blue Mountains, 7100 feet, July 28, 1936, J. W. Thompson 13430. Wallowa County: meadows, Mirror Lake, Eagle Cap Peak, Wallowa Mountains, 7500 feet, September 24, 1938, C. W. Shar- smith 3971. The subspecies of Gentiana calycosa may be distinguished as follows: Subspecies asepala Stems conspicuously decumbent, 5-10 (12) cm. long. Leaves 1.0-1.5(1.7) cm. long. Corollas 3—4 cm. long. Calyx membranaceous, deeply incised, lobes obsolete, or inconspicuous and mostly subulate, 1.0-1.5(2.0) cm. long. A Great Basin-Intermountain race, apparently confined to Utah, Nevada — and Idaho. Intermediate plants are known from eastern Oregon. Subspecies typica Stems erect, or slightly decumbent at the base, (8)10—-20 em. long. Leaves (1.5)2.0-3.0 cm. long Corollas 3.5-4.5(5.0) em. long. Calyx membranaceous mainly in the sinuses, not at all or rarely incised, lobes foliaceous, ovate, elliptic, or lanceolate, 5-10(15) mm. long. The population of the Rocky Moun- tains and the Sierra Nevada, ex- tending into Alberta and British Columbia. 1942] MAGUIRE: ARNICA IN ALASKA AND YUKON 153 The smaller, frequently almost procumbent subspecies is pri- marily and sharply set off from the typical population by the critical calyx distinction. It further occupies a clear-cut southern geographical range. Interestingly there is considerable similar- ity in its calyx characters to those of Gentiana Parryi of the south- ern Rocky Mountains, suggesting that this latter species might be likewise a southern, but more complete segregate of G. calycosa subsp. typica. Intermountain Herbarium, Utah State Agricultural College, Logan. May 11, 1941 ARNICA IN ALASKA AND YUKON Bassetrt MAGUIRE For some time the manuscript of a monograph of the genus Arnica has been completed, and is awaiting publication. During the course of the work the writer had been asked by Dr. Eric Hul- tén to contribute the account of Arnica for the “Flora of Alaska and Yukon,” now appearing in parts. Because of the immediacy of the needs of Dr. Hultén, and because appearance of the “mono- graph” does not, in the near future, seem probable, the following new entities, new names, and new combinations, for the most part, are herewith extracted from that study. Only the principal synonomy is given for the new names and combinations. Se- quences and numbering of entities is that employed in the treat- ment prepared for Hultén’s “Flora.” la. Arnica aupina (L.) Olin subsp. angustifolia (Vahl) comb. nov. A. angustifolia Vahl, Fl. Dan. 3. 1816; A. alpina (L.) Olin var. angustifolia Fernald, Rhodora 36:96. 1984. Ib. Arnica atpina (L.) Olin subsp. attenuata (Greene) comb. nov. A. attenuata Greene, Pittonia 4: 170. 1900. Ie. Arnica aLtpina (L.) Olin subsp. tomentosa (Macoun) comb. nov. A. tomentosa Macoun, Ottawa Nat. 13: 168. 1899; A. tomentosa Greene, Pittonia 4: 168. 1900; A. pulchella Fernald, Rhodora 27: 18. 1915. 2a. ARNICA LOUISEANA Farr subsp. frigida (Meyer) comb. nov. A. frigida Meyer ex Ilijin, Trav. Musc. Bot. Acad. Sc. U.S.S.R. 19: 112. 1926; A. nutans Rydb. N. Am. Fl. 34: 828. 1927; A. Sancti- Laurenti Rydb., N. Am. Fl. 84: 828. 1927. a. Var. genuina nom. nov. d. frigida Meyer ex Ilijin l.c. b. Var. Mendenhallii (Rydb.) comb. nov. A. Mendenhalli Rydb. N. Am. Fl. 34: 329. 1927. ce. Var. brevifolia (Rydb.) comb. nov. A. brevifolia Rydb. Nevam. Fl. 84: 329. 1927. d. Var. illiamnae (Rydb.) comb. nov. A. Illiamnae Rydb. Noam. Fl 34-331. 1927. 154 MADRONO [Vol. 6 e. Var. pilosa var. nov. Caulibus 5-10 (15) em. altis, foliis 2.0-4.0 (5.0) cm. longis, denticulatis; herbaceo exigue piloso; pedunculis dense pilosis; pericliniis densus subflavis lanato-pi- losis; capitulis nutantibus, aut plus frequenter erectis; acheniis glabratis, aut plus frequenter sparse super hirsutis. Type. Igloo Creek, McKinley National Park, July 11, 1932, | Joseph Dixon 29. Deposited in the Herbarium of the University of California; isotype at the United States National Herbarium. This population, seemingly confined to the Mount McKinley area, is strongly marked. Plants with erect heads closely re- semble specimens of A. alpina subsp. tomentosa, to which indeed they had mostly been referred. The basally or entirely glabrous achenes, and the yellowish periclinial pubescence seem, however, definitely to relate the McKinley plants to d. louiseana. 4a, ARNIcA corpiFoLIA Hook. subsp. genuina nom. nov. A. cor- difolia Hook., Fl. Bor. Am. 1:33. 18384. a. Var. pumila (Rydb.) comb. nov. A. pumila Rydb. Mem. N. Y. Bot. Gard. 1: 433. 1900. 5a. Arnica Cuamissonis Less. subsp. genuina nom. nov. A. Chamissonis Less., Linnaea 6: 238. 1881. a. Var. Typica Regel, Suppl. Ind. Sachal 151. 1864. b. Var. interior var. nov. Foliis inferioribus plus petiolatis, 2.5-3.5 cm. latis; capitulis 15-18 mm. latis; corollis disci 7-8 (10) mm. longis. Type. Palliser, July 30, 1906, S. Brown 770. Type deposited in the Gray Herbarium; isotypes at United States National Her- barium, New York Botanical Garden, Academy of Natural Sci- ences, Philadelphia, Missouri Botanical Garden. The variety interior is the common inland form, and by far the greatest population, extending southward into Canada. It differs from, although completely intergrading with, the maritime var. typica Regel in having mostly petiolate rather than sessile lower cauline leaves, smaller heads (18-20 cm. broad in var. typica), and shorter disc flowers (9—11 mm. long in var. typica). 6a. ARNICA AMPLEXIFOLIA Rydb. subsp. genuina nom. nov. 4. amplexifolia Rydb. Mem. N. Y. Bot. Gard. 1: 4384. 1900; A. am- plexicaulis Nutt. Trans. Am. Phil. Soc. ns. 7: 408. 1841, non Wall. 1837. a. Var. borealis (Rydb.) comb. nov. A. borealis Rydb. NvAme Flo34s 35d 927. 6b. Arnica aMpLExiFoLiaA Rvdb. subsp. prima subsp. nov. Rhi- zomate longo, 2-8 mm. diametro, sparse radicellis vestito; caule 4.0—7.5 dm. alto, simplici, gracili, non furcato; foliis caulinis 5—7 jugis, elliptico-lanceolatis remote serrato-dentatis, 2-3 cm. latis, 6-8 cm. longis, capitulis 1-3, hemispherico-campanulatis, 12—15 mm. altis, pericliniis moderatis pilosis, longi stipitato-glandulosis. 1942] MAGUIRE: ARNICA IN ALASKA AND YUKON 155 Type. Kodiak, August 28, 1867, A. Kellogg 231. Deposited at the Academy of Natural Sciences, Philadelphia; isotypes at Gray Herbarium, United States National Herbarium, Missouri Botanical Garden, and a specimen of questionable authenticity at the New York Botanical Garden. The subspecies occurs only in maritime Alaska and on the Alaskan islands. This subspecies may be separated from subsp. genuina by the following key: Cauline leaves 5—7 pairs, the lower 2 or 3 pairs petioled, mar- gins only inconspicuously dentate-serrate; mostly with 1-3 heads; plants of the Alaskan Coast region and islands ... A. amplexifolia subsp. prima Cauline leaves 5-12 pairs, mostly all sessile, margins mostly conspicuously sharply serrate-dentate; mostly with 5(3)— 7(9) heads; occurring from Alaska to California and AVRORNG AINA Meet ee eet rae eee eaten ee eee cies Solas wea ee hy A.amplexifolia subsp. genuina This clearly distinct race is interesting and important. It shows much similarity to Arnica Chamissonis and A. ampleaifolia subsp. genuina, in leaf character and pubescence, and particularly resembles the former in the tendency to produce long unap- pendaged rhizomes. Indeed the intermediate characters of this rather small group of plants strongly suggest the possible deriva- tion of the remainder of the subgenus Chamissonis through A. Chamissonis. Or, again, this entity of primarily insular distribu- tion might prove to be hybrid between A. amplewxifolia and A. Chamissonis. 8a. Arnica Lesstneu Greene subsp. genuina nom. nov. A. Les- singit Greene, Pittonia 4: 167. 1900. 8b. Arnica Lesstnam Greene subsp. Norbergii Hultén and Ma- guire. Caulibus 2.5-3.5 dm. altis, foliis caulinis 5-6 jugis, (3) 5-8 cm. longis, 1-1.5 em. latis, ellipticis, acutis, herbacio sparse aut moderate moniliformo-pilosis. Type. Orca, August 14, 1937, I. L. Norberg. Deposited at Lund; isotype at Utah State College. So striking and distinctive are the type specimens in height and in the numerous pairs of narrower cauline leaves (3—4 pairs in subsp. genuina) that the authors venture to designate them as a subspecies. Were it not for approaching intermediates (Hultén 8144 from Juneau) they should be tempted to propose these plants as representing a separate species. A somewhat similar plant, Anderson 2A 375 from Mendenhall, is questionably referred here. It may be found that the Norberg specimens represent sole dis- parities and hence are best relegated to varietal rank. Intermountain Herbarium, Utah State Agricultural College, Logan, Utah. 156 MADRONO [Vol. 6 A COMPARISON OF THE EMBRYOGENY OF PICEA AND ABIES J. T. BucHHOLZ Picea SmituiAaNa Boiss. The embryogeny of Picea Smithiana Boiss. was observed in the spring of 1986 in trees planted on the campus of Stanford University. Though it was impossible to fol- low the entire embryogeny, the stages of suspensor formation were observed in considerable detail and this account will serve to present more fully than heretofore the differences between the suspensor of a spruce and that of a fir, which will also be described in greater detail. Picea Smithiana is a species introduced into cultivation more than a century ago from southern Asia. Its native region is given as Afganistan and the Himalayas. Its embryogeny appears to be very similar to that of Picea Abies (L.) Karst., (P. excelsa), and several American spruces which the writer has examined pre- viously and is probably typical for the genus Picea. The development of the proembryo was not observed. Miyake (7) gave us a description of the proembryo of Picea excelsa in all stages. Another record in the form of drawings made nearly a century ago by Schacht (8) describing the late stages in the pro- embryo of Picea glauca (Moench) Voss has been generally over- looked because it was given under the name Abies alba Michx., a synonym of Picea glauca. Four tiers of four cells each are found in the proembryo. Pinus and nearly all genera of Pinaceae (Abietineae) have this same type of proembryo (2, 4). The uppermost tier of cells is incompletely surrounded by walls, and these four nuclei soon dis- appear. The uppermost tier of walled cells is the rosette, the name given to this structure by Schacht. Below this is a tier of suspensor cells which elongate in unison, and push the embryonic tier forward into the female gametophyte. The embryonic tier divides very soon after the elongation of the suspensor has begun. This precocious division of the embryonic cell to form an addi- tional tier is a feature wherein the spruce differs from the fir. In Abies the tier of embryonic cells may remain undivided for a con- siderable period, during which the suspensor becomes very long. Figure 1 shows the earliest stage of the embryo of P. Smithiana obtained from dissections. The division of the embryonic cells into two tiers has already taken place, although the cells of the suspensor tier have only become elongated in the ratio of about 6:1. Figure 2 shows a later stage in which the four lowest cells have divided again, so that three cells are now replacing each of EXPLANATION OF THE Ficures. Puate 15. Puate 15. Successive STaces In EMBRYOGENY OF Picea SMITHIANA. Figs. 1-6, X 60 (fig. 2 an optical section). s=primary suspensor, e=tier of embryonic cells; €:, €2, €s=successive tiers of embryonal tubes that constitute secondary additions to the suspensor system. 1942] BUCHHOLZ: EMBRYOGENY OF PICEA AND ABIES 157 Puate 15. Successtve STacGes IN EMBRYOGENY OF PICEA SMITHIANA. 158 MADRONO [Vol.6 the four original embryonic cells as found in the lowest tier of the proembryo. In figure 3, the tier of cells next to the suspensor is beginning to elongate. This elongation is succeeded by that of the next tier, and by that of adjacent cells in tiers still to be added by the division of the terminal tier. Figure 4 shows a sus- pensor system in which the s, e; and e, tiers have elongated. For convenience the first tier is called the primary suspensor tier and the series of cells that elongate in successive 4-celled tiers may be designated as embryonal tubes e;, es, e3, and ey. The e, is the last tier to be added to this regular succession before the terminal group of embryonic cells becomes more actively meristematic to increase greatly the number of cells in the entire embryo. The tiers that elongate subsequently to add the secondary suspensor come off in more irregular groups of cells. Figure 5 shows this stage in an embryo which has produced the e, sections of the sus- pensor. It is likely that the e; tier may at times be the last regular tier and in subsequent stages, the embryonal tubes lose their sym- metry, become interlocked and more irregular in their elongation as they build up a massive secondary suspensor. The writer can find no essential difference in appearance be- tween the tiers s, €1, €2, €3, etc., in Picea. There is probably no valid reason for making the distinctions implied by the names primary suspensor s and embryonal tubes of the successive orders €1, €2, ete., but this somewhat arbitrary designation is retained for the sake of uniformity and because it does seem to make a differ- ence when compared with other genera of conifers. Figure 6 shows an embryo of more than 80 cells in which about 8 cells are adjacent to e, tier, which will elongate to form a much more mas- sive addition to the suspensor. The early suspensor system is, therefore, segmented and made up of four or five 4-celled and regularly tiered tubes. Picea has a rosette in which nuclei remain visible for some time, but the nuclei in the rosette disappear without undergoing division about the time the nuclei of the primary suspensor cells disintegrate. The successive suspensor cells collapse in the order of their age so that their remains and those of the rosette are sometimes difh- cult to find. The cells in the successive suspensor segments be- come about 20 to 80 times as long as their original diameter before the nuclei disappear and the suspensor elements collapse. In Picea, simple polyembryony prevails. The fertilization of several eggs may give rise to several embryos which compete with each other, but the product of a single fertilized egg does not split up into several embryos, as in Pinus, Cedrus and Tsuga. The single embryo which survives usually has 8 to 10 cotyledons. Picea Omorika (Pancic) Purkyne. The Siberian spruce was examined a number of years ago from trees planted at the Arnold Arboretum. The stages obtained were somewhat older than those described above, but the succession of tiered suspensor elements 1942] BUCHHOLZ: EMBRYOGENY OF PICEA AND ABIES 159 could also be observed in this species. Since the oldest elements had collapsed, it was not possible to be certain that e, was the last 4-celled tier that had elongated, though an e; tier could be recog- nized. In respect to the suspensor system and simple poly- embryony, this Siberian spruce agrees with other spruces. The embryo as found in the mature seed of Picea Omorika had 5 to 6 cotyledons. AsBiEs veNnusTa (Dougl.) K. Koch. The material for a study of the embryogeny of the Santa Lucia fir came from trees planted on the campus of Stanford University. The trees in the Santa Lucia mountains, their native region, were found to be too inaccessible and too far removed from a laboratory. Those on the grounds of Stanford University could be studied within an hour after collec- tion. Their advantages were the early fertilization and the fact that the schedule of development fitted more conveniently into a research program which included other conifers. The proembryo of no species of Abies has been adequately investigated. Miyake (6) observed only the free nuclear stages and Hutchinson (5) added a few stages in a study of Abies bal- sama. Hutchinson’s account, especially his statement: “the pro- embryo of Abies ordinarily consists of eight cells only, in two tiers’ is misleading and might leave one with the impression that Abies has a program in its proembryo which differs from other conifers. Actually it is very similar to Pinus (4) and Picea (7) in its early stages. There is no doubt that the proembryo is the same as that described above for Picea except that the embryonic tier does not divide immediately. This tier divides only after considerable elongation of the suspensor tier. The three tiers of cells—rosette, suspensor, and embryonic— are still present and undivided in A. venusta when the suspensor has elongated considerably. This is shown in figure 7 and in fig- ure 8. Thus it is clear that there is really no serious hiatus in our knowledge concerning the proembryo of Abies. An opaque de- ‘posit has appeared over the rosette and all traces of the tier of relict nuclei have disappeared. In figure 9 each cell of the lowest tier of embryonic cells has divided so that this quartette of cells has now formed an octette arranged in two tiers. In figure 10 each of the original quartette of cells is represented by a progeny of four cells. The details of the embryo in figure 10 are shown better under higher magnification in two planes of focus in figures 10a and 10b. The pointed tip of the embryo has a very thick hyaline wall which is also shown best in figures 10a and 10b. The pointed ends are very close together at first (figs. 7-9), become separated with the growth and lateral enlargement of the embryo, and remain as tiny projecting knobs that serve to identify the four derivative cell groups. These hyaline tips may be seen in successive stages 160 MADRONO [Vol.6 shown by figures 11, 16, 17, and are obscurely recognizable in figure 18, in which the embryo has several hundred cells. Sometimes the vertical rows do not contribute equally to the embryonic mass. Figures 12 and 13 show conditions in which inequalities have appeared in the development of the progenies of the 4 original embryonic cells. These do not show the small knobs that are seen in figure 17. In the latter, and in figure 18, the evidence is fairly conclusive that the cell progenies of the four original embryonic cells have contributed equally to the embry- onic tissue. Simple polyembryony prevails in A. venusta and is the normal condition in all firs that have thus far been examined. The prod- uct of each fertilized egg gives rise to a single embryo below the suspensor, and whether the four terminal embryonic cells con- tribute equally, as indicated in figures 17 and 18, or unequally, as indicated in figures 12 and 13, appears to make no distinguish- able difference in later stages. The suspensor s, composed of four cells, is remarkable in the amount of its elongation. Each cell elongates to about 60 times its original diameter before the terminal embryonic cells undergo their first division, and eventually becomes 150 to 200 times this length before the secondary suspensor begins to elongate, and before the primary suspensor cells collapse. Figure 16 has a suspensor in which each cell is fully 200 times as long as the width of these cells in figures 7 and 8. Figure 17 has suspensor cells that are not quite as long, but the ratio is fully 160: 1. The latter has probably ceased elongation for the nuclei are no longer visible. The secondary suspensor is formed by the elongation of cells (usually spoken of as embryonal tubes), from the distal end of the embryo, as shown in successive stages by figures 12, 13 (pl. 16) and 16-20 (pl.17). As the enlarging embryo becomes wider, the number of elongating cells increases (figs. 18, 18), so that the secondary suspensor now becomes increasing'y more massive. The suspensor system may become greatly coiled and twisted. In dissections, some of these twists were removed, and those that are shown in the figures are not necessarily the ones that were most extreme before dissection. ‘The newest additions of the secon- dary suspensor elongate most rapidly. These projecting newly- EXPLANATION OF THE Ficures. PuLate 16. Piatt 16. STaAGes In EMBRYOGENY oF ABIES vENUSTA. Figs. 7-8, earliest stages dissected show 12 cells of 16-celled proembryo still undivided. Fig. 9, lowest (embryonic) tier of cells has divided. Fig. 10, later stage after four embryonic cells have each become 4-celled, with details of embryo shown in sur- face view in 10a and in lower focus in 10b (X147). Fig. 11, later stage with 8 cells in each of the four embryonic cells. Fig. 12, formation of secondary sus- pensor by elongation of cells from distal end of embryo. Fig. 13, later stage of same, showing rosette embryos derived from rosette cells, and well developed secondary suspensor with the suspensor cells now irregularly interlocked. Figs. 14-15, rosette embryos from embryo systems of stage similar to Figs. 12-13. All figures, except 10a and 10b, x 60. r=tier of rosette cells; s=primary sus- pensor; ¢=tier of embryonic cells. 1942] BUCHHOLZ: EMBRYOGENY OF PICEA AND ABIES 161 Puate 16. STAaGes IN EMBRYOGENY OF ABIES VENUSTA. 162 MADRONO [ Vol. 6 formed embryonal tubes usually ensnare any shorter embryos from another zygote that may be situated behind them and push the shorter rivals back toward the region of the archegonia. Abies venusta usually forms rosette cells. A majority of these form rosette embryos, as may be seen in the upper part of figure 13 and in figures 14 and 15. These secondary embryos are formed only after considerable delay, during which the primary suspensor becomes fully elongated. In early stages shown by figures 7 to 10, and even in figure 16, the rosette cells usually re- main undivided. In figure 17, each of the two rosette cells has divided once. Figures 14 and 15 were taken from embryos that had not only fully-elongated and collapsed primary suspensors, but the embryos had become massive and had developed secon- dary suspensors similar to those shown in figures 18 and 19. It was also observed that sometimes these rosette cells may fail to divide; the nuclei may disintegrate and the cells collapse without forming embryos. The later stages of the successful embryos are shown in figures 19 and 20. Abies venusta agrees with other members of this fam- ily of conifers in that the primordium of the stem tip appears as a slight protuberance before the cotyledonary primordia are formed. The embryo shown in figure 20 had 7 cotyledons surrounding and obscuring the primordium of the stem tip. The cotyledon number varied between 5 and 8. In a count of about 150 em- bryos, the mean number of cotyledons was 6.5. The embryo of figure 20 has a conspicuous calyptroperiblem. This structure, which replaces the root cap in conifers, merges with the secondary suspensor and surrounds the plerome of the root tip. A ridge surrounding the embryo of figure 20 just below the middle marks the outer margin of the calyptroperiblem. In respect to this feature, the embryo of A. venusta resembles that of Cedrus (3) more than that of Picea. Asies Prnsapo Boiss. Embryological material of the Spanish fir was obtained from reproductive specimens found on the Mills Estate at Millbrae and on the Flood estate near Menlo Park, both in California. Only the early stages in the embryogeny were satisfactory. In stages later than those shown in the accompany- ing figures (21-24), the embryos were all aborted and the entire gametophytes shriveled up so that no viable seeds were matured. In the early stages there was a remarkable similarity with the embryogeny of A. venusta. EXPLANATION OF THE Ficures. PLate 17. Pruate 17. Laver StTaces 1x Empryoceny or ABIES vENUSTA. Figs. 16-17, embryo systems showing extreme length of suspensor in a stage subsequent to Fig. 10, X60. Fig. 18, embryo on well-developed secondary suspensor (stage following Figs. 11-13), x 60. Fig. 19, stage showing primordium of stem tip, x20. Fig. 20, embryo after cotyledonary primordia have formed, X20. r= rosette cells; s= primary suspensor; e:= beginning of secondary suspensor. 1942] BUCHHOLZ: EMBRYOGENY OF PICEA AND ABIES 163 Puate 17. Larer STAGEs In EMBRYOGENY OF ABIES VENUSTA. 164 MADRONO - [Vol.6 Discussion In the writer's earliest investigations on polyembryony (1), where the similarity in Abies and Picea with respect to simple poly- embryony was stressed, the differences in the suspensor systems of these two genera was entirely overlooked. Unfortunately, these diagrams (copied by several authors) have been taken too literally. Much greater accuracy was achieved in a later publica- tion (2), but in the second set of drawings, the rosette cells had usually collapsed in the embryos with long suspensors—those that were selected for the illustrations. In re-examining these preparations, it is obvious that the single tier of primary suspensor cells of Abies balsamea collapses earlier and does not become as extreme in its elongation as the same structure in A. venusta. The difference between the spruce and fir is apparent in these old preparations and in the drawings made in 1980, which are fairly accurate, but the rosette cells and the older parts of the suspensor elements are so shriveled that they are very difficult to recognize. Another feature which makes the rosettes and suspensor struc- tures very difficult to study is not only the deposit shown above the rosette in nearly all of the figures, but a similar gummy or resinous deposit which surrounds the upper region of the suspensor system, omitted in all of the drawings. There were several objectives that motivated this investiga- tion. It was suspected that the embryogeny might offer features which would serve to separate A. venusta from the remaining firs and. place it, either in one of the other genera or in a new genus. It was also desirable to discover the basis for a clearer definition of the differences between the genera Picea and Abies, both of which have simple polyembryony. The taxonomic differences between Abies venusta and other firs are very marked. There are differences in the leaves, winter buds, bracts of the cone scale, and in the fact that the branches of this species are somewhat flexible and do not hold the cones erect asin other firs. The seed cones are inserted at nearly right angles on the twigs that bear them, are erect in early stages, but later their weight bends the slender twigs downward, so that many cones appear to be borne in horizontal positions, sometimes even drooping downward. The external differences may not be of great importance, but some of these characters such as the large peculiar winter buds and the very long, pointed leaves, so different from other firs, may have caused botanists to wonder whether this species may not represent something other than a true fir. In fact, several botanists, in discussing the unique external features in the Santa Lucia fir expressed the wish that the embryogeny be investigated in the hope that this might contribute new facts that might possibly prove to be decisive. Abies Pinsapo offered many external taxonomic features of contrast with A. venusta. How- 1942] BUCHHOLZ: EMBRYOGENY OF PICEA AND ABIES 165 Puate 18. Successive STAGES IN EMBRYOGENY OF ABies PinsApo. The paired embryo systems, Figs. 23-24, were dissected from the same ovules, X 60. $=primary suspensor; e=tier of embryonic cells. 166 MADRONO [Vol. € ever, the embryogenies of these two species are in close agree- ment in the early stages that were observed. Picea and Abies have a similar type of proembryo and both have simple polyembryony. Even though the proembryo of Abies has not been fully investigated, the three tiers of cells that remain are still undivided in the earliest stage shown in figure 7. The two genera differ, however, in the manner in which the suspensor sys- tem develops. In Picea, the primary suspensor tier (s) is fol- lowed by a succession of three or four similar additions, e1, eo, es, and e, before the secondary suspensor becomes irregular and may consist of many cells; in Abies a very long primary suspensor is formed which is directly succeeded by a more irregular many- celled secondary suspensor. In Picea, rosette cells are formed but they usually abort and have not been observed to develop into embryos. In Abies balsamea and A. Pinsapo, these rosette cells usually fail to develop as in Picea, but in A. venusta a considerable number of instances were observed in which rosette cells developed into very small embryos. Abies venusta agrees on the whole with the embryogeny of other species in this genus. It differs not only from Picea but also from Pseudotsuga (1, 2), in which no rosette cells are formed. Abies venusta differs from Pinus, Cedrus and T'suga, all of which have cleavage polyembryony. There is no embryological feature in A. venusta which would tend to segregate it from the genus Abies save the rosette embryos which develop in a little more than half of the embryos. Other embryological characters that are distinc- tive in this species are quantitative in nature. The embryology of A. venusta, therefore, offers little in support of the idea that this species should belong to another genus. It definitely does not belong to any of the other genera now recognized and does not show embryological features which are sufficiently unique to sug- gest its segregation into a new genus. SUMMARY 1. The early embryogeny of Picea Smithiana is described with special reference to the development of the suspensor system. 2. The embryogeny of Abies venusta is described and com- pared with other firs, including A. Pinsapo, and with Picea. 3. Both genera begin their development with the same type of proembryo. In Picea the embryonic cells divide early and form three or four additional tiers of cells that elongate successively in adding relatively short tiers of elongated sections to the suspensor system. In Abies the embryonic cells do not divide until the suspensor tier has become well elongated. The suspensor cells elongate as a single tier, eventually becoming as long as the 3 to 4 tiers of additions in Picea, or even longer. 4. Rosette cells are formed in both genera. In Picea and some species of Abies the rosette cells do not divide to form em- 1942] PURER: AMMOPHILA 167 bryos; in A. venusta the rosette cells may form embryos after the suspensor tier has become fully elongated. 5. On the whole, the embryogeny of Abies venusta agrees closely with other firs. It does not belong in one of the other genera of Pinaceae and there seems to be little in the embryogeny which would suggest the segregation of A. venusta into a distinct genus. Department of Botany, University of Illinois, Urbana, July, 1941. LITERATURE CITED 1. Bucunoiz, J.T. Polyembryony among Abietineae. Bot. Gaz. 169: 153-167. 1920. 2, ——_—_—————._ The pine embryo and the embryos of related genera. ‘Trans. Ill. State Acad. Sci. 23: 117-125. 1931. 3. Bucunoriz, J. T. and Epona M. Op. The anatomy of the embryo of Cedrus in the dormant stage. Am. Jour. Bot. 20: 35-44. 1933. 4. Frercuson, Marcaret C. Contributions to the life history of Pinus, with special reference to sporogenesis, the development of the gametophytes and fertilization. Proc. Wash. Acad. Sci. 6: 1-202. 1904. 5. Hutcuinson, A. H. Embryogeny of Abies. Bot. Gaz. 77: 280-289, 1924. 6. Miyake, K. Contribution to the fertilization and embryogeny of Abies bal- samea. Beih. Bot. Centralbl. 14: 134-144. 1903. 7, ————————... On the development of the sexual organs and fertilization in Picea excelsa. Ann. Bot. 17: 351-372. 1903. 8. Scuacut, H. Entwickelungsgeschichte des Pflanzenembryon. Amsterdam. 1850. ANATOMY AND ECOLOGY OF AMMOPHILA ARENARIA LINK Epitu A. PuRER During a period of three years, studies were made upon sand dune plants along the Pacific coast from Oregon to Baja Cali- fornia. Environmental factors were measured and a study of the anatomy of twelve species was undertaken. In the subsequent publication (Purer, Studies of Certain Coastal Sand Dune Plants of Southern California, Ecol. Monog. 6: 1-88, 1936) the genus Ammophila was omitted since it was not abundant in the particular areas where the instrumental work was carried on. Although it is not a native species, Ammophila arenaria Link has been successfully planted as a sand binder in a number of areas and is fairly well distributed along the Pacific coast. Speci- mens were examined at the herbaria in the following institutions, Stanford University (D) ; University of California (UC) ; and the University of Southern California (USC): Linnton, Oregon, Sep- tember, 1927, Thompson 3881 (D); Eureka Peninsula, Humboldt County, June, 1899, Dudley (D); Point Arena, Mendocino County, August, 1899, Davy and Blasdale 6046 (UC) ; one-half mile south of Lake Merced, San Francisco County, May, 1901, Dudley (D) ; Jazos Creek, San Mateo County, March, 1922, Bacigalupi (D) ; 168 MADRONO [ Vol. 6 Oceano, San Luis Obispo County, October, 1930, de Forest (USC) ; Hueneme, Ventura County, August, 1931, Purer 2213; strand at Spanish Bight, San Diego County, October, 1931, Purer 2311. Ammophila arenaria thrives in open places where there is little or no vegetation, where the wind is severe, the insolation great, and the soil unstable. As a rule this grass, commonly known as sand reed, grows alone on the dunes, although occasionally plants such as Franseria bipinnatifida may establish themselves on sand held by Ammophila. Concerning the stabilization of moving dune areas, A. S. Hitchcock (Controlling Sand Dunes in the United States and Europe, Nat. Geog. Mag. 15: 46, 1904) states: “Many plants have been tried, but the most satisfactory is beach grass (Ammophila arenaria Link). This grass grows naturally upon the sand dunes of the North Atlantic coast of Europe as far south as Morocco, and of America as far south as North Carolina, and also along our Great Lakes. This is the grass which was used in re- claiming land which is now Golden Gate Park in San Francisco.” The highway department of Oregon has also planted this species as a sand binder on dune areas of the coast. The sand reed grows in large clumps to a height of two to three feet, the stems and the roots being produced at the nodes of elongated rhizomes. From each node of the rhizomes five or six roots, about twenty-five centimeters long and usually bearing many secondary roots two to three centimeters in length, are pro- duced. The plant withstands partial covering by sand through the production of adventitious roots from the buried portions of its erect stems. The entire root system is smaller in proportion to the aerial parts than in other principal sand dune plants. The leaf blades which are long and gradually narrowed to a point, bend in all planes without breaking. The blades are mostly ver- tically placed, receiving the minimum amount of light; only old leaves are bent at an angle. Since the plant grows in tufts its leaf blades shade each other somewhat. In San Luis Obispo County the young leaves were found tightly rolled and the old leaves partially unrolled. Farther south, where the habitat is even more xeric, the leaves were always found tightly rolled; in northern California and Oregon the leaves were habitually par- tially unrolled. According to E. Pee-Laby (Etude anatomique de la feuille des graminées de la France. Ann. Sci. Nat. Bot. 8: 227- 346, 1898) such variation is due to the relative dryness of the habitat. EXPLANATION OF THE Ficures. PLate 19. Piate 19. AM™MoPrHILA ARENARIA. A, stem, transverse section; B, vascu- lar bundle, transverse section; C, folded blade, diagrammatic transverse section; D, blade, transverse section; EH, infolded adaxial epidermis of blade showing motor cells; f, epidermal cells and stomata on adaxial surface of blade; G, epi- dermal layer of blade showing stoma, transverse section; H, sheath, transverse section; [, epidermal layer of sheath showing stoma, transverse section; J, root, transverse section. 1942 | PURER: AMMOPHILA 169 Piate 19. AMMOPHILA ARENARIA LINK. 170 MADRONO [Vol. 6 The roots (pl. 19, fig. J) are short and fibrous, and when pulled from the soil easily lose their outer cortical layer. The epidermis is composed of small compactly-placed cells; inside of this is the cortex, composed of soft, closely placed parenchyma cells, which makes up about two-thirds of the root in cross-section, The endodermis is composed of very heavy-walled cells, the inner wall being especially thick while the pericycle is inconspicuous. The stele contains a few tracheae surrounded by sclerenchyma. Phloem is inconspicuous. The stem (pl. 19, figs. A, B) is of simple monocotyledonous type, compact, with few intercellular spaces, the small epidermal cells having all their walls of about equal thickness and bearing a thin layer of cuticle. Within this tissue are about five or six rows of cortical parenchyma cells and on the inner side of these a definite ring of sclerenchyma. The scattered vascular bundles are disposed in a relatively narrow zone occupying about two- fifths of the radius, the outer bundles being embedded in the ring of sclerenchyma. The central pith is definite and conspicuous, composed of large cells occupying about three-fifths of the diam- eter of the young stem; old stems are usually hollow. The outside epidermis of the sheath (pl. 19, figs. H, I) is cov- ered with thick cuticle, and bears at frequent intervals, short, uni- cellular, unbranched, pointed trichomes. There is a single row of closed vascular bundles with conspicuous sheaths. Scleren- chyma surrounds these, and expands to a broad, wedge-shaped mass, causing a slight protrusion of the epidermis. In the tissue between the bundles are large lacunae running vertically through the sheath. On the inner surface in parallel rows stomata occa- sionally occur. The blade and sheath portions of the leaf differ considerably in structure. The cylindrically rolled blade (pl. 19, figs. C, D) of the southern California plants is covered on the abaxial surface by a firm epidermis of small cells averaging 0.017 millimeters in their radial diameter. These bear a cuticle about 0.008 milli- meters in thickness, or more than 47 per cent of the radial diame- ter of these cells. The outer surface of the epidermis is regularly undulate, which permits the blade to unroll without splitting the cuticle. There are no stomata in this surface.. Below the epi- dermis are several rows of large sclerenchyma cells with moder- ately thick walls. The adaxial surface of the blade, the inner face of the rolled blade, is ridged and grooved. The ridges, alternately large and small, are approximately 0.55 millimeters and 0.25 millimeters in height. The epidermis here averages 0.012 millimeters in thick- ness, and bears abundant unicellular, unbranched, conical trich- omes with thick walls, varying in length from four to ten times the diameter of the epidermis; there is a cuticle about 0.002 milli- meters in thickness, or about 17 per cent of the radial diameter 1942] McMINN: NEW SPECIES OF CEANOTHUS 171 of these cells. The stomata (pl. 19, figs. F, G) which are confined to the sinuses, occur in parallel rows and number about twelve per square millimeter; the stomatal pore is about 0.0014 millimeters in its long diameter, with the guard cells and the subsidiary cells measuring about 0.040 by 0.035 millimeters. The central portion of each ridge contains sclerenchyma, with a single closed fibro- vascular bundle near the base of the ridge. Surrounding the grooves and extending up the sides of the ridges almost to their tops is a narrow band of chlorenchyma, which consists of small, more or less isodiametric, parenchyma cells containing numerous chloroplasts. The top of each ridge is completely filled with sclerenchyma to the point where the chlorenchyma begins. Groups of motor cells (pl. 19, fig. E) are found at the bottom of each of the grooves; the intercellular spaces are small and few. The ridged and grooved surface of the involute blade results in a withdrawal of the chlorenchyma from the light. Transpiration in the leaf is checked by inrolling, by the heavy cutinization of the abaxial surface, and by the infrequent stomata which occur only in the grooves of the inrolled epidermis where they are overlapped by the trichomes. San Diego, California, February 4, 1941. AN UNDESCRIBED SPECIES OF CEANOTHUS FROM CALIFORNIA Howarp EK. McMinn Ceanothus Masonii sp. nov. C. rigidus variation 1 McMinn, Contrib. Dudley Herb. 1: 145. 1980, in part. C. gloriosus var. exaltatus J. T. Howell, Leafl. West. Bot. 2: 44. 1987, in part. Bolinas Ceanothus. Frutex erectus vel erecto-patens, 6-18 dm. altus, ramis crassis arcuato-divaricatis, ramulis rigidis atro-fuscis vel purpureis, to- mentulosis demum glabrescentibus; folia opposita persistentia, laminis late ellipticis vel fere orbicularibus, 6-19 mm. longis, 5—12 mm. latis, basi rotundis apice rotundis truncatisve, aliquando emarginatis, supra atroviridibus nitidis glabris, subtus albidis sub microscopio inter venas canescentibus, crebre dentatis dentibus brevibus aut rare leviter sinuato-dentatis ad basim versus integris ; stipulae prominentes persistentes, 1.6-5 mm. longae; gemmae squamae fuscae glabrae vel leviter tomentulosae; inflorescentia subumbellata conglomerata, plerumque foliis binatis parvis sub- tentia, ramos breves (6-19 mm. longis) terminantia; flores atro- cyanei vel purpurei; fructus globosus, tricornutus, 5 mm. diame- _ tro, cornibus brevibus apicalibus subapicalibusve, sine crestis _ Intermediis instructus: Erect or erect-spreading shrub, 6—18 dm. tall, with stout rather stiff divaricate branches and rigid dark brown or purplish tomen- 172 MADRONO [Vol. 6 tulose branchlets, becoming glabrous in age; leaves opposite, ever- green; the blades broadly elliptical to oval or nearly orbicular, 6-19 mm. long, 5-12 mm. broad, rounded or sometimes cuneate at base, obtuse, rounded or truncate at apex, sometimes emargi- nate, dark green, glabrous and glossy above, grayish white and microscopically canescent between veins beneath,. margins with numerous short teeth or rarely slightly sinuate-dentate except near base; stipules prominent, persistent, 1.6-5 mm. long; bud scales brown, glabrous, or slightly tomentulose; flowers dark blue to purple, in many-flowered umbel-like clusters usually subtended by a pair of small leaves terminating short lateral branchlets 6-19 mm. long; fruit globose, about 5 mm. in diameter, with 3 short apical or subapical horns, without intermediate crests. Flowering period, March, April. Type. Along trail on east end of Bolinas Ridge, Marin County, California, April 23, 19383, McMinn 3044, deposited in the University of California Herbarium, Berkeley, no. 657,550. Other representative collections: McMinn 906, 5416, 5417; transplant series, McMinn 1574R, 15740, 1574Q; Eastwood & Howell 3838. Bolinas Ceanothus occurs on Bolinas Ridge, Marin County, California. It is very closely related to Ceanothus gloriosus var. exaltatus J. T. Howell. These two entities belong to the C. gloriosus-C. ramulosus-C. purpureus-C. divergens-C. confusus complex which occurs in the North Coast Ranges of California, in Marin, Sonoma, Napa and Mendocino counties. My first acquaintance with the entity was in February, 1923, when I collected seven small plants (tentatively referring them to C. rigidus var. grandi- folius Torr.) and transplanted them to the trial gardens at Mills College. On March 380, 1928, I revisited the area on Bolinas Ridge in company with Herbert L. Mason. We found vigorous mature plants associated with Ceanothus foliosus, Arctostaphylos sensitiva, A. virgata, Quercus Wislizenit var. frutescens, Sphacele calycina and Adenostoma fasciculatum, in an area of about two miles along the ridge. These plants occupied the drier habitats of the ridge crest. In the late summer of 1924, a fire burned over Bolinas Ridge and destroyed most of the plants. On December 20, 1925, I again collected along the ridge. Not a single old plant of Bolinas Ceanothus was found; all had been destroyed by the fire of 1924. However, seedlings were abundant along the trail throughout the area. Twenty-three seedlings, from 4 to 12 inches tall were taken up and transplanted to the trial gardens at Mills College. At this writing, just sixteen years later, all but one (1574R) of the trans- plants have died. This lone survivor is about 6 feet tall and has a spread of 18 by 18 feet. The trunk at the ground is about 8 inches in diameter. In October, 1941, Dr. Mason and I studied the species of Ceanothus occurring on the south slope of Mount Tamalpais and 1942] NOTES AND NEWS 173 along Bolinas Ridge. Ceanothus Masonii was the most abundant species along Bolinas Ridge in the area which had been burned in 1924. I do not know how many times the ridge has been burned over subsequently; however, since some of the plants appeared to be at least six or seven years old, no destructive fires have occurred since 1935. Many seedlings and young plants abound in and along the ridge trail which has been cleared from time to time for use as a fire road. In addition to many plants of typical Bolinas Ceanothus, a few plants with leaves simulating those of C. purpureus Jepson, and a few with leaves intermediate between the two, were observed. A few plants with the large leaves and habit of growth of C. gloriosus var. exaltatus and others with smaller leaves intermediate between those of C. ramulosus (Greene) McMinn and C. purpureus were found growing along the ridge. These facts supported by additional observations made upon certain Ceanothus entities occurring in the North Coast Ranges, lead to the conclusion that Bolinas Ceanothus is a mem- ber of a large complex, which may consist of several species occupying different geographical and probably ecological niches. Mills College, California, November 27, 1941. NOTES AND NEWS RANGE EXTENSIONS IN Species oF WEsTERN NorTH AMERICA. New localities have been reported recently for the following species: Boyxinia Jamesit Engelm. var. HEUCHERIFoRMIS (Rydb.) Rosen- dahl. Growing in crevices of limestone cliffs, altitude 8800 feet, Canadian zone, above White Pine Lake, northeast slopes of Mount Magog, Cache County, Utah, July 17, 1986, Maguire 14046. This species is known from Colorado, Idaho and Nevada, but hereto- fore has not been reported from Utah. SAXIFRAGA ERIOPHORA S, Watson. This rare plant, apparently known previously only from the type locality in the Santa Cata- lina Mountains, Arizona, has been collected as follows: common, moist ravine slopes along stream course in yellow pine and oak, altitude 8500 feet, Pine Crest, Pinaleno (Graham) Mountains, Graham County, Arizona, April 17, 1935, B. §& R. Maguire 10545; altitude 8000 feet, May 26, 1936, B. §& R. Maguire 12012, May 28, 1935, B. §& R. Maguire 12014.—Bassett Maguire, Intermountain Herbarium, Utah State Agricultural College, Logan. ERIODICTYON capiTatuM Eastwood. Previously known only from Pine Canyon on Burton Mesa, five miles north of Lompoc, this species was discovered in a canyon on the James J. Hollister ranch, approximately five miles northeast of Point Conception, Santa Barbara County, California. Here, on a west-facing slope, elevation 900 feet, at the head of the west fork of Barranca 174 MADRONO [Vol. 6 Honda, the shrub is associated with Ceanothus cuneatus, C. papil- losus, Phacelia ramosissima var. suffrutescens, Quercus agrifolia and Salvia mellifera (July 20, 1941, Tucker 342, University of Cali- fornia Herbarium).—Joun M. Tucker, Department of Botany, University of California, Berkeley. ORNAMENTAL SHRUBS AND Woopy VINEs oF THE PaciFic Coast. By Evelyn Graham and Howard E. McMinn; published October, 1941, under the auspices of Mills College, California, by the Gillick Press, Berkeley, California; price $3.00. In this volume of 259 pages the authors have presented the first general treatise on the shrubs and vines cultivated in California. The descrip- tions are brief and for the most part entirely non-technical, al- though as a convenience a comprehensive glossary of botanical terms is included. An introductory chapter presents a concise account of flowering plant organs and their functions for those who may not have had previous training in botany. Practical keys are supplied for all genera and species treated. The book is well illustrated with twenty-two plates from natural color photographs, also with 144 figures consisting of both photographs and line drawings. With this volume in hand any interested per- son should be able to name most of the commonly cultivated orna- mental shrubs and vines of the state -——E. Crum. WeeEps or Catirornia. By W. W. Robbins, Margaret K. Bellue, and Walter S. Ball; published, 1941, by the State Department. of Agriculture, Sacramento, California; price $2.00. This useful work, succeeding Smiley’s “Weeds of California,” long since out of print, is the result of years of cooperation in research and field practice between the Division of Botany, College of Agriculture, University of California, Davis, and the Division of Plant Indus- try, State Department of Agriculture. Descriptions, origins and distributions of 693 species of weedy plants are included; in addi- tion the control of weeds is discussed and weeds of special crops and soils are listed. ‘Technical terms in keys and descriptions are avoided as much as possible. The volume is well illustrated with line drawings, photographs and 24 colored plates from the work of Lena Scott Harris. Sixteen maps showing the approximate distribution of certain important species are included.—A. Carrer. Purant Hunters 1n THE ANvDES. By T. Harper Goodspeed; pub- lished October, 1941, by Farrar and Rinehart, Inc.; price $5.00. In this attractive volume are recounted the adventures of the members of the recent University of California Botanical Garden Expeditions to South America, especially to the Andean regions of Peru and Chile. Dr. Goodspeed, Director of the Botanical Garden and leader of the expeditions, has presented in a very readable manner material of both botanical and general interest. The book consists of 429 pages and is profusely illustrated with 1942] PROCEEDINGS OF THE CALIFORNIA BOTANICAL SOCIETY 175 photographs not only of the plant life and topography of the regions visited but also of certain social aspects of the South American scene.—E. Crum. Desert Witp Fiowers. By E. C. Jaeger; second edition, pub- lished October, 1941, Stanford University Press; price $3.50. In this second edition a sixteen page popular key has been added that seems adequate when used in combination with the excellent drawings. Necessary corrections in the text have been made. The completeness of this attractive guide to the plants commends it to all who enjoy the deserts of California and their adjacent borders.—Davw D. Keck. The fourteenth annual meeting of the Western Society of Naturalists was held at Stanford University, December 29 to 31, 1941. An important feature of the program was a symposium dealing with the genetic basis of evolution. The following papers were presented: “Genetic Evolutionary Processes in Crepis,’ Pro- fessor E. B. Babcock, University of California; “The Pattern of Relationships Revealed by Morphologic, Ecologic and Cytoge- netic Evidence,’ Dr. Jens Clausen, Carnegie Institution of Wash- ington Laboratory, Stanford University; “The Sterility Barrier,” Dr. R. B. Goldschmidt, University of California; ““Where Does Adaptation Come In?” Dr. F. B. Sumner, Scripps Institution of Oceanography, University of California. Dr. Herbert L. Mason, University of California, presided and led the subsequent dis- cussion. Mr. J. Francis Macbride and Dr. Francis Drouet, members of the Herbarium staff, Field Museum, Chicago, spent some time in California this fall. During part of October they collected algae, mainly in the desert regions of southern California. Mr. Macbride, however, spent nearly two months in Berkeley where he was studying the South American collections in the University of California Herbarium in connection with his work on the “Flora of Peru.” PROCEEDINGS OF THE CALIFORNIA BOTANICAL SOCIETY May 15,1941. Meeting, 2093 Life Sciences Building, Univer- sity of California, Berkeley, at 7:45 P.M. Dr. G. Ledyard Steb- bins, Jr., Chairman of the Program Committee, occupied the chair. Dr. Norman H. Boke, of the University of California Botanical Garden, Berkeley, gave a lecture on “The Structure of the Cactus Plant.” The talk was illustrated by prepared sections of cactus tissue, shown through a microprojector, and by a collection of _ representative cactaceous genera loaned by Mr. Jack Whitehead. 176 MADRONO [ Vol. 6 September 18, 1941. Meeting, 2093 Life Sciences Building, University of California, Berkeley, at 7:45 P.M. The President, Professor E. B. Babcock, occupied the chair. Dr. Ralph Emerson, Instructor in Botany, University of California, Berkeley, pre- sented an illustrated lecture on, “An Interpretation of the ‘New Systematics’ in the Lower Fungi.” October 16, 1941. Meeting, 2503 Life Sciences Building, University of California, Berkeley, at 7:45 P. M. The President, Professor E. B. Babcock, occupied the chair. The President appointed a Membership Committee, consisting of Dr. Alan A. Beetle, Dr. Ralph Emerson and Dr. Reed C. Rollins. Mr. E. Yale Dawson, Teaching Assistant in Botany, University of California, Berkeley, presented an illustrated lecture on, “Exploring for Algae in the Gulf of California.” November 27, 1941. Meeting, 2093 Life Sciences Building, University of California, Berkeley, at 7:45 P. M. The President, Professor E. B. Babcock, occupied the chair. Professor H. E. McMinn, Chairman of the Nominating Committee, submitted the names of the following candidates: President, Dr. Alva R. Davis; First Vice-President, Dr. Palmer Stockwell; Second Vice-Presi- dent, Dr. Reed C. Rollins; Secretary, Dr. Mildred E. Mathias; Treasurer, Dr. William M. Hiesey. A motion, made and sec- onded, to accept the report of the nominating committee, was passed unanimously. There were no nominations from the floor. Dr. Howard S. Reed, Professor of Plant Physiology, University of California, Berkeley, presented an illustrated lecture on, “The Migrations of the Citrus Fruits.” Following are the names of persons who have afhliated with the California Botanical Society since the publication of the list of members, April, 1940: Dr. Gordon D. Alcorn; Dr. Vada H. Allen; Mr. William H. Baker; Dr. Fred A, Barkley; Dr. J. T. Barrettg Mr. Kay H. Beach; Mrs. Margaret K. Bellue; Mr. W. Sidney Boyle; Dr. Muriel Bradley; Mr. Conrad M. Brown; Mr. Alfred J. Burrows; Dr. Marion S. Cave; Dr. Walter P. Cottam; Dr. Alden S. Crafts; Dr. Robert A. Darrow; Miss Bernice E. Doyel; Mr. M. B. Dunkle; Mr. Erl H. Ellis; Dr. Ralph Emerson; Dr. Katherine Esau; Professor A. O. Garrett; Mr. M. French Gilman; Dr. T. Harper Goodspeed; Mr. Arthur H. Holmgren; Mr. J. Wendell Howe; Mrs. Lilla Leach; Mr. Harlan Lewis; Dr. W. H. Long; Mr. Elmer Lorenz; Mr. Gregory S. Lyon; Miss Lois McCasky; Mr. James B. McNair; Mr. Oliver V. Matthews; Mr. Roy D. Metcalf; Mr. Reid Moran; Mr. Thomas Morley; Dr. W. C. Muenscher; Mr. David G. Nichols; Mrs. Doris K. Niles; Mr. M. A. Nobs; Mr. Donald Culross Peattie; Mr. Frank W. Peirson; Mr. R. L. Pie- meisel; Dr. Ethel I. Sanborn; Dr. Michael Shapovalov; Mr. Ed- ward Stuhl; Miss Helen Talbot; Mr. John M. Tucker.—Lincotn CONSTANCE, Secretary. 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 Single numbers 2.2/3.5): 0.75 The subscription price of MADRONO is $2.50 per year. We solicit your pat- ronage, Address all orders to: William Hiesey Bus. Mer. Carnegie Institution of Washington Stanford University, California hay ' ¢ ny 2) } My F Wy i i « } VOLUME VI NUMBER 6 MADRONO A WEST AMERICAN JOURNAL OF BOTANY Contents DEVELOPMENT OF THE FEMALE GAMETOPHYTE IN ERYTHRONIUM HELENAE AND ERYTHRONIUM TUOLUMNENSE, Marion S. Cave ..................0005. 177 Some CHEmIcAL PRoPerties OF EvUcALyptus IN RELATION TO THEIR EvoLu- TiIonARY Status, James B. McNair ............ 00... ec ees 181 GRASSLAND AND RELATED VEGETATION IN NortTHERN Mexico, Forrest Shreve 190 teaPAnRAt, #. P. Cronemilter oo). je ee a Us eas wie se BOS Nores on PoLtemoniIACcEAE, Hebert L. Mason ............... 0.2.0.0 00.020 200 EsuropHyton HeEtter: A Vain GENUS OF THE ORCHIDACEAE, Louis O. RARER) CE LOMO EWAN MEOW eho Oakes ts al), ara ty ASTER) ayn 205 Review: Harlan P. Kelsey and William A. Dayton, Standardized Plant PMC MACTTET PETE (hs, NEASON) cir Welk case 6 kG eleth cele aie ele cy Galen de See 206 PROCEEDINGS OF THE CALIFORNIA BOTANICAL SOCIETY ...................... 208 Published at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania April, 1942 BY: ae MADRONO A WEST AMERICAN JOURNAL OF BOTANY Board of Editors Hersert L. Mason, University of California, Berkeley, Chairman. LeRoy Asrams, Stanford University, California. Epcar Anperson, Missouri Botanical Garden, St. Louis. Lyman Benson, University of Arizona, Tucson. Hersert F. Copetann, Sacramento Junior College, Sacramento, California. Ivan M. Jonunston, Arnold Arboretum, Jamaica Plain, Massachusetts. Mitprep E. Marutas, University of California, Berkeley. Bassett Macume, Utah State Agricultural College, Logan. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board—EtuHet Crum Department of Botany, University of California, Berkeley Business Manager—Wit1t1iam 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: 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. 1942] CAVE: FEMALE GAMETOPHYTE IN ERYTHRONIUM 177 DEVELOPMENT OF THE FEMALE GAMETOPHYTE IN ERYTHRONIUM HELENAE AND ERYTHRONIUM TUOLUMNENSE Marion S. Cave The type of development of the female gametophyte which Bambacioni (2) described for Fritillaria in 1928 has been shown to occur in the first five of the ten genera listed by Hutchinson (10) in the tribe Tulipeae: Erythronium, Fritillaria, Tulipa, Gagea, Lilium, Lloydia, Nomocharis, Notholirion, Giraldiella, and Calochor- tus. In this type of development, of the four nuclei resulting from the two divisions of meiosis, three migrate to the chalazal end of the embryo sac while the fourth remains at the micropylar end. At the following division the latter divides normally giving rise to two haploid nuclei. The chromosomes of the three divid- ing chalazal nuclei become aligned on one spindle, thus giving rise to two daughter nuclei, each of which has 8n chromosomes. This second group of four nuclei goes through the final division to produce an eight-nucleate gametophyte with four haploid nuclei at the micropylar end and four triploid at the chalazal. The primary endosperm nucleus is formed by fusion of a polar nucleus from each group and is thus 4n. Representatives of the five genera in which the Fritillaria type of female gametophyte development is known are: Fritillaria per- sica (2); Tulipa Gesneriana (3), and a triploid form of the same (4); Erythronium Dens-canis (9), and E. japonicum (138); Lilium Henry (6), L. philippinense (15), L. tigrinum (20), and Cardio- crinum cordatum Makino (= Lilium cordifolium Thunberg) (12); Gagea minima and G. lutea (11, 19), G. ova, G. graminifolia and G. tenera (14). So far as the author knows no work has yet been published on Lloydia, Giraldiella, Notholirion and Nomocharis. Calochortus (5) is the only genus of the tribe thus far shown to have the “‘nor- mal” type of macrosporogenesis and development of the female gametophyte. In Erythronium both the Fritillaria and Adoza types of develop- ment have been described. Hruby (9) described the process in Erythronium Dens-canis with twelve pairs of chromosomes up to the stage where three nuclei are seen at the chalazal end and it may be assumed that this species would hold to the Fritillaria type. Oikawa (13) has followed the complete development in Erythronium japonicum (n=12) and found it to be of the Fritil- laria type. However, Cooper (7) has found the Adoza type in a 22-paired Erythronium albidum. Schaffner (17) described game- tophyte development for Erythronium americanum and E. albidum (both with twelve pairs of chromosomes). From his description and figures development seems to follow the Adoxa scheme, but Manroxo, Vol. 6, pp. 177-208. April 17, 1942. 178 MADRONO [Vol.6 since this work was done prior to 1928 there may be a possibility that Schaffner missed seeing the chalazal fusion as did all other workers before this time. Guerin (8) stated that the develop- ment in Erythronium Dens-canis is analagous to that which has been known “depuis longtemps chez Lilium, Tulipa, et Fritillaria.” According to Applegate (1) there are three to four species of Erythronium in Eurasia, five in North America east of the Rockies, and fifteen west of the Rockies. Since those of eastern North America and Europe and Asia apparently show differences in the — type of development of the female gametophyte, it was thought that study of this point in some of the western North American species would be of interest. Investigation of two Californian species, Erythronium helenae Applegate of section Concolorae and i. tuolumnense Applegate of section Pardalinae was therefore undertaken. Mareriats. Ovaries of these two species of Erythronium were fixed in CRAF, dehydrated by normal butyl alcohol, and em- bedded in paraffin. Sections were cut at 10, 15, and 20 microns. Three stains were employed: Heidenhain’s iron alum haematoxy- lin, Stockwell’s modification of Fleming’s triple stain, and the Feulgen stain, counterstained with fast green. Haematoxylin was perhaps the best for sections at 10 microns but the Feulgen and fast green was by far the best for thicker sections. DEVELOPMENT OF THE FEMALE GAMETOPHYTE The embryo sac mother cell (the archesporial cell) is located directly below a single layer of nucellus cells (pl. 20, figs. 1 and 2). The cytoplasm is finely vacuolate throughout with a some- what denser layer around the spindle. At metaphase of the heterotypic division (pl. 20, fig. 3) it was impossible to determine the exact number of bivalents since all metaphase plates seen were in more than one section and some of the bivalents may have | been cut. However, there seemed to be around twelve in E£. helenae. No heterotypic mitoses were observed in E. tuolumnense, but somatic plates indicated no large number such as Cooper (7) found in EF. albidum. The two nuclei resulting from the heterotypic division lie at opposite ends of the embryo sac (pl. 21, fig. 4). There is no rest- ing period at this time and the homeotypic division proceeds at once. The spindles of this division also are surrounded by a dense layer of cytoplasm (pl. 21, figs. 5 and 6). No vacuole is yet observable in the sac. At the end of this division there are four haploid nuclei in the sac. In Erythronium helenae, but not in E. tuolumnense, faint lines which are apparently fibers are often seen connecting the nuclei. Text figure 1 which is a camera lucida drawing of figure 8 (pl. 22) shows that fibers connect the two central nuclei not only with each other but also with the nuclei at the micropylar and chalazal ends. Stenar’s (18) figures 8b and 7c show connecting fibers in 00 XIV —. evitejoy ‘a Ul UoTstATp oIdAjosozyoy Jo oseydeyo “E “Sly ‘asuaumnjon) “A UL UOTstaAIp oIdAjosojyoy Jo oseydoig °Z ‘BSI ‘apuajay “Ay UL [Jad LayzouL oes OAIQUIa IO [TeLLodsopody [Sly “WOINOUHLAYY NI ALAHdOLIWVS TIVWAA GO LNAWdOTIATQ “0% TLVIG Te see aa eat Pe ee ny re ‘OOL X ITV. ‘Seppurds punowe uolso1 d1ruIse]doyAD osSUspP ZION ‘avUa]aY “47 UL UOISIAIP PUODIS JO ossvYydodg “9 “BIy~ “asuauunjonz “WY UL UOISIATP puodes Jo sseydotg "GBI ‘apuajay “ff UL UOISTATp od AjOA9}OY WOLF BuUYl[NSat lapNu OMT, “Pf “Sly “WAINOYHLAUY NI ALAHAOLAWVOS AIVWaAT AO ENAWAOTIATGQE “1G ALVIg 1942] CAVE: FEMALE GAMETOPHYTE IN ERYTHRONIUM 179 Gagea lutea and G. minima in the same configuration. Figure 7 (pl. 22) shows three nuclei at the micropylar end connected by fibers and the lowest of these furthermore connected with the nucleus at the chalazal end (in the adjacent section). Bamba- cioni and Giombini’s (3) figure 9 in Tulipa Gesneriana shows a similar arrangement of fibers. In Ro- manov’s (14) photomicrographs of Gagea graminifolia in his figure 7c the three chalazal nuclei can be seen con- nected by faint fibers. Joshi (11) men- tions secondary spindle fibers arising to connect the homeotypic spindles in Gagea fascicularis. Since these four haploid nuclei are the result of meiosis and therefore cor- respond to the four macrospores of a “normal” macrosporogenesis which are separated by cell walls, it may be that the fibers under discussion represent vestigial traces of cell wall initiation between the spores. If walls should form across all the fibers in figures 7 and 8 (pl. 22) each nucleus would then 6 be separated frum every other as in Fie. 1. Four haploid Spore formation. A somewhat similar nuclei in Erythronium hel- condition exists in a young multicellu- enae all connected by fibers Jar endosperm where fibers arise to con- (camera lucida drawing of pect all the nuclei, and walls separating embryo sac shown in pl. 22, fig. 8). them then appear across the fibers throughout the endosperm. However, no walls are produced between the haploid nuclei in the embryo sac and the fibers have completely disappeared by the beginning of third prophase. In both species three of the four haploid nuclei migrate to the chalazal end of the sac. In E. tuolumnense a large vacuole sepa- rates the three chalazal nuclei from the micropylar and a smaller vacuole is found at the center of the three (pl. 22, fig. 9). In E. helenae no vacuole is present until the second four-nucleate stage (pl. 23, fig. 12 and pl. 24, fig. 14). At prophase of the third division the three chalazal nuclei start to fuse (pl. 238, figs. 10 and 11). Often from telophase of the second division through metaphase and telophase of the third in E. helenae many small bodies in the cytoplasm which stain black with haematoxylin, but do not stain with the Feulgen tech- nique, are seen (pl. 28, fig. 12). They are pictured by Schaffner (17) in E. albidum in his figure 59 although this is apparently only the telophase of the first division. Hruby shows them at the telo- phase of the second division and suggests that they are leuco- 180 MADRONO [Vol. 6 plasts since they stain faint blue with picroindigocarmine. The writer has seen them also at the third division in Lilium longiflorum and Sargant (16) shows them in Lilium Martagon. They were also noted by Stenar (18) in Gagea lutea. In the third division it is apparent that there are many more chromosomes in the mitosis at the chalazal end than in that at the micropylar (pl. 28, fig. 12 and pl. 24, fig. 13). Figure 13 (pl. 24) is a polar view of the chalazal nucleus at anaphase of the third division. All the chromosomes are not in focus but many more than the haploid number seem to be present. In both species the second four-nucleate stage is charactem ized by two large elongated nuclei at the chalazal end separated by a vacuole from the two more nearly spherical nuclei at the micropylar end (pl. 24, fig. 14). A fourth division occurs to give rise to an eight-nucleate female gametophyte (pl. 24, fig. 15). The inner chalazal nucleus divides into two nuclei by means of a more or less abortive mitosis (pl. 24, fig. 15). In flowers which had been open several days fusion of the polar nuclei had not yet taken place. SuMMARY 1. A study was made of the female gametophyte of two Cali- fornian species of Erythronium, E. helenae Applegate and E. tuo- lumnense Applegate, with a view to comparing their development with that found in eastern North American and Eurasian species. 2. Three of the four haploid nuclei resulting from meiosis migrate to the chalazal end of the embryo sac while the fourth goes to the micropylar end. In the following division the micro- pylar nucleus divides normally giving rise to two haploid nuclei, whereas the chromosomes of the three chalazal nuclei become aligned on one spindle and produce two 8n daughter nuclei. Each of the four nuclei now found in the embryo sac divides nor- mally once again to give rise to an eight-nucleate gametophyte with three haploid nuclei at the micropylar end, three triploid at the chalazal, and one haploid and one triploid polar nucleus in the center. 3. In the two Californian species of Erythronium investigated development of the female gametophyte therefore follows the Fritillaria scheme as is true for the Eurasian species, while the Adoxa scheme occurs in Erythronium albidum from eastern North America. Department of Botany, University of California, Berkeley, September 5, 1941. LITERATURE CITED 1. Apvprecatr, E. I. The genus Hrythronium: a taxonomic and distributional study of the western North American species. Madrofio 3: 58-113. 1935. ‘OOL X ITY ‘SepTonoea Aq popJearedas asuaumnjony “A jo joponu projdey «nog °6 “SIT (1 ‘SY 4X0} 99S) SNDOJ JO JNO IIe YOY Jo sUIOS ‘sdoqy AQ Poa}ooUUOD [[B anUa]—Y “ff UL leponu projdey «nog ‘g “SI “Uses oq UB). Sdaqy SuLTd9UUOD 4Jnq UOT}IeS JUDE pe vy} UL ST SNoponuU [ezZeTeYO sy} Ssaaqy Aq PeyoouUoy lo[onu projdey ANOJ IY} JO IIY} LAVUAIOY “PF CL ol a | “WOINOUH LAMY NI ALAHdOLAWVD PIVWAE AO TNAWdOTIATC “CG ILVId “a ur aseydojo} party, ‘21 SIA ‘00OL X ITV ‘pue [ezeyeyo ye sosseUl UT}PEUIOAYO Josie, pue Ssotpog yzlep sjou :anuajay ‘aqpurds 9uoO UO PoUSI[e SuIM0d9q TepPoNU [eze[RYO JO SoUIOSOWOAYO OT “Sly UeY} oye, ATUSITS ‘TL ‘Sty ‘apuajay “yp ut oseydoad pany, ‘OL (Sty “WOINOUHMLAUY NI ALAHAOLANVD ATVWad AO ENAWAOTIATG “§Z ALVId “anuajay a JO ay ‘gq ult sseydeue Aydojyourt eet ee Ae 2 > 3 o[PUloy 94 MITA LBTOg Bd] onu-}Ys ‘SL ‘SI TH “WOINOUWH SI 31 LAU “apua] aU NI a Taal AHdOW 93R} AWVD S r TIVWdH WO Ny J}RI[ONU Ino} puo DIS v NAW dOTHATCT I Re | 00g X ILV ‘ODUa]AY 'S ALVIG 1942] CAVE: FEMALE GAMETOPHYTE IN ERYTHRONIUM 181 2. Bampacioni, V. Richerche sulla ecologia e sulla embriologia di Fritillaria persica L. Ann. di Bot. 18: 7-37. 1928. 3. BAMBACIONI, V. and Giomsini, A. Sullo sviluppo del gametofito femminile in Tulipa Gesneriana L. Ann. di Bot. 18: 373-386. 1930. 4, Brettows, J. M. and Bamrorp, Ronatp. Megagametophyte development in a triploid tulip. Bot. Gaz. 102: 699-711. 1941. 5. Cave, Marion S. Megasporogenesis and embryo sac development in Calo- chortus. Amer. Jour. Bot. 28: 390-394. 1941. 6. Coorrr, D. C. Macrosporogenesis and development of the embryo sac of Lilium Henryi. Bot. Gaz. 97: 346-355. 1935. 7, ——_——————.. Development of megagametophyte in Erythronium albidum. Bot. Gaz. 100: 862-867. 1939. 8. GueERIn, P. Le developpement de loeuf et la polyembryonie chez Ery- thronium Dens-canis (Liliacées). Compt. Rend. Acad. Paris 191: 1369— 1372. 1930. 9. Hrury, K. A contribution to the cytology and embryology of Erythronium Dens-canis L. Bull. Inter. Acad. Sci. Bohéme. Pp. 1-9. 1934. 10. Hurcuinson, JouHn. The families of flowering plants. II. Monocotyledons. London. 1934. ll. Josu1, A. C. Development of the embryo-sac of Gagea fascicularis Salisb. Bull. Torr. Bot. Club 67: 155-158. 1940. 12. Orxawa, K. A note on the development of the embryo-sac in Cardiocrinum cordatum. Sci. Rep. of the Tohoku Imp. Univ. ser. 4, Biol., 11: 303- 306. 1937. 13. ————_———. The embryo sac of Erythronium japonicum. Bot. Mag. Tok. 54: 366-369. 1940. 14. Romanov, I.D. Die Embryosackentwicklung in der Gattung Gagea Salisb. Planta 25: 438-458. 1936. 15. Santos, J. K. Macrosporogenesis in Lilium philippinense Baker. Cytol. Fujii Jub. Vol. pp. 822-835. 1937. 16. Sarcant, E. The formation of sexual nuclei in Lilium Martagon. I. Oogenesis. Ann. of Bot. 10: 445-477. 1896. 17. ScHarrner, J. H. A contribution to the life history and cytology of Ery- thronium. Bot. Gaz. 31: 369-387. 1901. 18. Stenar, Hence. Uber die Entwicklung des siebenkernigen Embryosackes bei Gagea lutea Ker., nebst einigen Bemerkungen tiber die Reduktion- steilung bei Gagea minima Ker. Svensk bot. Tid. 21: 344-360. 1927. 19. Westercarp, M. A cytological study of Gagea spathacea, with a note on the chromosome number and embryo-sac formation in G. minima. Compt. Rend. des Trav. du Lab. Carlsberg. Sér. Physiol. 21: 437-451. 1936. 20. WestFaLL, J. J. Cytological studies of Lilium tigrinum. Bot. Gaz. 101: 550-581. 1940. , SOME CHEMICAL PROPERTIES OF EUCALYPTUS IN RELATION TO THEIR EVOLUTIONARY STATUS JamMES B. McNair This paper presents comparisons of morphological and chemi- cal characteristics of members of the genus Eucalyptus and shows that primitive species are primitive both morphologically and chemically and more recent species are advanced both morpho- logically and chemically. Chemical advance involves oxidation. But chemical and morphological advances do not necessarily progress hand-in-hand. Baker and Smith (1, 2) after thirty years work classified the genus chemically and phylogenetically. Their phylogenetic ar- 182 GROUP I PINENE C,H, CINEOLE C,H,0,4% (FEATHER VENATION MADRONO GROUP IL PINENE CH 0 6 CINEOLE CHO, 25 °/, [ Vol. 6 GROUP IIa, It PINENE C,H CINEOLE CoH,0, 49-64 °/ STRINGYBARKS 19 an 13-14 18 Hee a 10) 25 IRONBARKS 62-28. 42 é 0 0 58-BB8t-87b-5!-75b-74 b-76b-B3b-894-72b- BEb-73b 8 a8 | Ns Q 2746-45-44 794-60-90b-49-48-8.4b-54-70 80b-7 7b-78b-B5b- lb B 29-34-32-35-0 47-50-61-55-56-53-52- 67-57 8 neo Si / ere 63-68-69-66-65-59-64-82 a 394 38-37- 40-30-31 ; 5————— 56 7 ; 4G. : -33-43 GROUP IZo,Zb GROUP GROUP WI GROUP Wa, We PINENE LESS PINENE PINENE CINEOLE 15-3 °% CINEOLE 56-46% PHELLANDRENE CiHig CINEOLE 16% PHELLANDRENE CINEOLE 13°% PHELLANDRENE PHELLANDRENE PIPERITONE CyH,O (BUTTERFLYWING VENATION) AROMADENDRAL [CoH 20 Cio toe 135-136 eo I25q-l27-128 142 I55a 1494-150 o-!Sla-106 Wt-15 307 !520 PEPPERMINTS \ 37-139-lA3 159b-I58b-1574-O pcs 144VI-F107Y-94.-082-93.-iO9W-6W-1I4V-II3U-9 24 140-26 138 16lb-160b 95lWa 141 129 105 We-I26W1-1594.0-163 0-162 0-132 104 Wb-l65b-1645-I666 99e-!Ola-l00a- 960 156 b-I31M-O 134-133-103 We \47-145-146-148 Fic. 1. Phylogenetic tree of Eucalyptus species portraying both systematic and chemical relationships. rangement is shown in figure 1 and their seven chemical groups are given in Table 1. If we consider the groups of Baker and Smith in relation to the phylogenetic tree made by them it is ap- parent that each group represents a horizontal cross section of their phylogenetic tree (fig. 1) and consequently sometimes in- cludes heterogeneous species which are products of different branches of descent (not natural groups or genetic sequences) and sometimes includes reversions. In other words each Baker and Smith group represents a stratum or phase in the evolutionary development of the genus. So we find (fig. 1) that group II in- cludes three separate natural sequences, 22. E. Wilkinsonia, 23. E. eugenioides, and 24. E. umbra all descendants of 18. E. nigra; 42. E. peniculata from 9. E. botryoides; and 36. E. maculata and its numer- ous following from FE. terminalis. Group VI has five separate sys- tematic series and in groups IVa and IVb are found examples of natural sequences on a morphological basis which include a vari- ety of interspersed chemical groups. For example 144. E. obliqua of group VI chemically is followed morphologically by 107. E.— tereticornis of group V chemically, followed by 94. E. tereticornis var. cineolofera of group IV class (a) chemically, ete. Consequently it is apparent from figure 1 that these plants 1942] McNAIR: CHEMICAL PROPERTIES OF EUCALYPTUS 183 may sometimes combine advanced systematic characters with more primitive chemical characteristics, for example, 105 IVb which is chemically in group IV but morphologically in group VII. In still other cases primitive morphological characters seem to have been retained by plants in chemically advanced positions, for example, 114 V between 116 V and 113 V. It may be con- cluded, therefore, that the development of chemical character- istics and morphological characters do not necessarily proceed hand-in-hand; one may proceed more rapidly or less rapidly than the other. An excellent example of this is seen in the case of FE. dives, E. radiata and E. micrantha. Eucalyptus dives, the common broad leaf peppermint, occurs in Australia over vast areas and the oil from the leaves has become of importance owing to the occurrence in it of from 40 to 50 per cent of the ketone piperitone, a commercial source of thymol and menthol. With the increased economic demand it'»was found that certain oils said to be obtained from the leaves of E. dives only yielded 5 to 20 per cent of the ketone. These were at first re- garded as adulterated. It was, however, shown that they were genuine oils and that E. dives existed in at least four varieties which were morphologically absolutely indistinguishable both in the field or in the herbarium. At first sight this difference might be expected to be due to soil or climatic conditions but this is not the case, since the different varieties may grow side by side in the field and breed true when grown in pots (7, 8,9, 10). The young seedlings from all four kinds were morphologically identical, yet when tiny fragments of the leaves, even from plants only one and one-half inches in height were rubbed between the fingers the characteristic odor of each particular kind was readily detected. As a result of investigations by Penfold and Morrison (11, 12, 13) extending over a period of ten years it has been found possible to separate morphologically identical trees of E. dives, E. radiata and E. micrantha into groups as a result of hybridization based upon the chemical composition of the volatile oils. INTERRELATIONSHIPS BETWEEN PRINCIPAL CONSTITUENTS The principle constituents of Fucalyptus oils are the two ter- penes, pinene and phellandrene, the oxide cineole, the group of aldehydes and a ketone known collectively as aromadendral and the ketone piperitone. PineneE. The pinene occurring in Eucalyptus oils is alpha pinene and it is found in two forms, one which deflects a ray of polarized light to the right (dextro) and the other which deflects to the left (laevo). PHELLANDRENE. This terpene is somewhat extensively dis- tributed in the oils of certain groups of Eucalypti. It is more pro- nounced in those belonging to the more recent end of the genus, and it occurs more abundantly in those species common to the [ Vol. 6 8°85 — 088" 0°% | [PAS $9898" ie | i= £S06° iG | Or — 9606" ~ MADRONO ISBLIIAV ¢ . joyoo|e DD SI yuao 19d 0g aa Ayaeas —0) S}ied ur et ayroads AYTIGNTOS SSsr 1 FUOSSICes qyuosaid § ‘1aae | ae (Q)ILA S6LV'T quoSoad 4 ISL] GI “AdAR = OF Secret: cast en (®)IIA CYST I sone ete) istsier isc |P ne, levrieaselsiar eo telire quasaid ra | “IDAR | — OF | quasaid TEN Z'ON OOSTAlg es ea quesoid d1Bd QT ‘doe | dnoax — OF qussaid A I3.1e'] c8sor L go fendu, derdar tes ee All” Be) Sah eets eran res ter xe yuasoid OF “IQAPR + OF wuts Meuse. c.4) moms (q) AI HOSP Tee yuosoid qyuoasqe 9G “IIAP SUIYST + OF -uyarp | (®) AT OCOTs sul. = qyuesqe yuasqe A ON | +9G | = eBaeT | (Q)ITT OLOLT ie o1e1 yuosqe 6h “1oAe 0S-0F | odaieT | (®) IIT TON COLP a yuasqe yuesqe GZ ‘ADAG dno1ry — OF Id 1B] at aie] GGipip> (os teee Ae yuasqe | gp “AOA IIB | ISLC] I ‘DO 006 XOpul DATPRAJIY OH") eee jeqd Arg O ei e) OH Ore) -11odig ene : : [eipuop -eUlOLY SET SES) udip -ue][e4d I3eR -yud0.10d OH") g[OIUIA -IVULSOYV ) SdNOUr) TIM SALMATVONYY AO SAILYAdOUd TVOIWAH,) GNV 'IVOISAHG ANV SAONVISHNAS TVdIONIYd dO SLNNIOWY 184 ad 2 Ae) guoulg ‘ON dno1y (NVINVWSV], ONV NVI ‘[ aTav I, 1942] McNAIR: CHEMICAL PROPERTIES OF EUCALYPTUS 185 eastern and southeastern portion of the continent of Australia and ‘Tasmania. CineoLte. The constitution of cineole is probably that of cinyl oxide, CipHigO. It has been described under the names eucalyp- tol (its most common commercial name) and cajuputol. When terpin is dehydrated cineole is produced. Cineole is an internal ether produced by the elimination of water between the two hydroxyl groups in terpin. Although we have no direct evi- dence as to the mechanism of the formation of cineole in plants, its very frequent occurrence in oils containing g-terpinene can be accounted for readily if it be assumed to be formed from either a-pinene or q-terpineol, when terpin may be regarded as an inter- mediate product (16, 382). The oils from the group of Eucalypti known as “‘gums” usually contain a fairly large amount of cineole, together with pinene, and in the case of many members belonging to this group, the cineole increases in amount when the oils are stored. This increase in cineole through possible oxidation during storage apparently con- firms the fact that cineole is an oxidation product of pinene. AROMADENDRAL. The term aromadendral is used to denote the presence of one or more members of a group of characteristic aldehydes and a ketone in Eucalyptus oils. These aldehydes in- clude cuminaldehyde (cuminal), phellandral and cryptal; they do not seem to occur in the oils of the earlier members of the genus (the pinene yielding group), nor in those of the more recent species, particularly those in which phellandrene is the more pro- nounced terpene. In these latter species the characteristic con- stituent is the ketone piperitone. It may also be stated as a general rule that cymene is present in either larger or smaller amount in the oils of species containing these aldehydes. Wallach (17) has shown that the oxidation of B-phellandrene produced a glycol, which on treating with dilute H2SO,4 gives dihydro- and tetrahydrocuminaldehydes. Molecular re-arrange- ment of a somewhat similar character may perhaps take place naturally. PirpERITONE. Piperitone is the peppermint ketone of Eucalyp- tus oils. It is an unsaturated ketone C,9HigO with one double bond. It appears to occur only in the oils of species occupying the more recent end of the genus and is not found in the oil of any member of the groups occupying the anterior position, in the evo- lutionary sequence of the genus. In the oils of most species, piperi- tone is found associated with the corresponding |-rotatory secon- dary alcohol piperitol. Phellandrene in Eucalyptus oils is often associated with |-piperitone, but not always, although it may be accepted that the most pronounced phellandrene Eucalyptus oils always contain this ketone in smaller or larger amounts. Hughesdon, Smith and Read (4) have directed attention to the fact that l-piperitone always occurs in nature in association with 186 MADRONO [Vol.6 |.49 92 92 fey L48 91 x< x : a) . oe 30 a7 = < 2 iS) = (20 = 45 are O : Ee e gy -89 1.46 in a 89 46 uy 88 I I To Tw me Wo» VW Vw Vw OIL. GROUPS OIL. GROUPS CINEOLE CINEOLE PERCENT OF CINEOLE Row es PERCENT OF CINEOLE IL Te Tw Ma Vo VV Vil@ VIL) OIL. GROUPS OIL. GROUPS Fic. 2. Graphs showing the specific gravities, refractive indices and per- centages of cineole of Eucalyptus oil groups. l-g-phellandrene and I-piperitol and it is possible that some rela- tionship exists between these substances. In the laboratory it has been found that the reduction of either d- l- or dl-piperitone with sodium in alcoholic solution yields dl-isomenthols and. dl- menthols, some dl-q-phellandrene being formed simultaneously. CoMPARISON OF SPECIFIC GRAVITIES AND REFRACTIVE INDICES Based on their specific gravities (and concomitant refractive indices) Eucalyptus oils may be divided into two large groups. The first and most primitive group includes groups I, II, IIa and IIIb of Baker and Smith (2). (Table I and text fig. 2). The second and less primitive group includes groups IVa, IVb, V, VI, Vila and VIIb of Baker and Smith (2). In figure 2 it so happens that a straight line will pass near all 1942] McNAIR: CHEMICAL PROPERTIES OF EUCALYPTUS 187 the points representing the specific gravities of groups I to IIIb inclusive and that another straight line will pass near all the points representing refractive indices. Therefore, the rule that a high specific gravity is accompanied by a low refractive index and vice versa for plant volatile oils (5) is confirmed. It is noticeable that a straight line cannot be drawn through all of the five points representing the refractive indices of groups IV to VIIb, and that the line of closest fit is a curve. Conse- quently it is obvious that the mixture of substances which consti- tutes the volatile oils of at least some of the members of this group are different from those of the groups I to IIIb where straight lines can be drawn approximately through all the points repre- senting the specific gravities and another straight line will lie near to all of the points representing refractive indices. Differ- ent mixtures of substances are also indicated in figure 2 by the fact that for groups I to IIIb the refractive index decreases with advance in plant evolutionary position while for groups IV to VII the reverse is true. Likewise there is a difference in the direc- tional change in specific gravities for groups I to IIIb and IV to VII. In groups I to IIIb the specific gravity increases with ad- vance in plant evolutionary position while in groups IV to VII the reverse takes place. INCREASE IN OXYGENATED BODIES WITH EVOLUTIONARY PROGRESS According to the general theory for angiosperm volatile oil behavior, the specific gravity should increase and the refractive index should decrease with advance in evolution (6). The oils of groups [Va to VIIb contain principally pinene, its oxidation pro- duct cineole, and phellandrene and its oxidation products, cuminal, cryptal, phellandrol and piperitone. From Table 1 it is seen that pinene and cineole are diminishing while the other substances mentioned are increasing with progress from IVa to VIIb. The removal of a heavy substance from the oil would decrease its spe- cific gravity. Therefore the subtraction of pinene which has a specific gravity of 0.865 at 15 degrees would not tend to decrease the specific gravity of the oil because its specific gravity is lower than that of the average specific gravity of groups IVa to VIIb (Table 1). However, if cineole be taken out the specific gravity would tend to lower because cineole with its specific gravity of 0.930 at 15 degrees is greater than the average specific gravity of the oils of the groups here considered. The content of cineole rapidly diminishes (56 per cent to 3 per cent) in these groups. The other main oxygenated bodies of these groups also have spe- cific gravities near that of cineole. Consequently neither the amounts of these compounds nor that of cineole increases from groups [Va to VIIb enough to change the diminishing trend of the specific gravity. Should the total amount of the oxygenated bodies increase from IVa to VIIb it is obvious that the specific 188 MADRONO [Vol. 6 gravity would also increase. Consequently we may conclude that the volatile oils of Eucalyptus species when in their native habitat, increase in specific gravity with the increase in plant evolution except where the oxygenated bodies decrease in amount with the increase in evolution. The refractive index of cineole (1.4596 at 20 degrees) is lower than the average refractive index for the group oils. Conse- quently its removal would tend to raise the refractive index of the group oils. Otherwise the refractive index would increase with evolutionary progress. Table 1 shows that chemical analysis has found the number and variety of oxidation products to increase with advance in evolutionary position in the genus. Cineole, the oxidation prod- uct of pinene, is found in all of the groups. Aromadendral (which is mostly cuminal and cryptal oxidation products of phel- lJandrene) is found in groups III, IV and V. Piperitone with which is associated phellandrol, both oxidation products of phel- landrene are found in group VII. The increase in the number and variety of oxidation products is likewise shown by the increase in the solubility of the crude oils in 80 per cent alcohol from groups IVa to VIIb (Table 1). For the oxidation products are soluble in alcohol whereas the terpenes, pinene and phellandrene are comparatively insoluble. The amount of cineole rapidly decreases from group [Va to VIIb and yet the solubility increases. Consequently some other oxidation products must replace the removed cineole. OptTicaL RoTATION A glance at Table 1 shows clearly that the average optical rotation becomes more laevo-rotatory as the evolutionary se- quence of the groups advance. There are a number of dextro- and laevo-rotatory substances in the oils. There are present in varying amounts both dextro- and laevo-pinene and d- and ]-phellandrene. There are also |l-piperitone, d-terpinol and d- eudesmol. Most of the pinene is dextro- and most of the phel- landrene is laevo-rotatory; d-pinene accounts for the predomi- nant d-rotation in groups I, II, Hla and IIIb and ]-phellandrene and |-piperitone for the predominant ]-rotation in the remaining groups. ONTOGENY AND PHYLOGENY The Eucaly pti furnish an excellent chemical counterpart to the morphological theory of Haeckel (3) that ontogeny recapitulates phylogeny, that the organism in its development is to a great extent an epitome of the form modifications undergone by the suc- cessive ancestors of the species in the course of their historic de- velopment. Oil from the younger seedlings contains more d- pinene and less cineole (that is more hydrocarbon and less oxy- genated products) than does that from the saplings two or three 1942] McNAIR: CHEMICAL PROPERTIES OF EUCALYPTUS 189 years old, and the maximum cineole content is reached in the oil collected from older trees (2). This is true also for the leaves which are reproduced from lopped old trees, and the oil from seven months “suckers” contains more cineole and less pinene than does that from twelve months old seedlings, while that from fifteen months old “suckers” follows the same rule in respect to two and one-half year old seedlings. The increase in the amount of cineole with progress in evolution in Large Group No. 1 of the genus is shown in Table 1. SUMMARY There has been orderly evolution in volatile oil characteristics of the Eucalypti which may be correlated with changes in morpho- logical characteristics. However, oil groups represent cross-sec- tions of the phylogenetic tree and not necessarily genetic se- quences or natural groups of the systematist. Botanically distinct species are generally distinguished by their chemical constituents and also in some cases where morphological examination shows little or no difference, chemical analysis of the oil reveals the existence of completely distinct varieties. With progress in evolution the amount of the hydrocarbon (terpene), pinene and its oxidation product cineole increase to a maximum, then decrease; another hydrocarbon (terpene) phel- landrene and its products form a second series of compounds which is evidenced first by the appearance of aromadendral (an aldehyde and ketone mixture containing cuminaldehyde and eryptal) ; next phellandrene becomes a pronounced constituent and finally its ketone piperitone increases in amount. The specific gravity increases with the increase in plant evolu- tion, and the refractive index decreases with the increase in plant evolution, except where the oxygenated bodies decrease in amount with the increase in evolution. The number and variety of oxidation products increase with advance in evolutionary position in the genus. There is a tendency for the optical rotation to become more laevo-rotatory with advance in evolution caused principally by a decrease in d-pinene and an increase in ]-phellandrene and its ketone I|-piperitone. Morphological and chemical phylogeny have their counter- parts in ontogeny. Los Angeles, California, July 3, 1941. LITERATURE CITED 1. Baxer, R. T. and H. G. Smirn. A research on the Eucalypts. Sydney. First edition. 1902. 2, ————_————. A research on the Eucalypts. Sydney. Second edition. 1920. 3. Harcxet, E. H. Natiirliche Schépfungsgeschichte. Berlin. First edition. 1867. 4. HucGuespon, R. S., H. G. Smiry and J. Reap. Piperitone. Part VI. The reduction of piperitone. Jour. Chem. Soc. (London) 123: 2916-2925. 1923. 190 MADRONO ' [Vol.6 5. McNair, J. B. Some properties of plant substances in relation to climate of habitat—volatile oils, saponins, cyanogenetic glucosides, and carbo- hydrates. Amer. Jour. Bot. 19: 168-193. 1932. 6. ———__————.. The evolutionary status of plant families in relation to some chemical properties. Amer. Jour. Bot. 21: 427-452. 1934. 7. Prenroxtp, A. R. and F. R. Morrison. The occurrence of a number of varie- ties of Eucalyptus dives as determined by chemical analyses of the essential oils. Jour. and Proc. Roy. Soc. N.S. Wales 61: 54-67. 1927. 8. ————_———_.. Essential oils of Eucalyptus micrantha (De Candolle) and E. haemastoma (Smith). Jour. and Proc. Roy. Soc. N.S. Wales 61: 267-275. 1927. “9, ————————.. Occurrence of a number of varieties of Eucalyptus dives as determined by chemical analysis of the essential oils. II. with remarks on the o-cresol method for estimation of cineole. Jour. and Proc. Roy. Soc. N.S. Wales 62: 72-78. 1928. 10. —-——————.. Occurrence of a number of varieties of EKucalyptus dives as determined by chemical analysis of the essential oils. III. Jour. and Proc. Roy. Soc. N.S. Wales 63: 79-84. 1929. 11. ———————.. Occurrence of a number of varieties of Eucalyptus radiata (EK. numerosa) as determined by chemical analyses of the essential oils. Jour. and Proc. Roy. Soc. N.S. Wales 66: 181-193. 1932. 12, ——_—_—_—_—__.. Essential oils of Eucalyptus micrantha (De Candolle) in- cluding a form rich in piperitone. II. Jour. and Proc. Roy. Soc. N.S. Wales 67: 351-363. 1934. 13. ————————.. Value of chemistry in the botanical identification of eco- nomic plants. Australian Chem. Inst. Jour. and Proc. 2: 36-41. 1935. 14, ———————_.. Physiological forms of the Eucalypts. Perfumery and Essential Oil Record 26: 178-179. 1935. 15. ———————.. Essential oils of Eucalyptus australiana (Baker and Smith) and its physiological forms. I. Jour. and Proc. Roy. Soc. N.S. Wales 69: 111-122. 1935. 16. StmonsEN, J. L. The terpenes. Vol. 1. Cambridge, England. 1931. 17. WatiacH, O. Zur Kenntniss der Terpene und der atherischen Oele. 72 Abhandlung, Ueber B-phellandren. Justus Liebig’s Annalen der Chemie 340: 1-16. 1905. GRASSLAND AND RELATED VEGETATION IN NORTHERN MEXICO Forrest SHREVE The extensive grassland area of the central United States exhibits its optimum development in Kansas and Nebraska, and extends south to the Mexican boundary only after suffering local- ization in occurrence and modification in character. The plains and gently falling outwash slopes of southern New Mexico and western Texas are largely occupied by a very open type of arid grassland in which Yucca, Nolina, Dasylirion, Agave, Opuntia and various shrubs are conspicuous. This is a transition region, in which the conditions are intermediate between the optimum ones for grassland and for desert. The vegetation is formed by an infiltration of plants from each of these vegetations, with very few dominant species that are distinctive of the transition region. In both of the states mentioned and also in southeastern Arizona there are areas of true grassland growing in favorable valleys or 1942 | SHREVE: GRASSLAND IN NORTHERN MEXICO 191 circling the higher mountains, and in all cases occupying rela- tively deep soils at elevations of 1500 to 1800 meters. At higher elevations in northern Mexico, where climatic and soil conditions are favorable, there are some large areas of grass- land. These are chiefly along the eastern base of the Sierra Madre Occidental and around the higher mountains of northern Coahuila. Grasses are also important in the transitions from grassland to encinal (evergreen oak woodland), to juniper or pinyon woodland, and to oak chaparral. Grasses are likewise important in the cactus-acacia-grassland (“cactus savanna’’), which lies between the southernmost areas of desert and the grassland. In northern Chihuahua there are a number of “‘llanos,” or grass covered plains, which are not related to the true grass- land and its transitions but are distinctly a desert association occurring also in Arizona and Sonora. The aim of this paper is to describe briefly each of these types of vegetation in which grasses play a conspicuous part. Nothing will be said here in reference to the minor role played by grasses in the desert, their greater importance in the arid bushland which occupies the low elevations of northeastern Mexico, nor their secondary role in the forests of northern Mexico. Prior to the beginning of the Madero revolution the grass- lands of northern Mexico were a source of great wealth. The destruction and expropriation of most of the great cattle ranches gave the grasslands a period of over fifteen years with relatively light utilization. During the last seven years the agrarian pro- gram of the Mexican Government has brought some of the best areas of grass into cultivation. A magnificent natural sod has been destroyed in a worthy effort to help inexperienced and poorly equipped farmers to raise profitable crops of corn by dry farming methods. The recent heavy importation of foodstuffs by Mexico, for the first time in its history, is eloquent proof of the difficulties which this program has encountered. The most extensive grasslands are in central Chihuahua, covering an area indicated on the map accompanying a paper by the writer in this journal (4). A continuous belt of varying width extends south through Durango and Zacatecas into Aguas- calientes and northern Jalisco. Most of these areas are in the elevated valleys which lie between the eastern base of the Sierra Madre and the subsidiary ranges which parallel it on the east. In northern Chihuahua grassland is found in suitable situations between elevations of 1600 and approximately 2150 meters. In northern Durango there is a very gradual ascent from the central basin of the Mexican plateau to the summit of the Cuchillo de Zarca at 2000 meters. On this ascent in the region west of Mapimi the desert shrubbery is gradually replaced by grasses, and typical grassland is first met slightly below 1800 meters. In southern Durango, south of the Rio Nazas, grassland is first en- countered at 1925 meters and covers a large area in the central 192 MADRONO [| Vol. 6 and southern part of the state. Between Nombre de Dios, Durango, and Sombrerete, Zacatecas, broad grassland valleys rise to an elevation of 2500 meters. Similar areas also extend north from the city of Durango but are interrupted by extensive malpais areas. In the 1000 kilometers between northern Chi- huahua and central Zacatecas it will be noted that there is a slight increase in the elevations between which grassland is found. In northern Coahuila the scattered belts of grassland lie between 1500 and 1800 meters. The western face of the Sierra Madre Occidental is more pre- cipitous than the eastern and active denudation has prevented the development of large areas suitable for domination by grasses. In northeastern Sonora there are areas and belts of grassland in localities north of Moctezuma and east of Magda- lena. With decreasing altitude the transition from grass to desert is rapid. South of Moctezuma the principal display of grasses is in foothills at elevations of 450 to 900 meters, where coarse bunch grasses grow in the open stands of oak. The Sonoran grasslands are rich in species but differ in composition from the areas in Texas and Chihuahua. On the eastern side of the Mexican plateau, along the western base of the Sierra Madre Oriental, the physical conditions differ greatly from those on the western side. The mountain axis is not so continuously elevated and has no parallel subsidiary ranges on its landward side. The highest peaks are Cerro Potosi (3800 m.) and Pefion Nevado (3664 m.). Between them, and north of the former, are gaps so low that many desert plants extend over the divide into the drainage of the Gulf of Mexico. Another important feature with reference to grassland is the prevalence of limestone along the mountain front in Coahuila and Nuevo Leon. The reluctant weathering of limestone leaves the soil shallow, the surface stony and the deep pockets of soil few. In Mexico, as in the southwestern United States, the lime- stone soils do not support heavy stands of grass, and desert in- variably extends to much higher elevations on limestone than on other types of rock and their derived soils. The nearest approach to areas of typical grassland on the eastern side of the plateau has been found between Mier y Noriega and Soledad, Nuevo Leon, in the lee of the Pefion Nevado range. The elevation ranges from 1700 to 2000 meters and the rainfall at a single adjacent station is 500 millimeters. On the pediments and gently rounded ridges which parallel the base of the mountains the cover is rarely more than 50 per cent grass, the remainder being other herbaceous perennials. There are frequent mottes of Quercus cordifolia from 5 to 6 decimeters high, as well as other shrubs and semi-succulents of greater frequency than in typical grassland. The characteristic grasses in this region are Bouteloua gracilis,. Ficure 1. Ficure 2. Puate 25. GRASSLAND IN NorTHERN MEXICO. PiatTE 25. GRAsstANnD IN NortTuern Mexico. Fig. 1. Looking northeast across the grassland plateau of southern Durango from the lower edge of pinyon woodland in mountains near La Purisima, at 2400 meters elevation. Fig. 2. Looking across bajada with desert shrubbery of Larrea and Acacia vernicosa to small Hilaria Nano, 65 kilometers northeast of Camargo, Chi- huahua, at 1400 meters elevation. Pee bet Me 7 ane POR Tory aes 1942 | SHREVE: GRASSLAND IN NORTHERN MEXICO 193 Triodia grandiflora, Hilaria cenchroides, Lycurus phleoides, and a small slender form of Bouteloua curtipendula. The commonest associated plants are Zinnia anomala, Dichondra argentea, Dyscho- riste decumbens, Dyssodia setifolia, Acalypha phleoides, Florestina tripteris and Houstonia rubra. The precipitation in the Mexican grassland lies approximately between 400 and 500 millimeters on the west side of the central basin and between 500 and 600 millimeters on the east side of the mountains of northern Coahuila. The annual average for the city of Chihuahua, just below the edge of the grassland, is 385 millimeters, while for Parral, just above the grassland, it is 517 millimeters. In the city of Durango the average rainfall is 463 millimeters and at Villa Madero it is 488 millimeters, both localities being in cultivated country that was originally grass- land. Charcas, San Luis Potosi, is at 2060 meters elevation on the eastern edge of the grassland, which is only locally and poorly developed there. The average annual precipitation at Charcas is 411 millimeters. Throughout the grassland areas the late winter and spring are dry and the four months, June to Sep- tember, receive from 64 to 77 per cent of the annual precipitation. The areas herein designated as grassland are occupied by “short” grasses forming a sod or turf which covers 80 per cent or more of the surface. Cacti over 20 centimeters in height, shrubs and trees are rare or absent. Yucca, Nolina, Dasylirion and Hechtia are infrequent except at the lowest altitudes and in the transition from grassland to desert. Closely associated with the grasses is a large number of herbaceous root perennials, many of which are prostrate, low, or of habit and leaf size which make them inconspicuous. The structural and social features of the Mexican grasslands resemble those of the central United States, as described by Clements (1), Weaver and Fitzpatrick (5), Gates (2) and others. The floristic composition is similar to that of western Texas but differs materially from that of Kansas. Among the character- istic grasses of the latter state Agropyron Smithii, Eragrostis spec- tabilis, Sporobolus asper, Aristida oligantha, Scheddonardus panicu- latus, Stipa spartea and S. comata are absent or very uncommon in Mexico. Andropogon scoparius is absent and A. furcatus is wide- spread but nowhere abundant. Koeleria cristata is abundant only at high elevations in the grassy forests. Throughout the most extensive grass areas of Chihuahua and Durango the species of Bouteloua greatly dominate over the rep- resentatives of other genera. The commonest of these is B. graci- lis, which is estimated to form at least 80 per cent of the cover in approximately 60 per cent of the grassland area. In the southern extension of the grassland, in Zacatecas and Jalisco, Bouteloua hirsuta, B. radicosa, Hilaria cenchroides or Sporobolus trichodes alternate or associate as the most common species. Throughout the grassland area large coarse grasses occur spo- 194 MADRONO [ Vol. 6 radically or in isolated colonies, including Andropogon saccharoides, A. barbinodis, Stipa eminens, S. clandestina, Sporobolus airoides, Elyonurus tripsacoides and Trichloris mendocina. Relatively moist areas are heavily carpeted by Buchloé dactyloides. Dry localities with shallow soil at low elevations are thickly covered with Triodia pulchella or well defined colonies of Scleropogon brevi- folius. In general, however, the grasslands are as monotonous in composition as they are in their physiognomy. The following list includes the dominant and frequently re- curring grasses of the grassland areas of northern Mexico. The names are approximately in the order of abundance. Bouteloua gracilis Andropogon saccharoides Bouteloua curtipendula Aristida ternipes Bouteloua chondrosioides Eragrostis lugens Aristida divaricata Eragrostis intermedia Eragrostis mexicana Hilaria cenchroides Bouteloua radicosa Muhlenbergia monticola Triodia pilosa Stipa editorum Stipa eminens Triodia mutica Eragrostis diffusa Scleropogon brevifolius Lycurus phleoides Sporobolus trichodes Buchloé dactyloides Pappophorum Wrightii Triodia grandiflora Setaria macrostachya Grasses which are less abundant over the entire area or only locally common are the following. _ Andropogon perforatus Muhlenbergia rigida Aristida adscensionis Panicum Hallii Aristida glauca Panicum obtusum Aristida hamulosa Setaria geniculata Bouteloua eriopoda Setaria Grisebachii Bouteloua filiformis Sporobolus airoides Bouteloua hirsuta Sporobolus Poiretii Bouteloua Rothrockii Stipa clandestina Elyonurus tripsacoides Stipa tenuissima Eragrostis limbata Trichloris mendocina Leptochloa dubia Triodia grandiflora Muhlenbergia polycaulis Triodia pulchella TRANSITION FROM GRASSLAND TO ENCINAL At the same altitudes occupied by grassland there also occur extensive stands of encinal (evergreen oak woodland) or open — stands of juniper or pinyon. These are almost invariably con- fined to hills and abrupt slopes or to rocky ground with shallow soil, while grassland occupies level or gently sloping areas with soil from 15 to 50 centimeters or more in depth. In open stands of woodland the floor supports a light cover of grasses which 1942 | SHREVE: GRASSLAND IN NORTHERN MEXICO 195 differs little in appearance and composition from pure grassland. Where the soil is rocky and irregular in depth, and the trees are abundant, a rich variety of herbaceous perennials forms more of the cover than do the grasses. In central Chihuahua the lower edge of the encinal is often encountered at 1500 meters in very open stands of Quercus chi- huahuensis, Q. santaclarensis or Q. Emoryi. In southern Durango encinal fails to find extensive areas of suitable conditions below 2150 meters, and in several districts grassland extends up to 2500 meters. In both states the texture and depth of soil appear to be the deciding conditions for the dominance of grasses or the ap- pearance of trees. The critical season for trees is the dry period extending from February to May. Nothing is known about the comparative soil moisture of the deep soils and the rocky terrain in this region. If conditions are analogous to those investigated in Arizona the pockets and layers of the rocky soil have a higher moisture content in dry periods than the deep uniform soil has. All evergreen oaks defoliate in the early spring and simultane- ously form a complete new crop of leaves. In Chihuahua in 1937, after a dry winter and delayed summer rains, the oaks were nearly leafless in July and new leaves first began to appear early in August. Even in their more favorable habitat the oaks must often be thus brought near the margin of their drought resis- tance. At such times they might not be able to survive on deep level soil except near streamways. In northern Coahuila the upper edge of the grassland com- monly merges into shrubbery from 1 to 2 meters in height. This vegetation forms a closed cover at slightly higher elevations and on north slopes. It so closely resembles the Pacific coast chap- arral in its life forms, social aspects, and generic composition that it may well be designated by the same name. The Mexican chaparral has been described by Muller (3) as manifested on the lower western slopes of Cerro Potosi, in Nuevo Leon. The lower, open edge of the chaparral is dominated in Coahuila by shrubby oaks, notably Quercus invaginata, Q. cordifolia, Q. Pringlei, Q. hy- poxantha, and Q. intricata, or else by small individuals of arbores- cent species. Commonly associated with the oaks, or locally out- numbering them, are Cowania plicata, Arctostaphylos pungens, Microrhamnus ericoides, Ceanothus lanuginosus, Mimosa biuncifera, Amelanchier denticulata, Rhus microphylla, Berberis trifoliata and Cercocarpus mojadensis. The prevailing limestone of northern Coahuila is not favorable to the attainment of large size by oaks and pinyons. On steep and moderate slopes the chaparral forms heavy stands and grasses are nearly absent. On level ground and gentle slopes the shrubbery is open and there is a ground cover of grasses and herbs. These circumstances bring about consider- able variation in the altitude at which grassland merges into chap- _arral, and at which chaparral becomes dominant. The changes 196 MADRONO [ Vol. 6 most commonly take place between elevations of 1500 to 1800 meters, Cactus-Acacis-GRASSLAND Along its southwestern margin the Chihuahuan Desert is bounded by a distinctive type of vegetation which lies between the desert and grassland. This is essentially a thin cover of short grass with a continuous open stand of small trees and tall platy- opuntias. The striking physiognomy of this vegetation might suggest that it be designated by the loosely used term “savanna,” which is often misapplied to any association of grasses and trees. True savanna is characterized by large harsh-leaved grasses of a type which rarely forms closed communities in temperate North America. The characteristic tree is Acacia tortuosa, which usually has a broad flat crown, and the only other tree is a much less fre- quent undescribed species of Prosopis. The platyopuntias are Opuntia streptacantha and O. durangensis, which are erect, have broad joints 30 to 40 centimeters long, and reach a height of 3 to 6 meters. The cacti usually outnumber the trees but are widely spaced or in open groups. Shrubs are very uncommon and nearly limited to Celtis pallida and Acacia paucispina. Semi-succulents are sparingly represented by Yucca carnerosana, which reaches a height of 5 to 12 meters. Cacti other than those mentioned are very uncommon, the one most frequently seen being the ubiqui- tous Opuntia imbricata. The representation of grasses corre- sponds closely with that found in the open grassland. Herbaceous perennials are more abundant in species and individuals than they are in the grassland. Many perennial herbs and several grasses grow in the cactus-acacia-grassland which are not found in the grassland. These are largely confined to the heavy shade of the tall opuntias, whereas the light shade of Acacia, which executes considerable movement during the day, is occupied by the prevailing sod. It seems scarcely allowable to regard the cactus-acacia-grass- land as a transition between desert and grassland. In physiog- nomy it carries no suggestion of either. It has most of the typical grasses of the grassland but its flora includes none of the charac- teristic plants of the desert except the occasional trees of Prosopis. Also, the three characteristic large plants are found neither in desert nor grassland. The lower edge of the cactus-acacia-grassland is reached at about 1800 meters and its upper edge at 2000 meters, or excep- tionally as high as 2200 meters. It is found only on plains, bajadas and gently rolling surfaces with soil at least 1 to 2 meters deep. The commonest type of soil is a ruddy brown clay of vol- canic origin, containing from 5 to 20 per cent of well worn rock particles from 0.5 to 3 centimeters in diameter. In two localities typical stands were seen on granitic loam (near Sain Alto and near Pinos, in Zacatecas). 1942 | SHREVE: GRASSLAND IN NORTHERN MEXICO 197 This type of vegetation has been observed only in southern Durango, Zacatecas, northern Jalisco and southern San Luis Po- tosi. It has been studied between Yerbanis and Villa Madero, Durango, between Victoria, Durango, and Rio Grande, Zacatecas. between Concepcion del Oro and Santa Maria Majion, and near Pinos, in Zacatecas, and in the District of Lagos, Jalisco. The principal variations from the typical conditions described are east of Santa Maria Mafion, where the two opuntias form an unusually close stand, and near Arriega, San Luis Potosi, where the opuntias are dense but not more than 2 meters high, and the sod of grasses very open. There are no available rainfall records within the range of this vegetation but the annual total can be interpolated as proba- bly between 350 and 425 millimeters. The most abundant herbaceous perennials of the cactus- acacia-grassland are the following. Dichondra argentea Stevia salicifolia Evolvulus alsinoides Commelina scabra Alternanthera repens Oxalis albicans Sanvitalia ocymoides Zornia diphylla Plantago mexicana Cassia crotalarioides Guilleminea densa Tagetes lucida Spergularia mexicana Acalypha neomexicana Ipomoea costellata Dyschoriste decumbens Drymaria arenarioides Verbena teucriifolia Cyperus seslerioides Phaseolus heterophyllus Gaura coccinea Polygala compacta Tetraclea Coulteri Sisyrinchium tenuifolium Many of these plants are familiar components of the grass- land as far as northern Chihuahua and western Texas, and it is noteworthy that the list includes no plants of southern range which are here near their northern limits. The grassland and its variants are here near their southern limit, and also many of their characteristic plants reach the southern end of their ranges in Jalisco. Tue Hivarra Lianos Nearly every map of Mexico indicates that the center of the northern plateau is occupied by a single extensive undrained basin, styled the ““Bolson de Mapimi.” As a matter of fact the area includes a large number of independent undrained basins, or bolsons, varying from a few hundred to several thousand hectares in size. Along the eastern and western edges of the plateau there are an undetermined number of bolsons which re- ceive copious drainage waters from the slopes of the bordering mountains. The centers of these bolsons are occupied by alka- line flats and seasonal lakes. Other bolsons in the center of the plateau receive limited drainage from their surrounding hills and bajadas and are without a central playa, or lake bed. In these 198 MADRONO -[Vol.6. the floor of the bolson is nearly level and the soil is deep and fine in texture. In such situations are found the llanos, or nearly pure Hilaria grassland areas, which collectively cover a large ex- panse in northern Chihuahua. The largest are the “Llano de los Gigantes” and the “Llano de los Christianos,’ which are the northernmost ones that have been observed. The southernmost one that has been noted is north of Naica, in the district of Meoqui, Chihuahua. The elevation ranges from 1100 to 13800 meters. As previously intimated the llanos are not to be regarded as part of the climatic grassland formation but rather as a desert association controlled by soil conditions. Surrounding and over- looking the llanos are invariably long outwash slopes covered with typical desert of Larrea, Flourensia, Acacia vernicosa and many cacti. Smaller llanos of the same character occur under very similar conditions in southern Arizona and northern Sonora, down to elevations less than 300 meters. From 80 to 90 per cent of the cover in the llanos is formed by Hilaria mutica, growing continuously or in close-set tussocks. On the higher llanos there is often a small percentage of Bouteloua gracilis and B. eriopoda. ‘There are a few very common non-gra- mineous associates, including Florestina tripteris, Viguiera phenax and Xanthocephalum gymnospermoides. In addition to the regions that have been briefly described in this paper there are at least two other types of vegetation in Mexico in which grasses play a prominent part. One of these is the arid bushland of the low plains of Nuevo Leon and northern Tamaulipas, in which the open shrubbery may or may not be carpeted with annual and perennial grasses. The other is the coastal belt of Vera Cruz, with its tall coarse grass areas merging into marshes or into savanna. Neither of these vegetations has received the ecological investigation which they merit. They have little in common with the grassy areas of the highlands, either in ecological features or in floristic composition. The eco- logical importance of the grass family in northern Mexico may be surmised from its wealth, for in the ten states north of the Tropic of Cancer nearly five hundred species are known. Carnegie Institution of Washington, Desert Investigations, Tucson, Arizona. LITERATURE CITED 1. Crements, F. E. Plant Indicators. Carnegie Inst. Wash. Publ. no. 290. 1920. 2. Gates, F. C. Grasses in Kansas. Rept. Kans. State Bd. Agric. 55: 1-349. 1937. 3. Mutter, C. H. Relations of the Vegetation and Climatic Types in Nuevo Leon, Mexico. Amer. Midl. Nat. 21: 687-729. 1939. 4. Sureve, Forrest. Observations on the Vegetation of Chihuahua. Madrono 5: 1-13. 1939. 5. Weaver, J. E. and T. J. Frrzparricx. The Prairie. Ecol. Monogr. 4: 109- 295. 1934. 1942] CRONEMILLER: CHAPARRAL 199 CHAPARRAL F. P. CRONEMILLER Chaparral is a term commonly used in California for the dense brushfields of the Upper Sonoran life zone. This cover type is similar to the macchie and garique of the Mediterranean region and is a product of the set of climatic conditions peculiar to these areas. Its shrubby components have been termed quite accu- rately broad sclerophyll vegetation. The purpose of this paper is to give a chronicle of the derivation of the term “chaparral” and to encourage its use and adoption by the technician. Chaparral evolved from chabarra, the Basque word for a scrub oak of the Pyrenees. The Spaniard adapted it to “‘a dwarf ever- green oak” and spelled it chaparro. He did not develop the word “chaparral,” however, as he used the term “‘garique’”’ for the cover type composed of this and species of similar growth. On his arrival in the New World he was faced with a tremendous job of inventing place names. The vast number of saints furnished an abundance of names for important places, while descriptive terms were given those of secondary importance. The convenient suf- fix -al, meaning “place of,’ naturally was often used. Pinal, alisal, sausal, designated pine groves, sycamore flats, and willow thickets. For the cover types of dense evergreen scrub oaks, chaparral was invented. Quickly the term came to be applied to similar cover types, and this is its usage today. Colloquially it has been applied to individual species such as one of the Acacia species in Mexico and to Ceanothus cuneatus in California. Gen- erally the proper meaning has adhered: a place (cover type) of evergreen shrubs or dwarf trees. In addition to chaparral the Spanish Californian used a term, chamiso (or chamisal), to designate open brush areas composed of small shrubs. The original term, chamiza, meant simply “kindl- ing wood.” It is not certain that he ever got entirely away from this connotation, but the vaquero used it in opposition to chapar- ral. To him chaparral was that kind of brush one could not ride a horse through; through chamiso or chamisal, one could. Cali- fornians have anglicized the word to “chamise”’ and applied it to but a single species, Adenostoma fasiculatum. In New Mexico, chamiso refers to Atriplex canescens, and in Mexico to other spe- cies and types. It would appear that the term ‘‘chamise’’ is hardly tenable, although widely used by the layman and some technicians. Chaparral is without doubt a needed word; it is in general use, and is recognized by Webster and other lexicographers. The technician should not hesitate to use it. United States Forest Service, San Francisco, California, February 25, 1942 200 MADRONO [ Vol. 6 NOTES ON POLEMONIACEAE Hersert L. Mason GiLia INconspicua Dougl. ex Hook. Bot. Mag. pl. 2883. 1829, nomen confusum. Ipomopsis inconspicua Smith, Exot. Bot. pl. 14. 1805. Cantua parviflora Pursh, Fl. Am. Sept. 2: 780. 1814. Gilia parviflora Spreng. Syst. Veg. 1: 626. 1825. In the year 1805 J. E. Smith published in his ‘““Exotic Botany” the name Ipomopsis inconspicua based upon plants cultivated in England. The plants were “raised in 1793 by Mr. Thos. Hoy, F. L. S. at Sion House, from seed brought, if I mistake not, from America. Mr. Sowerby sketched it in November of that year.” Pursh, in 1814, transferred J. inconspicua Smith to the genus Can- tua, renaming it C. parviflora. However, his doubt as to its origin in North America was expressed in the following words, “I insert this plant on the authority of Exotic Botany; but at the same time I doubt very much of its being a native of North America, and more strongly suspect it to come from Mexico.” Sprengel trans- ferred the species to Gilia in 1825 making the combination G. parviflora (Pursh) Spreng., reporting it from North America and citing Cantua parviflora Pursh and Ipomopsis inconspicua Smith as synonyms. Hooker in 1829 published a manuscript name of Douglas whereby Douglas referred Ipomopsis inconspicua Smith to the genus Gilia as G. inconspicua (Smith) Douglas. AIl three of the above mentioned names were cited as synonyms. Hooker’s remarks are enlightening, “Of the authors who have hitherto described this plant, Smith alone has seen specimens which were cultivated at Sion House, in 1793, from seed which he supposed to be brought from some part of America. Pursh imagined it to be a native of America: but it was reserved for the indefatigable Mr. Douglas to determine its exact locality. He discovered it in the woodless tracts, or sandy barrens on the Southern branches of the river Columbia, on the Northwest coast of America, growing under the shade of Purshia (Tigarea. Ph.) tridentata and some species of Artemisia.” | The descriptions by Smith and by Hooker are each accom- panied by illustrations in color. It is obvious that the plants illustrated in each case are the same species. It is also obvious that the plants illustrated are not of a type known as yet from the interior of the Columbia River region in northwestern America. Nor does Hooker’s illustration agree with the Douglas specimen in the Hooker Herbarium labeled in Hooker’s handwriting “Gilia inconspicua Douglas.’ The description given by Hooker more nearly fits the plant illustrated. It would seem that the descrip- tion was drawn from fresh material as was the illustration, and that an error was made in the source of the seed that gave rise to the plant. The description and the illustration of Hooker fit Ipomopsis inconspicua Smith, whatever that species may be. 201 MASON: NOTES ON POLEMONIACEAE 1942 | Figs. a-c, Linan- Puiate 26. Linantuus WIGGINSII AND GiILIA CLOKEYI. thus Wigginsii; figs. d-f, Gilia Clokeyi. 202 MADRONO [Vol.6 Later botanists have largely gauged their concept of Giulia inconspicua on the specimen of Douglas rather than on the de- scription and illustration of either Smith or Hooker. As a result, the name G. inconspicua Dougl. ex Hook. is erroneously applied to the plant of the arid interior of the Great Basin of North America. Smith’s plant was grown in England and described before any botanist had traversed that portion of the interior of America. The next oldest available name for this group of plants is Gilia sinuata Dougl. ex Benth. based on specimens collected by Douglas from near the confluence of the Okanogan River with the Co- lumbia. This entity has been regarded by different authors as distinct either specifically or varietally from G. inconspicua Doug]. ex Hook. or as completely synonymous with that species. At any rate the plants represented by these names constitute an exceedingly variable genetic complex, and an_ eco-genetic analysis must be made before a clear understanding of the taxon- omy of the group is possible. The identity of the plants represented by the name Ipomopsis inconspicua Smith is not easy to determine. It seems probable that if they were derived from western North America they must have come from the coastal region. In North America plants of the group now passing under the name Gilia multicaulis Benth. are an excellent match for the illustrations of Smith and Hooker and could well have been collected at Monterey, San Francisco or Bodega, points visited by most of the early exploring expedi- tions. In both illustrations mentioned above, only portions of the plants are represented and these are not sufficiently complete to make identification certain. On the other hand, Giulia laciniata Benth. and G. valdiviensis Griseb. of South America are also close. Because of the permanent uncertainty as to the identity of Smith’s original material the name Giulia inconspicua should be designated as a nomen confusum. Gilia Clokeyi sp. nov. Herba annua erecta, 5-30 cm. alta, plerumque pseudoscapo evidente, 15-25 mm. alto; caules infra plerumque simplices in inflorescentibus ramosi; cotyledones lineari-spatulati, 12-16 mm. longi, 2—2.5 mm. lati, in petiolos gra- ciles attenuati; folia in ambitu anguste oblonga, alterna, basi non rosulata in inflorescentibus abrupte reducta bracteata, pinnate vel bipinnate lobata, lobis aliquanto remotis, foliorum inferiorum segmentis ultimis ovatis, superiorum lanceolatis, glabra vel rarius sparse floccosa tandem glabrescentia, textura delicata; pedicelli atque ramuli florales ultimi glandulis paucis nigris capitellatis ; calyx infra sinus membranaceus, membranis aliquanto distensis; corolla infundibuliformis, 6-9 mm. longa, pallido-coerulea vel fere alba, faucibus luteis, 1-1.5 mm. longis, tubo luteo, intus glabro, 3 mm. longo, lobis rhomboideis, circa 3 mm. longis; stamina in sinubus corollae affixa, antheris subsessilibus, circa 0.6 mm. longis, albis vel pallido-coeruleis, polline antheris similiter tincto ; 1942 | MASON: NOTES ON POLEMONIACEAE 203 pistilum 3-9 mm. altum, stylo apice diviso, ramis 0.5-1 mm. longis; capsula late ovoidea; semina matura non visa. Erect annual 5-30 em. high, usually with an evident pseudo- scape 15-25 mm. high; stems usually simple below, openly branched in the inflorescence; cotyledons linear spatulate, 12-16 mm. long, 2—2.5 mm. wide, the lower one-third or one-half nar- rowed to a slender petiole; leaves narrowly oblong in outline, alternate, not in a basal rosette, becoming abruptly reduced and bracteate in the inflorescence, pinnately, or sometimes bipin- nately lobed, the lobes somewhat remote, the ultimate segments of the lower leaves ovate, those of the upper lanceolate, glabrous, or more rarely sparingly floccose and becoming glabrate, texture thin and delicate; pedicels and ultimate branches of the inflores- cence with a few black tack-shaped glands; calyx membranous below the sinuses, the membrane somewhat distended; corolla funnelform, 6-9 mm. long, pale blue to almost white, lobes rhom- boid about 3 mm. long, throat and tube yellow, throat 1—-1.5 mm. long, the tube 3 mm. long, glabrous within; stamens inserted in the sinuses of the corolla lobes, anthers subsessile about 0.6 mm. long, white to pale blue, pollen of the same color; pistil 3-9 mm. high, style divided at the tip, the three branches 0.5—-1 mm. long; capsule broadly ovoid, mature seed not seen. Type. Larrea belt, altitude 1200 meters, north base of lime- stone ledge, Red Rocks, Charleston Mountains, Clark County, Nevada, March 31, 1940, 1. W. Clokey 8599 (Clokey Herbarium at the Herbarium of the University of California). Other collec- tions. Talus slopes above Wilson’s Ranch, Charleston Moun- tains, Clark County, Nevada, May 3, 1939, Bassett Maguire 16620. Gilia Clokeyt has been passing with a complex group of plants as Gilia inconspicua (Smith) Dougl., a name which, as has been indicated above, is of uncertain identity. The newly recognized entity may be distinguished readily by the lack of a rosette of congested leaves at the base of the stem, also by the essentially glabrous herbage, remotely lobed leaf blades and long narrow cotyledons. Linanthus Wigginsii sp. nov. Herba annua, 3-12 cm. alta simplex vel pauciramosa; internodia 5-50 mm. longa, pilis densis sparsisve, debilibus brevibus contortis, tandem glabrescentibus ; cotyledones lineari-spatulati, 3-5 mm. longi; folia infra saltem opposita aliquando in inflorescentibus subopposita vel alterna, palmatim in 3 (2—5) segmentis linearibus 5—20 mm. longis incisa, vel inferiora et superiora quandoque simplicia; flores non con- gesti, solitarii vel gemini in axillis foliorum superiorum; pedicelli graciles inaequales, 3-20 mm. longi; calyx 4-5 mm. longus, tubo 1 mm. longo, lobis linearibus 4 mm. longis, marginibus inferiori- bus ad sinus calloso-incrassatis infra sinus membranis parvis; corolla late infundibuliformis alba, longitudine rare 10 mm. exce- dens, tubo calycem aequante, vel subaequante, extus pubescente intus glabro, faucibus abrupte expansis 1.5—-2 mm. longis, lobis 204 MADRONO [ Vol. 6 obovatis 5 mm. longis, 2-3.5 mm. latis; stamina faucium in parte inferiora afhxa, filamentis filiformibus glabris, 1.5 mm. longis, antheris luteis, 0.5 mm. longis e faucibus exsertis; ovarium 1 mm. altum, stylo 3-5 mm. longo ad medium in tres divisionibus lineari- bus inciso; fructus non visus; semina non visa. Slender annual 3—12 cm. high, simple or with a few branches; internodes 5-50 mm. long with dense or sparse, weak short twisted hairs, becoming glabrate; cotyledons linear-spatulate, 3— 5 mm. long; leaves opposite at least below, sometimes suboppo- site or alternate in the inflorescence, palmately cleft into 2 to 5 (normally 3) linear divisions, 5-20 mm. long, occasionally the lower and sometimes the upper simple; flowers solitary or in pairs in the upper leaf axils on slender unequal pedicels, showing little tendency toward congestion; pedicels 3-20 mm. long; calyx 4—5 mm. long, the tube 1 mm. long, the linear lobes 4 mm. long, lower margins of the lobes callous thickened toward and in the sinus, the sinus with a small membrane; corolla broadly funnel- form, white, rarely exceeding 10 mm. in length, the tube equal or subequal to the calyx, pubescent externally, glabrous within; throat abruptly expanded, 1.5—2 mm. long, lobes obovate, 5 mm. long, 2—3.5 mm. wide; stamens inserted on the lower half of the throat, filaments threadlike, glabrous, 1.5 mm. long, anthers 0.5 mm. long, yellow, exserted from the throat; ovary 1 mm. high, style 3-5 mm. long, cleft to almost one-half its length into three linear divisions; fruit and seeds not seen. Type. Southern end of Santa Maria plains, Baja California, Mexico, February 5, 1935, Ira L. Wiggins 7557 (Dudley Herba- rium, Stanford University, 263704; isotype, Herbarium of the University of California, 659206). Linanthus Wigginsi is closely related to L. Nuttalli var. flori- bundus (Greene) McMinn from which it differs, however, in a number of significant characters. These differences may be best expressed by a key: Plants perennial from a woody base, 1-5 dm. high; internodes stout; leaves 5 to 8 cleft, rarely subopposite; many-flow- ered, flowers sessile or short-pedicelled; calyx pubescent; corolla 12-20 mm. long, lobes 7-10 mm. long, spatulate to ovate; stamens included, filaments 1 mm. long, anthers Ob mm: One lke hase ee ee Dy ge ML L. Nuttallii var. floribundus Plants slender annuals, 3-12 cm. high; internodes slender; leaves 2 to 5 cleft, usually subopposite in the inflores- cence; few-flowered; flowers short-pedicelled to long- pedicelled; calyx glabrous; corolla 8-10 mm. long, lobes 5 mm. long, obovate; stamens exserted, filaments 1.5 mm. long, anthers 70:5, mm plong. 3... ee eee L. Wiggins Linanthus Wigginsiit seems definitely to establish a place for L. Nuttallii (Gray) Greene and its allies in the genus Linanthus rather than in the genus Leptodactylon where it has been placed by some authors. Its position in this latter genus has been main- tained solely upon its perennial woody habit. The relationship 1942 | MASON: NOTES ON POLEMONIACEAE 205 between L. Nuttall and L. Wigginsit is so close as to make generic separation impossible. Similarity between the two species may be noted in a tendency of the flowers to occur in pairs in the leaf axils, in the unequal pedicels of the paired flowers, in the shape, color and pubescence of the corolla, and in the very similar de- tails of the calyx such as the thickening of the margins of the lobe and the nature of the membrane in the sinuses. Department of Botany, University of California, Berkeley, February, 1942. EBUROPHYTON HELLER: A VALID GENUS OF THE ORCHIDACEAE Lovis O. WILLIAMS EBUROPHYTON AUSTINAE (Gray) Heller, Muhlenbergia 1: 49. 1904. Chloraea Austinae Gray, Proc. Am. Acad. 12: 88. 1876. Cephalanthera oregana Reichb. f., Linnaea 41: 538. 1876. Cephal- anthera Austinae Heller, Cat. N. Am. Pl. ed. 2, p. 4. 1900. Sera- pias Austinae A. A. Eaton, Proc. Biol. Soc. Wash. 21: 66. 1908. In 1876 Asa Gray described an orchid from California which he called Chloraea Austinae. Chloraea is a genus of orchids occur- ring in South America from the Falkland Islands north to Peru, with its greatest concentration of species in the Andes of Chile. Chloraea occurs mainly in open habitats and quite often in very hard, sterile soil. So far as I know no member of the genus is saprophytic. In the same year, 1876, H. G. Reichenbach de- scribed the same species, from a specimen collected by Nuttall, under the name of Cephalanthera oregana. Cephalanthera is a genus primarily of Europe and adjacent regions but one in which the species are not saprophytic. The third generic name was that applied by Heller in 1904, Eburophyton, a name designed to con- tain the single species in question. In 1908 A. A. Eaton placed the species in still a fourth genus as Serapias Austinae. Although the species has no special character to recommend its being placed in this genus it is here that it has been treated most often. Ames in “Enumeration of the Orchids of the United States and Canada” (1924), the most authoritative work yet published on the region covered, placed the species here. Eburophyton Austinae is at once excluded from Serapias by its anther which is attached by a slender filament and is not solidly attached as in Serapias. From Chloraea it is distinguished by its saprophytic habit, by the lip being divided into an epichile and hypochile, with the hypochile gibbous at the base. From Cephal- anthera the distinction is more difficult but the scarious nature of the leaves, saprophytic habit and geographical distribution would seem to indicate a separate genus. Botanical Museum, Harvard University, Cambridge, Massachusetts, 206 MADRONO [ Vol. 6 REVIEW Standardized Plant Names. By Haruan P. Kexsey anp WiLtiam A. Dayton, prepared for the American Joint Committee on Horticultural Nomenclature. Second edition. Pp. xvi+675. J. Horace McFarland Company, Harrisburg, Pennsylvania. 1942. $10.50. As announced in the preface to the second edition “the pur- pose of ‘Standardized Plant Names’ is to bring intelligent order out of the chaos in names of plants and plant products existing the world over.” We learn from the preface that the authors believe that the confusion in plant nomenclature is due to mistakes in identification and labeling and to disagreement in opinion, prac- tice and judgment among botanists; we learn also that the gen- eral basis of the work of the committee appointed to bring “intel- ligent order’ out of this chaos is “to agree arbitrarily upon some one name for each plant, by which name it can be designated for a definite term of years’; we learn to our chagrin that “a right label on the wrong plant or plant product may cause even more loss or disaster than the wrong label on the right plant or 1 pia product” (page viii). Throughout the preface there are presented certain rules whereby, as implied in the title, the names of plants are to be standardized. Although these rules are not definitely organized into a precise code of nomenclature, with a little patience one can collect and arrange them. The authors are aware of the rules adopted by the International Conference in London in 1930 but some of the rules promulgated by this conference from the stand- point of the present authors are inadmissable. They therefore accept what they like of these rules and reject what they do not like. Botanists have had similar experiences in bringing “‘intelli- gent order” out of chaos and were hampered by a similar group working from a nationalistic point of view, which if we mistake not, stemmed from the same habitat as this effort. It seems to the reviewer that since plants and plant products are a consider- able item in international trade, their nomenclature is at once placed upon an international basis. Therefore, any attempt to standardize these names should result from international coopera- tion. This it would seem, is the intelligent basis for procedure and by not accepting it the authors violate the first point of their purpose. It should be pointed out that the authors do not state specific- ally that their object is to develop a code of nomenclature but rather to “agree arbitrarily upon some one name for each plant.” However, to standardize names implies that you have standards whereby this is to be accomplished. Moreover, if the rules men- tioned in the preface are not to be used as a guide, why are they published in conjunction with plant names under a title that implies a code of standardization? The reader must assume that 1942 | REVIEW 207 the authors intended that these rules should serve as a code or basis for the standardization of the nomenclature developed. Several of the rules apply only to scientific names which are under the jurisdiction of the International Committee on Botani- cal Nomenclature and hence are not the concern of this or any other committee on horticultural names. Since horticultural interests are represented on the International Committee any use of scientific names at variance with the International Rules is un- justified and serves only to create confusion. Although it is always in order to recommend new rules to clarify plant nomen- clature it is never in order for a small group to reject rules adopted through international cooperation. These rules cannot correct mistakes in identification nor can they prevent, disagreement in opinion, practice and judgment among botanists; these being the principal causes of confusion in the opinion of the authors. Rules of nomenclature should be aimed solely at matters which cause confusion in the names of plants. The consolidation of compound names adopted by the committee serves no useful purpose and more often than not, causes a name to violate the international code rule involving pronunciation and excessively long words. For example, try to pronounce “Gianthyssop”, “‘Pussyears”’, “Sevenyearapple’”’, “Globeamaranth’’, and “Holyghostflower.” Consider ““Browneye Babyblue-eyes Nemophila.”” We are not joking, this is the name the authors propose for Nemophila atomaria var. discoidalis (page 397). This rule as construed and practiced by these authors in many instances places the problem of the naming of plants upon a plane of ridiculousness never before achieved. We again call your attention to “Gianthyssop.”’ To what language does it belong? It certainly is not English. These and hundreds of other proposed common names are a combination of good English words that by compounding have been translated into an atrocious form of pigeon English. When you write it, ““Browneye Babyblue-eyes Nemophila” gets under the three-word limit by its beautiful eye- lashes, but don’t try to pronounce it as three words. After ac- cepting certain rules the authors do not hesitate to practice methods of circumventing them. Although opposing the capital- ization of specific and varietal names in the scientific nomencla- ture of plants, it is of interest to note that for common names no such concern is felt. Every word of a name is capitalized and in compounding words the. capital letter is retained. Thus we find orthographical monstrosities such as “‘RockyMountain’’, “SanDiego”, “EastIndies” and “‘SantaRosalIsland.”’ With respect to the application of the principle of priority to horticultural names certain difficulties need to be considered. Priority is a sound principle upon which to establish a stable nomenclature. It demands, however, that every name be docu- mented with authority so that its position in the priority sequence 208 MADRONO [Vol. 6 can be established. Some of the names in this work that have been presented by societies such as the American Carnation Soci- ety and the Chrysanthemum Society are documented at least to originator but even in these the date is sometimes omitted. Unless there is maintained a specimen voucher or an adequate illustration we believe that it will be impossible at some future date to check against errors in identification and the misuse of names. The majority of the names used in the work give no evidence of sup- plying any basis whereby priority may be established if one should choose to apply it. To put horticultural names in a posi- tion to be standardized would require a major revolution in the practices of horticulturists with respect to giving and document- ing the names of their plants. A system could be worked out by a qualified international committee for documenting names already in use and rules might be set up to which, in the future, authors would be required to conform. We are not concerned in nomenclature with the allocation of credit or honor for discovery or horticultural selection, we are concerned with responsibility for applying the correct name to a plant. There are many more points that might be discussed in con- nection with this work and there is much that deserves credit. The chief criticism to be leveled against it is that the committee conceived of its problem as being one of giving plants names rather than one of proposing methods whereby the horticulturist could adjust his practices in naming plants so that standardiza- tion would be possible and nomenclature could be stabilized in as simple and dignified a manner as possible-—HeErsert L. Mason, Department of Botany, University of California, Berkeley, Cali- fornia. PROCEEDINGS OF THE CALIFORNIA BOTANICAL SOCIETY January 22,1942. Meeting, 103 Wheeler Hall, University of California, Berkeley, at 7:45 P.M. The retiring president, Pro- fessor E. B. Babcock, opened the meeting. Dr. Mildred E. Mathias requested, with regret, that the nominating committee withdraw her name from the list of nominations for 1942. The name of C. R. Quick was substituted by due process for nomina- tion to the secretaryship. The following new officers were in- stalled: Dr. Alva R. Davis, President; Dr. Palmer Stockwell, First Vice-President; Dr. Reed C. Rollins, Second Vice-President; Mr. Clarence R. Quick, Secretary; Dr. William M. Heisey, Treasurer. Dr. Davis, incoming president, introduced Dr. Reed C. Rollins, Instructor in Botany, Stanford University, who presented an interesting illustrated lecture on “The geographical distribution, speciation, and natural variation in Arabis and other Cruciferae.” 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 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 ae NUMBER 7 MADRONO A WEST AMERICAN JOURNAL OF BOTANY JUL 2 41942 £ 4 é 47. \) ONAL MUSE Contents THE SIGNIFICANCE OF CERTAIN Ptant Names, Carl Sumner Knopf Notes on THE Fora oF THE CHARLESTON Mountains, CLARK CouNTY, Nevapa. IV. Asrracarus, fra W. Clokey Tue Tyre Locariry or Potystichum Lemmoni UNberwoop, Harold St. John Far WESTERN NOVELTIES IN SALIX, Carleton R. Ball Notes anp News PROCEEDINGS OF THE CALIFORNIA BOTANICAL SOCIETY Published at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania July, 1942 MADRONO A WEST AMERICAN JOURNAL OF BOTANY Board of Editors Hersert L. Mason, University of California, Berkeley, Chairman. LeRoy Asrams, Stanford University, California. Epcar Anperson, Missouri Botanical Garden, St. Louis. Lyman Benson, University of Arizona, Tucson. Hersert F’. Copetann, Sacramento Junior College, Sacramento, California. Ivan M. Jonnston, Arnold Arboretum, Jamaica Plain, Massachusetts. Mixprep E. Maruias, University of California, Berkeley. Bassett Macuime, Utah State Agricultural College, Logan. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board—EtTuHet Crum Department of Botany, University of California, Berkeley Business Manager—Win.1am Hiesey 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. 1942] KNOPF: SIGNIFICANCE OF PLANT NAMES 209 THE SIGNIFICANCE OF CERTAIN PLANT NAMES Cart SUMNER KwNopFr Botanical terminology is filled with oddities. An ancient Roman would probably find much amusement in the atrocious Latinesque mongrels, denoting that Smith, Ph.D., found and clas- sified the Something-or-other Smithii. However, in many com- mon and technical designations there is hidden a veritable ro- mance of linguistic adventure, where research leads across seas and sands to natural habitats and original appellations. In giving derivation of English words and common scientific terms, dictionaries often stop with Latin or Greek forms. Occa- sionally, reference will be made to Arabic. Yet many Graeco- Latin words were dialectic modifications of borrowed Near Eastern terms which were names of articles of trade peddled by Aramean and Phoenician merchants. The family, Boraginaceae, has generally been identified with the Mediterranean littoral and eastward. Littre (Dictionnaire de la Langue Francaise), speaks of it as a “Plante Sudorifique, originaire d’Afrique et introduite par les Maures en Espagne.” It was long known for cardial and febrifuge properties. Gerarde (1597) noted that borage was used in salads “‘to make the mind glad.” Britain’s famed “cool tankard”’ combined the leaves with wine, water, lemon and sugar. Great healing power was accredited to borage. The roots yielded brown and purple dye. If this plant or some special use of it was introduced into Spain by the Moors, an oriental ancestry of its name would be logical. The scattered variant forms are easily recognizable, as German boretsch, French bourranche, Italian borraggine, Spanish borraja, Latin borego, and in all probability, Greek pourakion. Some have tried to connect the term with Latin burra, “a hairy cloth,’ or French bourre, “animal hair,” since the group is notably hirsute. Among the Arabs the plant is known as barwaq. Boiled with olive oil and vinegar it ‘is a specific for jaundice. The root juice is used for skin eruptions; the juice of the leaves sometimes mixed with food “to cause excitement.” In Zerolo’s “‘Dic- cionario Enciclopedico de la Lengua Castellana,” the Spanish borrachuela is described as causing “cierta perturbacion.” Both Arab and Spaniard note the mild intoxication. Their terms are philologically related, the Arabic q passing over into Spanish ch and 9, borraja and bourrachuela, French bourrache. Linguists long pondered over the origin of the Spanish term borracho, “drunk.” Obviously it follows the same consonantal root pattern, b-r-g/j/q/ch always connected with some kind of exciting, mind-confusing state and a plant juice inducing it. An Maproxo, Vol. 6, pp. 209-240. July 20, 1942. 210 MADRONO [Vol.6 Arabic word, baraq, means “‘to be confused,” “weak,” or “with eyelids immovable.” A noun, built on the same root structure, bargat, means a “fit of confusion” or “blind perplexity.” In Spanish, confusion of judgment is borrachez. Since Arabic q is formed back of the palate, it passes easily into Spanish ch, while the dental t as readily becomes a voiced sibilant, z. Following the same intoxication motif, an Arabic wine basin is called an tbriq, and the Spanish leather wine bottle is a borracha. Again the familiar b-r-g/q/ch of borage is clear. In spring, every vivid patch of fiddleneck (the Amsinckia intermedia of Ore- gon and A. Douglasiana of California) or heliotrope, or forget-me- not (Myosotis) can give the scholarly observer the mild intoxica- tion of adventure—to Merrie England and old borage remedies; to France and her “plante sudorifique”’; to Moorish Spain and convivial borrachos; to Arabia concocting jaundice medicine; to Italy where Roman tongues twisted a foreign term from the. eastern provinces, while in Greece, Hellenic tongues did the same. The mucilaginous, sudorific, emollient, cooling, stimulating, hir- sute borage carries a story in its name. Cotton is equally revealing. The Arab long ago knew qutun. If he affixed the definite article it was al-qutun, which became Spanish algodon and Portuguese algadao. Spanish cotton cloth became coton. The ancient Greeks, still barbarians when the Phoenicians were civilized sophisticates, bought many a novelty from the Near East. The well-made kitunah became the most popular article of Greek clothing, the xiton.1. The name probably came from a Semitic root, k-t-n, “‘to clothe.’ In Assyrian, a word zatanu meant “protect.” Xutenu meant “protection” and, inci- dentally, sounded quite like the Greek wziton and Phoenician kitunah. An Assyro-Babylonian zittu was a “border,” like the ruffles on a garment. Through all the terms runs a basic idea—a garment; woven stuff; protection; clothing; fringe. The common throat sounds for the idea are a glottal stop, a dental, and a dental-nasal— k/q/x/-t-n. It became stabilized in Arabic qutun, and emerged in English as “cotton.” Back of it was Phoenicio-Aramaic kitunah; and back of that, Assyro-Babylonian zatanu and kitunnu. Speaking of cotton (of the family Malvaceae, genus Gos- sypium) reminds one that this family name, Latin malva, garbled into Anglo-Saxon mealwe, has also a distinct Near East connec- tion. When the Greeks called mallow malaze, they left a lin- guistic clue, pointing to some term that included an extra conso- nant, « The Greek malakos, suggested “soft”? and mallow has been used in medicine as an emollient. In the Greek words the extra sound of a guttural wi or kappa (k) is clear. 1X is used to denote the throaty, raspy, guttural, like a half-swallowed k or qg, indicated by Greek wi and Semitic weth. 1942] | CLOKEY: ASTRAGALUS 211 Turning to Near East literature, one finds in the great book of Job (80: 4) the old man’s plaint that everybody laughs at his suffering, and even the “salt-weed” or “mallow” cutters deride him. The Palestinian Negeb or south country is volcanic and saline. Spring torrents bring down mineral salts from the hills. Water holes turn salty and crystals often line the edges. Even the Babylonian texts refer to this salt country. The nomad population adapted its resources to their needs. Every edible or therapeutic plant had to yield its benefit. When Job mentioned the mallow cutters he used the term malluax. The final consonant, zeth, had much the same guttural sound as Greek ai. References to the salt lands in Psalms (107: 34) and Jeremiah (17:6) employed the same root. In Baby- lonia, a malaku was a sailor, one identified with salty waters. In passage from the cradle of civilization to and through the Medi- terranean tongues, the glottal stop, k, or guttural 2, could be easily lost, leaving the softer malva, malba, and mallow. But the presence of the extra consonant in some of the Greek terms pro- vides the clue pointing to oriental habitat and initial use of the mallow. The philological background of botanical nomenclature can not be expected to provide complete implementation for habitat and dispersion studies, but there are possible suggestions. Cer- tainly the linguist can clasp hands with the botanist when he finds basic word patterns such as b-r-g, k-t-n, or m-l-x stretching from Persia to the Pacific. Willamette University, Salem, Oregon, January 12, 1942. NOTES ON THE FLORA OF THE CHARLESTON MOUN- TAINS, CLARK COUNTY, NEVADA. IV. ASTRAGALUS Ira W. CLoKkry For assistance in the study of the Astragali of the Charleston Mountains and for affording me the use of the Pomona College Herbarium, including the Marcus E. Jones Herbarium, I wish to express thanks to Dr. Philip A. Munz. Appreciation is also ex- tended to the curators of the herbaria of the United States National Museum, the New York Botanical Garden and the Uni- versity of California for the loan of type and critical specimens. I also wish to thank Mr. Rupert C. Barneby for information about Nevada Astragali and for the preparation of the accompanying plate. 1 Previous notes in this series have appeared as follows: Madrofio 4: 128- 130. 1937; Bull. So. Calif. Acad. Sci. 37: 1-11. 1938, 38: 1-7. 1939. 212 MADRONO [Vol. 6 Type specimens of species herein described as new are in the Clokey Herbarium now on deposit at the University of California, Berkeley. KEY TO THE SPECIES OF ASTRAGALUS I. Perennials 1. Pods 1-celled A. Pods sessile. Pods leathery. Pods horizontal, 1.5 cm. or more long; low plants, stems decumbent. Pods strigose, tapering at base, narrowly lanceolate-linear, decidedly arcuate, 3- 4.5 cm. long. Leaflets elliptic; pubescence appressed, hairs with median attachment; flowers 2-3+Cm. lON aoe wes ee ee 1. A.amphioxys Leaflets oval to broadly obovate; pubes- cence loose, somewhat tangled, hairs with terminal attachment ........... 2. A. Tidestromi Pods long villous, obliquely ovoid with up- turned tips. Corolla purple. Pubescence of leaflets appressed; calyx tube about 10 mm. long, with nearly white hairs, teeth 2-3 mm. long; pods 1.5-2.5 em. Jong’. 2. ...04....% 3. A. Newberryi Pubescence of leaflets loosely villous; calyx tube 7-8 mm. long, with mostly black hairs, teeth 4-5 mm. long; pods 3 cm. or more long ..... 3a. A. Newberryi var. funereus Corollavcrimsonts...- eee oa, a 4, A. coccineus Pods erect, about 1 cm. long; flowers 7-10 mm. long; stems slender, 3-5 dm. long ... 5. A. humistratus var. sonorae Pods membranous, much inflated, speckled; su- tures equally convex; leaflets lance-linear to LUT Ver? 9 wm Maten of cere oe SR Te Rtomard: yee: 6. A. aequalis B. Pods stipitate. Pods leathery; stipe from very short to nearly as Tone aSsCalynts ses Ae er nA ie eee eee 7. A. Preusit Pods membranous, much inflated, mottled ...... 8. A.artipes 2. Pods completely or incompletely 2-celled A. Pods partially 2-celled; septum narrow. Pods stipitate, somewhat inflated. Stipe 3-5 mm. long. Pods leathery, erect, nearly straight; flowers white with purple tips, about 13 mm. ORG ih ee ee eo eee 9. A.arrectus var. remotus Pods membranous, mottled, strongly arcu- ate; flowers purple, 18-20 mm. long ... 10. A. Beckwithii var. purpureus Stipe very short; pods leathery, filled with pulp swheny ereen2 34.4. fa ll. A. praelongus Pods sessile, leathery, slightly arcuate; flowers purplish;°8—l0\mm. longe:.-.. 2.8 oe4) 2. 3. 12. A. mohavensis 1942 | CLOKEY: ASTRAGALUS B. Pods completely 2-celled or with the septum reach- ing almost to ventral suture, sessile. Pods papery, much inflated. Flowers white with purple tips; caespitose perennials, stems less than 1 dm. long; pods mottled, septum formed by protru- sions from both sutures; alpine or sub- alpine. Pods 2-3 cm. long, acuminate ellipsoid .... Pods about 1 cm. long, oval, with a slender beakei—2rmmpalONn ga os ceys oc was ook Flowers purple; stems erect, more than 3 dm. high; pods about 2 cm. long, rounded ovoid, septum formed by protrusion from dorsal suture only; Larrea or lower PUTO Ta 1S Ca ae ae ee aa eens ey erin a dagen ede N Pods coriaceous, not inflated. Plants 1-4 dm. high. Pods nearly straight. Pods white shaggy-woolly, 2-2.5 cm. long; flowers white with purple tips, 13-15 MATIN HOME oy eee se awe Manes Pods glabrous, 2.5-3 cm. long; flowers purple tinged, 7-8 mm. long ........ Pods strongly arcuate and strongly reticu- late, white strigose; flowers purple, 6-8 TUMTAD ONLY Geese exey de hain Sek cad cage en 2 Bae Plants acaulescent or subacaulescent, less than 1 dm. high. Leaflets 3-7, oblanceolate to obovate; calyx teeth, 3-4 mm. lONE seo hale Soa ag owes Leaflets 5-13, elliptic-oblanceolate; calyx teeth 1-1.5 mm. long .:+.............. 13.4 14. 15. 16. 17. 213 . platytropis . kernensis subsp. charlestonensis . lentiginosus var. Fremonti A. Minthorniae yee eA . bernardinus . hemigyrus calycosus Mancus II. Weak, decumbent annuals; flowers 3-5 mm. long, white or purple; pods sessile Racemes few-flowered; pods papery, linear, 1.5-2 cm. long. Pods 2-celled except towards tip. Keel with a short, rounded porrect beak ........ Keel with an acuminate, porrect beak .......... Pods 1-celled, septum from dorsal suture, if present, AINE TE MUM Cet do ne Seam tea dnrss Savihaew mead as ea ws Racemes dense and headlike; pods 2-celled, coriaceous, cross-wrinkled, obliquely ovoid, 3-4 mm. long .... 21. A 2la. A A 22: A. . Nuttallianus var. trichocarpus . Nuttallianus var. acutirostris . Nuttallianus var. imperfectus dispermus 1. AsTRAGALUS AMPHIOxyYs Gray, Proc. Am. Acad. 13: 366. 1878. Xylophacos amphioxys Rydb. Bull. Torr. Bot. Club 32: 662. 1906. Texas to southern Nevada, Arizona and northern Chihuahua. Local habitat, occasional in Larrea Belt at about 1000 meters: Cottonwood Springs, Clokey 8496; Wilson’s ranch, Maguire 18035. Blooms in April. 214 MADRONO [Vol. 6 2. Astragalus Tidestromii (Rydb.), comb. nov. Xylophacos melanocalyx Rydb. Bull. Torr. Bot. Club 52: 149. 1925; not Astragalus melanocalyx Boiss. Nouv. Mem. Soc. Nat. Hist. Mose. 12: 59. 1860. Xylophacos Tidestromiu Rydb. Bull. Torr. Bot. Club 52: 155. 1925. Astragalus Marcusjonesii Munz, Leafl. West. Bot. 8:50. 1941. Southwestern Utah, southern Nevada, northwestern Arizona and southeastern California. Local habitat, gravelly, brushy soil in Larrea and lower Juniper belts at elevations of 1100 to 1300 meters: Kyle Canyon, in flower, April 26, 1937, Clokey 7564; in fruit, May 20, 1987, Clokey 7563; Kyle Canyon Fan, Clokey 7996, in fruit, May 15, 1936, Clokey 8220; Las Vegas to Red Rocks, in flower and fruit, March 31, 1940, Clokey 8596; Wilson’s ranch, in fruit, May 27, 1919, Tidestrom 9661 (type of Xylophacos Tides- tromii). Astragalus Tidestromi is abundant at a station 3 to 4 miles from Wilson’s ranch, the type locality of A. Tidestromi, at the same elevation and in a similar environment. Studies in the field and herbarium show that there is considerable variation in the pubescence and in the pods. On the leaflets the pubescence varies from parallel and appressed to kinky and tangled. The hairs are always attached at the end and not in the middle as in A.amphiorys. 'The hairs on the calyx may be white, or white and black mixed. The pods vary from 8 to 4.5 centimeters in length and may be curved from a small are to over half a semicircle. — The seeds are reticulate, speckled with purple, 3.5 to 4 millimeters long by 2 to 2.5 millimeters wide. Both A. melanocalyz and A. Tidestromi were described originally from limited material. Our material has been compared with isotype specimens of A. melano- calyx and the type of A. Tidestromi. It is evident that these do not warrant even varietal distinction. 3. AsTRAGALUS NEWwBERRYI Gray, Proc. Am. Acad. 12: 55. 1876. Xylophacos Newberryi Rydb. Bull. Torr. Bot. Club 32: 662. 1906. Utah and central Nevada south to western New Mexico, Ari- zona and extreme eastern California. Local habitat, scattered as single plants or small groups in openings on brushy ground in upper Larrea, Juniper and lower Pinyon belts: Clark Canyon, Clokey 7168; Charleston Park, Clokey 7169; Harris Springs road, Clokey 7570; Kyle Canyon, Clokey 7569, 8404, 8405; Kyle Canyon Ranger Station, Train 2169; Kyle Canyon to Deer Creek, Clokey 7571; Lee Canyon, Clokey 7171; Trout Creek, Clokey 7170; below Wheeler Wells, Clokey 7167. Blooms about May 1. : 3a. AsTRAGALUS NEwserRyI Gray var. funereus (Jones) comb. nov. A. funereus Jones, Contr. W. Bot. 12:11. 1908. Xylophacos funereus Rydb. Bull. Torr. Bot. Club 52: 867. 1925. Astragalus Purshii Dougl. var. funereus Jepson, Fl. Calif. 2: 360. 19386. 1942] CLOKEY: ASTRAGALUS 215 Southern Nevada and southeastern California. Local habi- tat, scattered and rare; openings on gravelly soil in the upper Larrea to the lower Yellow Pine belts: Kyle Canyon, Clokey 7568; Kyle Canyon trailer camp, Train 1677. Blooms about May 1. A close relationship to A. Newberryi is evident from a study of the pods, and the larger size of both the flowers and pods warrants varietal standing. The pubescence of the calyx con- sists of both white and black hairs with either predominating. 4, ASTRAGALUS coccINEUS (Parry) Brandg. Zoe 2:72. 1891. A. Purshii Doug). var. coccineus Parry, West. Am. Sci. 7: 10. 1890. Xylophacos coccineus Heller, Muhl. 2: 217. 1906. Colorado and Mohave deserts of California; reported from the Charleston Mountains by Jepson (FI. Calif. 359. 1936). Should be looked for on lower foothills especially on the western side of the range. Blooms in April and May. 5. ASTRAGALUS HUMISTRATUS Gray var. SONORAE (Gray) Jones, Contr. W. Bot. 10: 58. 1902. A. Sonorae Gray, Pl. Wright. 2: 44. 1853. Batiophaca Sonorae Rydb. N. Am, Fl. 24: 3817. 1929. New Mexico, southern Nevada, Arizona and Sonora, Mexico. Very local in the Charleston Mountains: ridge above Charleston Park at an elevation of 2330 meters, associated with Pinus scopulorum, P. monophylla and Juniperus scopulorum, Clokey 8408. The vegetative parts of specimens from the Charleston Moun- tains are near the lower limits in measurements. Blooms in June. 6. Astragalus aequalis sp. nov. Herba perennis erecta, e basi ramosa, 3—7 dm. alta; caules striati strigosi; folia 6-12 cm. longa; stipulae liberae anguste triangulares, 2-3 mm. longae; foliola 9-15 (plerumque 11) anguste lineari-lanceolata vel linearia obtusa utrinque strigosa, 12—40 mm. longa, 2—4 mm. lata; racemi axillares, folia subtendentia excedentes; flores 6-12, 10 mm. longi, lutei; calyx strigosus, pilis albis vel nigris, tuba 4—-4.5 mm. longa, dentibus subulatis vel anguste triangularibus, 1—-1.5 mm. longis; vexillum obovatum, apice emarginata, 12 mm. longum, 9 mm. latum; alae quam vexillo paullo breviores, lamina oblonga, 6 mm. longa, auriculo rotundo, 1 mm. longo; carina alis aequans, lamina 5—6 mm. longa, auriculo basalari brevi instructa; legumen sessile persistens chartaceum multo inflatum uniloculatum ellipticum, 3.5—-4 cm. longum, 2 cm. latum, 1—2 cm. diametro, suturis sub- aequaliter convexis, sutura ventrali suleata (ca. 1 mm.), albo- pubescens stramineum purpureo-maculatum vel purpurascens stramineo-maculatum; semina fusca, 2.5-3 mm. longa, 2 mm. lata. Perennial, erect, branched from base, 38—7 dm. high; stems striate, strigose; leaves 6-12 cm. long; stipules free, narrowly triangular, 2-83 mm. long; leaflets 9-15 (usually 11), narrowly lance-linear to linear, obtuse, strigose on both sides, 12-40 mm. long, 2-4 mm. wide; racemes axillary, 6-12 flowered, extending 216 MADRONO [Vol. 6 above the subtending leaves; flowers 10 mm. long, yellow; calyx tube 4—4.5 mm. long; teeth subulate to narrowly triangular, one- fourth to one-third the length of the tube, strigose with white or black hairs; banner obovate, slightly notched, 12 by 9 mm.; wings slightly shorter, blade oblong, 6 mm. long, with a rounded auricle 1 mm. long; keel as long as the wings, blade 5-6 mm. long, with a short, rounded, basal auricle; pods sessile, persistent, papery, much inflated, 1-celled, 3.5—4 cm. long, elliptical, cross-section elliptical, 2 em. wide, 1 cm. deep to rounded, 1.5 cm. in diameter, sutures nearly equally convex, ventral suture sulcate about 1 mm., white-pubescent, straw colored speckled with purple to purplish speckled with straw color; seeds smooth, brown, 2.5—3 mm. long, 2 mm. wide. Occurs at scattered locations in the Charleston Mountains, Clark County, Nevada: Harris Springs road, associated with Juniperus utahensis, elevation 1900 meters, in fruit, June 4, 1937, Clokey 7572 (type); Kyle Canyon, with Pinus scopulorum, eleva- tion 2180 meters, in flower, May 10, 1936, Clokey 7172; elevation 2270 meters, in fruit, July 2, 1936, Clokey 7173; Lee Canyon, elevation 2450 meters, June 16, 1939, Alexander 791; ridge north of lower Lee Canyon, elevation 2000 meters, in fruit, June 6, 1936, Clokey 7174; Willow Creek at 1810 meters, in fruit, June 15, 1937, Tram 1997. Astragalus aequalis is most closely related to A. Douglasu (T. & G.) Gray and A. Douglasi var. Parishit (Gray) Jones. The three can be distinguished as follows: Pods attached to a minute boss, falling free from the calyx, dor- sal suture much more convex than the nearly straight ven- tral suture. Stipules 4 mm. long; leaflets 15-23, elliptic to oblong; calyx tube 3 mm. long; teeth subulate, at least half as long as the tube. West central California, coastal. ............ A. Douglasti Stipules 4-5 mm. long; leaflets 11-25, oblong to elliptic- obovate; calyx tube 4 mm. long; teeth deltoid, one fourth to one third as long as the tube. Southern California, westiof the deserts) saci. - oc. occislehe ee Nearer ee A. Douglas var. Parishit Pods not attached to a minute boss, falling with the calyx, ven- tral and dorsal sutures equally convex; stipules about 2 mm. long; leaflets 9-15 (usually 11), narrowly lance-linear to linear, 12-40 mm. long, 2-4 mm. wide; calyx tube 44.5 mm. long; teeth subulate, one fourth to one third as long as the tube. Charleston Mountains, Nevada ................ A. aequalis 7. AsTRAGALUS Preussiu Gray, Proc. Am. Acad. 6: 222. 1864. Phaca Preuss Rydb. Bull. Torr. Bot. Club 40: 47. 1913. Central Utah, central Arizona, southern Nevada to south- eastern California. Local habitat, sandy or gravelly calcareous soil in the Larrea Belt below 1200 meters: Cottonwood Springs ranch, Clokey 8460; Indian Springs, Clokey 8406. Blooms in April. 8. ASTRAGALUS ARTIPES Gray, Proc. Am. Acad. 13: 370. 1878. Phaca artipes Rydb. Bull. Torr. Bot. Club 32: 664. 1906. 1942] CLOKEY: ASTRAGALUS 217 Colorado to Nevada and Arizona. Local habitat, with Pinus scopulorum at an elevation of about 2700 meters: Lee Canyon, July 11, 1988, Train 2141. 9, ASTRAGALUS ARRECTUS Gray var. REMoTUs Jones, Rev. Astrag. 162. 19238. Tium remotum Rydb. N. Am. FI. 24: 3891. 1929. From La Madre Mountain to Good Springs, Clark County, Nevada. Local habitat, among limestone and sandstone rocks at elevations from 1100 to 1700 meters: Cottonwood Springs, Clokey 8407; Excelsior Canyon, Clokey 8713; Mountain Springs, Clokey 7998; Rocky Gap Springs, Clokey 8714; Wilson’s ranch, Maguire 18041, 18067. Blooms in April or May. 10. AsTRAGALUS BECKWITHIU Torr. & Gray var. PURPUREUS Jones, Zoe 3: 288. 18938. Phaca artemisiarum Rydb. Bull. Torr. Bot. Club 40: 48. 1918. Phaceomene artemisiarum Rydb. N. Am. FI. 24: 3838. 1929. Western Utah, eastern and southern Nevada. Local habitat, widely scattered in dry soil in upper Larrea, Juniper and Pinyon belts at elevations from 1800 to 2450 meters: Charleston Park, Alexander 590; Clark Canyon, Clokey & Anderson 7164, 7165; Cold Creek, Clokey 7989, Train 1976; Cold Creek Spring, Clokey 7565; Deer Creek road, Clokey 7566; Harris Springs road, Clokey 8643; Kyle Canyon trailer camp, Train 1692; below Wheeler Wells, Clokey 7166. Blooms in May. 11. AsTRAGALUS PRAELONGUS Sheldon, Minn. Bot. Stud. 1: 238. 1894. A. Pattersonti Gray var. praelongus Jones, Contr. W. Bot. 10: 65. 1902. Jonesiella praelonga Rydb. N. Am, Fl. 24: 404. 1929. Southern Nevada and southwestern Utah; reported from the Charleston Mountains by Jones (Rev. Astrag. 156. 19238). Should be looked for on the lower foothills. 12. AsTRAGALUS MOHAVENSIS Wats. Proc. Am. Acad. 20: 361. 1885. Brachyphragma mohavensis Rydb. N. Am. Fl. 24: 400. 1929. Mohave Desert, California and Nevada. Local habitat, scattered and scarce; gravelly soil in Juniper Belt at elevations from 1500 to 1800 meters: Harris Springs road, Clokey 8687; Kyle Canyon, Clokey 7990, 7991. Blooms in May. 13. ASTRAGALUS PLATYTROPIS Gray, Proc. Am. Acad. 6: 526. 1865. Phaca platytropis Rydb. Mem. N. Y. Bot. Gard. 1: 246. 1900. Cystium platytrope Rydb. Bull. Torr. Bot. Club 40: 50. 19138. Rare on isolated peaks; Beaverhead County, Montana; Tooele County, Utah; Elko, White Pine and Clark counties, Nevada; Sonora Pass, California. Local habitat. Gravelly slopes at or above timberline on Charleston Peak at elevations of 3400 to 3500 meters; associated with Pinus aristata: Charleston Peak, Clokey 5518, 7992, 8001; southwest slope of Charleston Peak, Train 2292. Blooms in late July. 218 MADRONO [Vol. 6 The Charleston Peak plants differ constantly from the typical form in the following characters: stipules 1.5—2 mm. long, leaflets 11-19, calyx teeth 1 mm. or less long. The illustration (plate 42) in “Revision of the North American Species of Astragalus” by M. E. Jones is inaccurate in showing the septum extending from the dorsal suture only. The septum is formed by protrusions from both sutures meeting in the center of the pod. The seeds are dark brown and mitten-shaped. 14, ASTRAGALUS KERNENSIS Jepson subsp. charlestonensis subsp. nov. A specie differt: foliolis 15-19, leguminibus 1 cm. longis. Caespitose, decumbent perennial; stems 1-1.5 dm. long, slender, strigose; leaves 6 cm. or less long; stipules deltoid, 2 mm. long, strigose; petioles white strigose; leaflets 15-19, well sepa- rated, elliptical to narrowly obovate, obtuse, 4-7 mm. long, stri- gose on the lower face, glabrous on upper; racemes axillary, shorter than the subtending leaves, 2-6 flowered; peduncles slender, 2-3 cm. long; racemes 1 cm. or less long, the inflated pods appearing capitate; flowers white except for the purple tip to the keel, 8-10 mm. long; calyx strigose with white and black hairs; the tube about 3 mm. long; teeth 0.5—-1 mm. long; banner 8-10 mm. long, 4-5 mm. wide, nearly erect, entire or minutely notched at apex; wings nearly as long as the banner, blade 5-6 mm. long, 1.5 mm. wide, with reflexed, basal auricle; keel purple tipped, about 7 mm. long; blade 3.5 mm. long, with reflexed basal auricle ; pods sessile, papery, strigose, mottled, 1 cm. long, septum formed by protrusions from both sutures, reaching the tip, only the ventral suture sulcate, oval to nearly globular, obtuse at both ends, with a slender beak 1-2 mm. long; seeds about 5, mitten- shaped, 2.3 mm. long, 2 mm. wide. Known only from Charleston Peak. With Pinus aristata, ele- vation 3200 meters, July 29, 1937, Clokey 7573 (type) ; west slope near Trout Creek, elevation 10,000 feet, June 26, 1926, Jaeger (Pomona). The oval to spherical pods, obtuse at both ends with the partition, formed by protrusions from both sutures, reaching the tip, making the pods completely 2-celled, shows relation to A. kernensis Jepson not to A. lentiginosus Dougl. var. sierrae Jones or other forms near A. lentiginosus. These all have the partition formed by a septum, from the dorsal suture only, which does not reach the tip. | The subspecies may be separated from the species as follows: Leaflets 11-15, pods 6-7 mm. long, 8000-8500 ft., Tulare County,.-Caltlornia. 008 cect ihc leer en ois ean cot mab ee eg A. kernensis Leaflets 15-19, pods 1 cm. long, 10,000—10,500 ft., Charleston Peak, Clark County \Nevadar-.220)- cas ee ee A. kernensis subsp. charlestonensis 15. AsTRAGALUS LENTIGINosus Dougl. var. Fremontn (Gray) Wats. Bot. King Expl. 66. 1871. A. Fremontiu Gray, in Torr. 1942] CLOKEY: ASTRAGALUS 219 Puate 27. Astracatus. Figs a—j, Astragalus aequalis Clokey: a, pod, dor- sal view, X 1; b, pod, lateral view, X 1; c, d, cross sections of pods, <1; e, flower, x3; f, banner, X 2; g, wing-petal, x 2; h, keel, X 2; i, seed, X 4; j, leaf, x1. Figs. k-o, Astragalus kernensis Jepson var. charlestonensis Clokey: k, longitudinal section of pod, X 2; I, transverse section of pod, X2; m, banner, X 2; n, wing- petal, X 2; 0, keel, x2. Figs. p-z, Astragalus hemigyrus Clokey: p, raceme and leaf, X1; q, pod, X 2; r, transverse section of fresh pod, x5; s, t, cross sections of dry pods, X 5; wu, seed, X 4; v, flower, X 2; w, calyx, X 3; , banner, X 2; y, wing- petal, x 2; z, keel, x 2. 220 MADRONO [Vol. 6 Pacif. R. R. Rep. 4: 80. 1857. Cystium Fremonti Rydb. N. Am. Fl]. 24: 407. 1929. Southern Utah to the Death Valley region of California, south to Mexico. Local habitat, locally abundant in rocky, brushy ground in the upper Larrea and lower Juniper belts: Kyle Can- yon, Clokey 7175, 7574, Train 1672; mouth of Pine Canyon, Clokey 8612; Trout Creek fan, Clokey § Anderson 7176; Wilson’s ranch, Maguire 16596. Blooms about May 1. 16. AstRacaLus MINTHORNIAE (Rydb.) Jepson, Fl. Calif. 2: 374. 1936. Hamosa Minthorniae Rydb. Bull. Torr. Bot. Club 54: LS lO2 Southern Nevada to the New York Mountains, California. Local habitat, gravelly flats and slopes in the Juniper Belt at elevations from 1700 to 2200 meters: Clark Canyon, Clokey & Anderson 7180; Kyle Canyon, Clokey 7177, 7575, Train 1686; Moun- tains Springs, Clokey 7997; below Wheeler Wells, Clokey 7179. Blooms in May. 17. ASTRAGALUS BERNARDINUS Jones, Proc. Calif. Acad. ser. 2, 5: 661. 1895. Hamosa bernardina Rydb. Bull. Torr. Bot. Club oar. Oe O27. Mohave Desert from the San Bernardino Mountains, Califor- nia; reported from the Charleston Mountains by Jones (Rev. Astrag. 258. 1923). Should be expected on the lower foothills. Blooms in early spring. 18. Astragalus hemigyrus sp. nov. Herba perennis humilis frutescens argyreo-canescens; caules numerosi ramosi, 1—4 dm. alti; folia adscendentes, 5-10 cm. longa; stipulae triangulares acuminatae, 2 mm. longae; foliola 7-11, 6-15 mm. longa elliptica, apice obtuso vel retuso; pedunculi et racemi quam foliis subten- dentibus paullo longiores; bracteae subulatae, 1 mm. longae; pedicelli in fructu reflexi, leguminibus horizontaliter patentibus; flores purpurei, 6-8 mm. longi; calyx strigosus, pilis albis vel nigris, tuba 3 mm. longa, dentibus subulatis, 2 mm. longis; vexil- lum obovatum; alae quam vexillo 1 mm. breviores, lunatae, apice rotundo, auriculo magno reflexo; carina alis aequans; legumen 2.5-3 cm. longum, 4-5 mm. latum, subsessile deciduum non in- flatum valide reticulatum, uniformiter arcuatum, basi acuto, apice acuto in rostro brevi gracili attenuato, stylo curvato persistenti, biloculatum vel subbiloculatum fere ad apicem, septo crasso ex sutura dorsali extendenti, valvis immaturis crassis paullo succu- lentis, maturis coriaceis, sutura ventrali paullo prominenti, dorsali suleata; semina compressa ad hilum alte emarginata, 2.5 mm. longa, 1.5 mm. lata. Low, bushy, silvery-canescent perennial; stems numerous, branched, 1—4 dm. high; leaves ascending, 5—10 cm. long; stipules deltoid-acuminate, about 2 mm. long; leaflets 7-11, 6-15 mm. long, elliptic, obtuse or retuse; peduncles and racemes somewhat longer than the subtending leaves; bracts subulate, 1 mm. long; 1942] CLOKEY: ASTRAGALUS 221 flowers purple, 6—8 mm, long; calyx strigose with white and black hairs; tube 3 mm. long; teeth subulate, 2 mm. long; banner obovate; wings 1 mm. shorter than banner, lunate, rounded at tip, with a large reflexed auricle; keel the same length as the wings, rounded above to a blunt tip, with a reflexed, basal auricle; pedicels reflexed in fruit, pods horizontally spreading; pods sub- sessile, deciduous, not inflated, strongly reticulated, uniformly arched to a half circle, acute at both ends, tapering to a short, slender beak surmounted by the curved persistent style, when green, walls thick, somewhat fleshy, cross-section circular, dry walls leathery, cross-section cordate, ventral suture somewhat raised, dorsal suture sulcate, 2-celled or almost so nearly to the tip by a thick-walled open septum from the dorsal suture, 2.5—3 em. long, 4-5 mm. high; seeds brown, mitten-shaped, 2.5 mm. long, 1.5 mm. wide. Growing on rock ledges south of Indian Springs in the Larrea Belt, elevation about 1250 meters, April 18, 1989, Clokey 8409 (type) ; Clokey 7996, 8593. Astragalus hemigyrus is most closely related to A. Layneae Greene from which it may be separated as follows: Stipules 7-10 mm. long; leaves near base of plant; leaflets 13-23, 1-1.5 cm. long; flowers white with purple tip, 15-20 mm. long; calyx 5-7 mm. long; pod 3-5 cm. long, 6-7 mm. wide, pilose-canescent with somewhat curly hairs, curvature of Pod most pronounced near tip |.......... «22.2.2 ...%0..5. A. Layneae Stipules 2 mm. long; leaves throughout length of stem; leaflets 7-11, 6-15 mm. long; flowers purple, 6-8 mm. long; calyx tube 3 mm. long, pods 2.5-3 cm. long, 4-5 mm. wide, strigose with short appressed hairs, curved nearly uniformly MH TROU GAN cag ee Pe SecA a wy ie OS el a ud eg Ge A. hemigyrus Jones (Rev. Astrag. 261. 1923) reports A. albens from Indian Springs, Charleston Mountains. No specimens to sub- stantiate this record are in the Jones Herbarium at Pomona Col- lege or in the National Herbarium where many of Jones’ first sets are deposited. Astragalus albens is a local species of the San Ber- nardino Mountains of California. Rydberg (Bull. Torrey Bot. Club 54: 22. 1927) calls attention to Jones’ description of the pods of A. albens “ ‘arched mostly to a circle, . . . when mature coriaceous, strongly corrugated, 2-3 cm. long, 8 mm. wide and high, flat for about 1 mm. high along the ventral suture and form- ing a thick wing, etc.’ In the type number the pod is only 1.5 cm. long, forming an arch of about one fourth of a circle, neither coriaceous nor corrugated.” The type specimen and other collec- tions from and near the type locality fit the original description. Jones’ description of the pods of A. albens would serve for the pods of A. hemigyrus. It is believed that there is no justification for including A. albens in the flora of the Charleston Mountains. 19. AstraGaLus catycosus Torr. in Wats. Bot. King Expl. 66. 1871. Hamosa calycosa Rydb. Bull. Torr. Bot. Club 40: 50. 1913. 222 MADRONO [Vol. 6 Western Wyoming and Idaho south to southern Nevada and eastern California. Local habitat, slopes in Juniper Belt at ele- vations of 2000 to 2200 meters: ridge along lower Lee Canyon, Clokey & Bean 7589, Clokey 8002; below Wheeler Wells, Clokey 7168. Blooms in June. 20. ASTRAGALUS MANCUS (Rydb.) Wheeler, Rhodora 40: 1386. 1938. Hamosa manca Rydb. Bull. Torr. Bot. Club 54:17. 1927. Northeastern to southern Nevada. Local habitat, slopes and hilltops from timberline with Pinus aristata at elevations of 3300 meters to 2600 meters with Pinus scopulorum: Charleston Peak, Clokey 5516; ridge south of Deer Creek, Clokey 8635; be- tween Kyle Canyon and Deer Creek, Clokey 8000, Alexander 792b; Lee Canyon, Lakivers & Hancock 514, Clokey 7999, S681, Train 2073, Alexander 792a. Blooms in late June or July. 21. AsTrracatus Nutrauuianus DC. var. TricHocarpus Torr. & Gray, Fl. N. Am. 1: 334. 1838. Hamosa austrina Small, ee Southeast. U. S. 618, 1832. 1902. Colorado to southern California south to Texas and Lower California. Local habitat, rocky ground in the Larrea and lower Pinyon belts at elevations below 1700 meters: Mountain Springs, Clokey § Anderson 7985. Blooms about May 1. 21a. Astracatus Nutratuianus DC, var. acuTirostris ( Wats.) Jepson, Fl. Calif. 2: 379. 19386. Astragalus acutirostris Wats. Proc. Am. Acad. 20: 860. 1885. Hamosa acutirostris Rydb. Bull. Torr. Bot. Club 54: 331. 1927. West central Nevada to the Sierra Nevada, south to the Colo- rado Desert, California. Reported from the Charleston Moun- tains by Jones (Rev. Astrag. 271. 1928). Should be looked for on the lower foothills in the early spring. 21b. Astracatus Nurrauuianus DC. var. mperrectus (Rydb.) Barneby, Leafl. West. Bot. 3: 109. 1942. Hamosa imperfecta Rydb. Bull. Torr. Bot. Club 54: 829. 1927. Nevada, Arizona and Lower California. Local habitat, dry, rocky soil in the Larrea Belt: ridge east of Wilson’s ranch, eleva- tion 1320 meters, Clokey 8712. Blooms about May 1. 22. ASTRAGALUS DISPERMUS Gray, Proc. Am. Acad. 13. 365. 1878. Hesperastragalus dispermus Heller, Muhl. 1: 187. 1906. Astragalus didymocarpus Hook. & Arn. var. dispermus Jepson, FI. Calif. 2: 376. 1936. Western Arizona, southern Nevada and California south to Lower California. Reported from the Charleston Mountains by Jones (Rev. Astrag. 285. 1923). Should be expected at the lower elevations. Blooms in March or April. South Pasadena, California, January 19, 1942. 1942] ST. JOHN: POLYSTICHUM LEMMONI 223 THE TYPE LOCALITY OF POLYSTICHUM LEMMONI UNDERWOOD Harowp St. Joun By detective methods it is often possible today to locate rather exactly the type locality of species described but not defi- nitely localized by earlier botanists. These notes are written to publish more information on the type locality of Polystichum Lemmoni Underwood (Our Native Ferns, ed. 6: 116-117, 1900). The published type locality was “Near Mt. Shasta, California (Lemmon). This appears to have been a loose usage of the geo- graphic term, as the strikingly distinctive, but rare, fern has not been subsequently rediscovered on Mount Shasta. William B. Cooke (Am. Fern Journ. 29: 109, 1939) in his account of the ferns of Mount Shasta proper, excluded P. Lemmoni from the list, concluding that Lemmon’s specimen, the type of the species, was probably not collected on Mount Shasta. This is in agreement with the detailed review and discussion of this type locality by Louis C. Wheeler (Am. Fern Journ. 27: 121-126, 1937). The type specimen is in the herbarium of the New York Botanical Garden. Two of the Lemmon collections there were labeled P. Lemmoni Underw. by Dr. L. M. Underwood, but neither definitely marked as the type. The first was collected on Mount Eddy, July 12, 1878; the second, near Shasta, California, July 1879. Dr. H. A. Gleason designated the second as the lectotype, since Shasta was the published type locality, and Dr. L. C. Wheeler at one place agreed with this choice. Wheeler discussed and cited the several Lemmon collections in different herbaria and their variously worded data and indicated that the type locality ‘near Shasta,’ did not apply to the former valley settlement called Shasta. Itis apparent that duplicate collections were distributed by Lemmon with varying statements of the locality data. The writer, while preparing a biography, has searched far and wide for botanical correspondence with Lemmon. That with C. B. Davenport does not help on this particular fern, but there is evidence in the letters from J. G. Lemmon to Professor D. C. Eaton, preserved in, and kindly made available by, the Stirling Library, Yale University. “Sierra Valley, Cal, August 26, °79. “Now for the most astonishing part of your letter—the new Aspidium. Is it possible that it is distinct? Why it is abundant in a certain valley not 30 miles west of Mt. Shasta, the old stamp- ing ground of hosts of botanists. I will send you full specimens as I have here a fine lot, . . . I was struck by the appearance of the Aspidium & gathered a lot of it for it looked such a marked variety. And I fear yet it may prove a munitum for some of the fronds are large & approached the type—in appearance. 224 MADRONO [Vol. 6 “Now for ‘habitat, soil, moisture, exposure, abundance, scar- city?’ ete. It is quite abundant on the side of the little valley at the headwaters of the South Fork of the upper Sacramento & along the south sloping side of Mt. Eddy, arising on the N. side of this valley. Protrudes from under rocks, a vast number of fronds together—more than any munitum I ever saw (which took my eye). The soil a dissolved granite, quite moist & loose, the inclination generally to the S. at a steep angle. “What is very singular is that a grove of the long lost Pinus Balfouriana extends over the same ground, two excellent things found in one day! No objection to the name you are kind enough to propose. Nothing so fine as a fern, and such favorites with the ladies !!”’ Lemmon later gave details of the occurrence of Pinus Bal- fourrana Jeffrey: ““A few trees at an altitude of 7,500 feet forming a dark-green belt on the south flank of one of the eastern spurs of Scott Mt., 20 miles west of Shasta, where Jeffrey detected it in 1852 (rediscovered by the writer, in 1878; only other California localities, a few trees near the headwaters of Kings River, in the Southern Sierra.” | (2nd. Bienn. Rept. Calif. State Bd. Forestry 1887-88: 71, 1888). Further on (pp. 86-87) Lemmon con- tinues, “Jeffrey noted his discovery, ‘Mountains between Shasta and Scott Valley, N. Cal. Lat. 40° 30’ to 41° 51’. Elevation 5,000 to 8,000 feet.’ . . . But so small are the groves, and so local their position, that they were not detected anew until August of 1878, when the writer, making his headquarters at Sisson, prosecuted a thorough search of the various intricate mountain ranges lying west of Shasta, and forming spurs of the diversified Scott Moun- tains. I noted the locality for publication in ‘Brewer's Botany of California,’ as ‘on the southern flanks of the Scott range of moun- tains, forming a dark-green belt, from 5,000 to 8,000 feet altitude, between the light-colored P. monticola below and P. albicaulis above it.’ ” It appears that at first Eaton thought the fern from “near Shasta” to be a new species and he wrote to Lemmon announcing that he would name it Aspidium Lemmoni in his honor. One can imagine the intense pleasure this gave to Lemmon, a fern-lover. Then, on further consideration, Eaton decided that it was not a new species but was identical with Aspidium mohrioides Bory of Southern Chile and Patagonia. He published this determination (Eaton, D. C., Ferns of N. Am. 2: 128, 1879; 251-254, pl. LXXX, figs. 4-9, 1880, and Torrey Bot. Club, Bull. 6: 360-361, 1879). On page 128 Eaton recounted this, “At first I believed it to be a distinct species, and proposed to name it after its discoverer, a gentleman whose own modesty has been the innocent reason why some Californian fern was not long ago named in his honor.” Wheeler (p. 122) quotes the two indefinite statements of the locality of this rare fern given by Eaton in his “Ferns of North 1942 | ST. JOHN: POLYSTICHUM LEMMONI 225 America,” but omits the third and more detailed one (p. 252), viz., “Mr. Lemmon writes that his fern grows in loose and moist granitic soil the root-stocks hidden under rocks, and a great many plants in one cluster. ‘It is very abundant on the side of a little valley at the headwaters of the South Fork of the Sacramento, and along the southern sloping side of Mount Eddy, which rises on the northern side of this valley.’ ”’ Several of the Lemmon collections seen by Wheeler were labeled Scott Valley, Siskiyou County, July 23, 1879, and Wheeler concludes (p. 128) that this is the real type locality. The Scott Mountains are a ridge connecting with Mount Eddy and running in a southwesterly direction from it. Only the northwestern slopes are in Siskiyou County. Scott Valley, drained by the Scott River into the Klamath, runs northerly from Scott Mountain and is about sixteen miles westerly of Mount Eddy. The headwaters of the Sacramento on the slopes of Mount Eddy are in Shasta County, about sixteen miles southwesterly of Mount Shasta, so the labels would indicate that Lemmon in 1879 found two locali- ties, one on Mount Eddy collecting there on both July 12, 1878 and July 23, 1879, and one on Scott Mountain. We also have Lemmon’s statement that he also found the fern in August, 1878 on the same day that he discovered a grove of Pinus Balfouriana on the southern side of Scott Mountain at 7,500 feet altitude. However, it is significant that in the contemporary letters and in the data on specimens furnished at the time to Gray and Eaton, that only the Mount Eddy locality is mentioned. Wheeler (p. 123) puzzled over a printed label of one speci- men indicating apparently that J. G. Lemmon and wife were col- lectors of a specimen dated 1879. Mr. Lemmon’s name was underlined and this would seem to indicate that he alone was the collector. It can now be positively so stated, since J. G. Lemmon and Miss Sara A. Plummer were not married till late November, 1880. Mr. Lemmon advertised for sale, specimens of “Aspidium Mohrioides Bory (New to North America),” on a handbill (Pacific Coast Flowers and Ferns, Distribution of 1880). He again enumerated this rare fern as Aspidium mohrioides Bory (Ferns of the Pacific Coast, ed. 1: 12, 1882). He named it the New Shasta Shield Fern, and located it at “Mt. Eddy, Head-waters Sacra- mento River; near Shasta Cal. 1879. (New Species!).” His insertion of the phrase ““New Species” was quaint, to say the least, when he accepted it, upon Eaton’s determination, as the old spe- cies described by Bory de St.-Vincent. Wheeler (p. 124-125) decided that this listing by Lemmon was based upon two of his own locality records, Mount Eddy, and near Mount Shasta (=Scott Valley), and this is now confirmed. The writer has two volumes of ““American Ferns,” quarto sized books bound in blue cloth with the title and ornamental designs 226 MADRONO [ Vol. 6 of ferns on both cover and title page. These books have no other printed words, and no indication of author. They have on each page a pressed fern collected by Lemmon, attached by strips and with an herbarium label with data. Each contains a specimen of Aspidium Mohrioides Bory and a printed label form with “U. S. Pacific Slope Flora, (California). Coll. by J. G. Lemmon and wife, Oakland, California, 188-.’”’ The data is in Mr. Lemmon’s handwriting. On one it is “Near Shasta, 8,600 ft. alt. Found elsewhere only in Patagonia and Falkland Is. S. Am. Jul. 1878.” On the other it is, “Near Shasta. N.Cal. ‘Only found elsewhere in Patagonia.’ July 1883.” These books were made up by Mr. and Mrs. Lemmon for sale to botanists and fern-lovers. Doubt- less many of the ferns found in our public herbaria were supplied in this book form. The two books are not identical in number of species or arrangement. It is not clear just how many times Lemmon revisited his localities in the Shasta region, but he kept dried specimens of this fern in stock and it was one of his most unique and desired collections. Dr. W. L. Jepson has informed the writer that Lemmon did not keep complete collection number books. None of any sort has survived. The sets of his plants sorted and distributed by Dr. Asa Gray were handled differently, but those isssued by Lem- mon himself were selected from the duplicate stock, each set individually, when ordered. Real duplicates of a single season’s collecting were thus prepared and issued over a period of decades, using whatever labels were then available, and Mr. Lemmon or Mrs. Lemmon inserting the written data (perhaps from memory) with variations in wording inevitable by this method. This doubtless explains the existence of the several apparently differ- ent habitats and localities for Lemmon’s collections of this one fern in the general vicinity of Mount Shasta. The lectotype designated by Gleason is ‘‘near Shasta, July 1879.” This lectotype lacks accurate locality data, especially since it is now known that the species does not occur on Mount Shasta. Later Wheeler (p. 122) more precisely chose the type locality as Scott Valley. He was apparently influenced by the existence of a specimen from Lemmon’s own herbarium, now at the University of California, labeled ‘Scott val. near Shasta, July 23,1879, J.G. Lemmon.” This agreed with the data published by Underwood, “Vicinity of Mount Shasta, Calif.” and was a speci- men retained by Lemmon. Other factors, however, provide argu- ments against Wheeler’s choice. The type specimen cannot be that in Lemmon’s herbarium, but must be one of the two speci- mens in the New York Botanical Garden labeled as the new spe- cies by Underwood himself. The Lemmon herbarium received by the University of California was the remnant left after Mr. and Mrs. Lemmon had eked out a living by selling their specimens. The remaining collections had poor data, many of these specimens 1942] BALL: SALIX 227 when incorporated in the Berkeley collection could only be labeled “‘Lemmon Herbarium” as there was no exact statement of data found. In any case, Underwood did not study or cite these particular specimens, rather the ones in New York. It seems well demonstrated that Mount Shasta was intended as the general area and cannot have been the exact locality. Though Lemmon collected this rare fern in Scott Valley, also on the south side of Scott Mountain, and also on the south side of Mount Eddy, the latter is here proposed as the lectotype locality. It is selected because of the detailed locality data supplied contemporarily by Lemmon to D. C. Eaton, and published by Eaton [for Aspidium mohrioides|, and of Lemmon’s own listing of Mount Eddy in his Ferns of the Pacific Coast. These statements are quoted here fully on a previous page. The exact taxonomic position of this fern continues to trouble the botanists. At first Eaton considered it a distinct species, then on reconsideration determined it as Aspidium [= Polystichum] mohrioides Bory, and later H. Christ agreed. Underwood sepa- rated it as a new species, Polystichum Lemmoni which was accepted by Piper and by Maxon. Now, Professor Fernald (Rhodora 26: 92. 1924) has made the northern plant.a variety of the South American one and evaluated its characters. He classifies it as Poly- stichum mohrioides (Bory) Pres] var. Lemmoni (Underw.) Fernald. The writer has not made a detailed revision of this group, but he recently compared the North American P. Lemmoni with good material of P. mohrioides from the far extreme of South America, and was struck by their dissimilarity. For the time being he is content to follow Underwood and Maxon, and to accept Poly- stichum Lemmoni Underwood as a species. University of Hawaii, Honolulu, April 14, 1942 FAR WESTERN NOVELTIES IN SALIX CaRLETON R. BAL Activities of collectors continue to bring to light hitherto unrecognized variations in willows. Continuing studies of re- lationships indicate a need for new combinations which better represent actual affinities. This paper contains some novelties in each category. | The abbreviations for herbaria containing specimens cited are as follows: BPI, National Arboretum Herbarium, Bureau of Plant Industry, United States Department of Agriculture; CAS, Her- barium of the California Academy of Sciences; CRB, Salix herba- rium of Carleton R. Ball; CUA, Herbarium of the Catholic Uni- versity of America; USN, United States National Herbarium; SU, Herbarium of Stanford University; UC, Herbarium of the University of California. 228 MADRONO [Vol. 6 New VARIETIES OF SALIX PULCHRA CHAMISSO Salix pulchra, with rather broadly elliptical leaves acute at both ends, is a species of far northwestern North America. It occurs throughout Alaska, except the southeastern portion, and in northwestern British Columbia, most of Yukon Territory, and the lower Mackenzie Valley. Like many species of Salix, it exhibits both broad-leaved and narrow-leaved variations from the normal or average. The Swedish salicologist, Andersson, frequently added varieties lati- folia and angustifolia when describing new species, or reviewing old ones, as in S. Richardsoni. The late American salicologist, Bebb, also often described one or both variations, as in S. laevigata and S. glaucophylla. More recently, Fernald has separated the Labradorean S. cordifolia into a series of varieties based on leaf size, Shape, and pubescence. Still more recently, Schneider has erected varieties of S. anglorum, S. ovalifolia and others on leaf width and shape. The writer also has followed this practice in S. lasiandra, S. lutea, S. glauca, S. reticulata and other species. Because these variations in S. pulchra render difficult its com- plete recognition from descriptions of the more typical material, it seems desirable to describe as varieties its two chief leaf variations. Salix PuLcHRA Cham. var. Looffiae var. nov. E forma typica speciei differt foliis anguste vel late obovatis vel obovato-ovalibus, apice rotundatis vel apiculatis vel terminalibus acutis. The variety Looffiae differs from the species in having its leaf-blades narrowly to broadly obovate, or obovate-oval, rounded to apiculate at apex or the distal leaves acute. Common dimen- sions in centimeters are: 1 xX 2—2.5, 1.5 x 8—3.5, 2 x 8-4, 2.5 x 3.5—4, 3 x 4, 2.5-3 x 5 and 3-3.5 x 6-7. It is a pleasure to name this willow for Ethel H. (Mrs. Henry B.) Looff of Oak Harbor, Washington, who has collected on Kodiak Island during two seasons. Her critical ecological work has done much to explain the distribution and the peculiar ex- pression of arctic willows on that island. Specimens referred to this variety are listed below. Most of the plants from coastal areas are recorded as of prostrate or de- pressed habit, but this is true also of specimens of typical S. pulchra. The specimen collected by Setchell appears to have been erect in habit. SouTHERN Axaska. Kodiak Island: Alitak, prostrate, moun- tain slope, eastern exposure, altitude 500 feet, May 26, 1940, Ethel H. & Henry B. Looff 1191 (type, pistillate, CRB, 3 sheets) ; prostrate, southern exposure, 119SA (CRB, pist.) ; decumbent, in mixed moss and grass association, altitude 75 m., W. J. Eyerdam 2047 (CRB, pistillate) ; no specific locality, Roland Snodgrass 39 (CRB); altitude 1385-2500 feet, Gulkana to Paxson, Wm. A. & 1942] BALL: SALIX 229 Clara B. Setchell 77 (CRB, UC). Western Atasxa. St. Paul Island, Pribilof Islands, August 7, 1891, C. H. Merriam (USN) ; plants browsed by muskoxen, Nunivak Island, O. J. Murie 2060 (CRB); St. Matthew Island, Coville § Kearney 2086a (USN); St. Lawrence Island, Northeast Cape, Coville § Kearney 2001 (USN) ; King Island, Bering Sea, J. P. Anderson 3607A (CRB). Nortu- ERN AxuaskA. Walker Lake, Kobuk River, August 21, 1901, W. C. Mendenhall (USN); Alatna River, 65 miles above mouth, July 23, 1901, Mendenhall (USN); Beaver, Yukon River, W. A. & C. B. Setchell 408 (CRB, UC, USN); Circle City, Yukon River, W.A.& C. B. Setchell 392 (CRB, UC, USN). The more vigorous plants, like those from the Yukon River and the Alatna River (tributary of the Koyukuk), have elongated seasonal shoots with broad leaves at the outer end and relatively narrower blades below. SaLix putcHRA Cham. var. Palmeri var. nov. E forma typica speciei differt foliis anguste oblongis vel elliptico-oblongis vel anguste elliptico-oblanceolatis, apice acutis vel acuminatis, 0.8- 1.5 cm. latis, 4-6(8) cm. longis. The variety Palmeri differs from the species in having nar- rowly oblong, elliptic-oblong, or narrowly elliptic-oblanceolate leaf blades, 0.8—1.5 centimeters wide, 4-6 or 8 centimeters long, and acute to acuminate at the apex. Common dimensions in centimeters are: 0.7 x 3.5, 0.8—1 x 4, 0.8—-1.2 x 5, 1-1.8 x 6, 1.5 x 7, and 1-2 x 8. A form described by Andersson of Sweden requires con- sideration. In 1858, he published a paper on American willows simultaneously in Sweden (Oefvers. Kon. Vet.-Akad. Forh. 15: 100-133) and America (Proc. Am. Acad. Arts & Sci. 4: 50-78). In the Swedish paper (p. 123), he described S. phylicoides, be- lieved by Bebb to be S. pulchra but by Schneider to be a different Asiatic species, and added: “-latifolia.” “-angustifolia,” without description or indication of rank. In the American paper, edited and annotated by Asa Gray, these entities are called forms and described (p. 64). The second reads: “Forma angustifolia: foliis 1-2 pollicaribus 14 poll. latis lanceolatis integerrimis.’ Gray doubtless got this information from Andersson. In his Mono- graphia Salicum, 1867, Andersson described this form as “angusti- folia: foliis lanceolato-linearibus, 4—5 pollices longis medio 144-34 poll. latis....’ In his treatment of world Salices (DC. Prodro- mus 16 fase. 2: 190-323. 1868), he writes “2, argentifolia: foliis lanceolato-linearibus 2—8 poll. longis. ...” (p. 245). The name argentifolia doubtless was a typographical error but the striking discrepancies in leaf length remain unexplained. No localities or collectors are cited in any paper. Because of this omission, the discrepancies noted, and the uncertain identity of S. phylicoides, Andersson’s form must be disregarded. The specimens referred to var. Palmeri are listed below. The 230 MADRONO [Vol. 6 variety apparently is most common in a belt extending south to north through central Alaska, with an outlier to the west on Nor- ton Sound and to the east on the Arctic Coast of northeastern Yukon Territory. The type (Palmer 121) was collected in the Matanuska Valley of south-central Alaska. Another specimen of the type collection is in the herbarium of the Fish and Wildlife Service at the Research Laboratory of the Patuxent Wildlife Refuge near Beltsville, Maryland. It is a pleasure to name this variety for L. J. Palmer of the United States Fish and Wildlife Service at Fairbanks, whose collections of plants browsed by moose and reindeer have done much to increase our knowledge of their distribution, ecology, and taxonomy. SouTH-CENTRAL Ataska. Alaska Peninsula, Kukak Bay, Coville § Kearney 1633 (CRB, USN); Kenai Peninsula, between Skilak and Tustumena lakes (moose-browse reconnaisance), L. J. Pal- mer, 1, 6, 22, 32, 36, 56, 66 (CRB); Olga Bay, Kodiak Island, alti- tude above 1600 feet, KE. H. §& H. B. Looff 356 (CRB); mountain shrub type, Matanuska Valley, Palmer 121 (CRB, type) ; Richard- son Highway, Gulkana to Paxson, W. A. & C. B. Setchell S1 (CRB, UC) ; Summit Lake, W. A. § C. B. Setchell 105 (CRB, UC) ; Mount McKinley National Park, Savage River Headquarters, altitude 3800-4200 feet, W. A. §& C. B. Setchell 185, 186, 189, 193 (CRB, UC). West-centrat Ataska. Norton Sound: Egavik, August 11, 1931, C. H. Rouse 13, 14 (CRB); Pastolik Reindeer Camp, Rouse 26 (CRB). Cerntrat AtasKa. Vicinity of Fairbanks: Pedro Dome, altitude 2800 feet, August 11, 1909, R. S. Kellogg (USN) ; McKinley Creek, tributary of Forty-mile River, Murie 141 (USN). Steece Highway, Fairbanks to Circle City, Twelve-mile Summit, altitude 3225 feet, W. A. §& C. B. Setchell 551 (CRB, UC). Yukon Territory. Near delta of Mackenzie River, Mackenzie Bay, Shingle Point, 69 north lat., 137 west long., A. Dutilly 18141 (CUA), 16743, 5174, (CRB, CUA): New SPECIFIC AND VARIETAL COMBINATIONS Following study of more adequate material, it often becomes necessary to elevate varieties to specific rank, to degrade specific entities to varietal rank, or to transfer varieties from one species to another. Changes of all three kinds are made herein. Salix Walpoleii (Coville & Ball) comb. nov. Salix Farrae Walpolew Coville and Ball, Bot. Gaz. 71: 435-486. 1921. It was a mistake to arrange this northern willow as a variety of the more southern species belonging to Section Cordatae. Fur- ther study and more abundant material show that it is most closely related to S. pyrifolia Andersson (S. balsamifera Barratt), which Schneider separated from the Cordatae and made the basis of a new section, Balsamiferae. Our plant, however, appears to be more properly regarded as a second species in that section than as a variety of S. pyrifolia. 1942] BALL: SALIX 231 The description given at the time of varietal publication still is reasonably adequate. The branchlets are somewhat more pu- bescent than was then indicated, the distal leaves on vigorous shoots may be subovate and rounded at base, the stipules reach 8 millimeters, and the styles sometimes are 0.4 millimeters long. A comparison of some contrasting characters of S. pyrifolia and S. Walpolei is given below. OrGAN S. pyrifolia S. Walpoleu Young branchlets glabrous more or less pubescent Blades shape ovate or ovate-lanceolate elliptical-lanceolate to obovate base cordate or rounded acute or distal roundish margin serrulate or crenate-ser- entire or remotely cren- rulate ulate venation shallowly rugose reticulate Stipules wanting or minute 2-6 or 8 mm. long Pistillate aments 3-8 cm. long 2-5 or 6 cm. long Capsules 7-9 mm. long 5-7 mm. long Pedicels 2.5-3.5 mm. long 1.0-1.5 mm. long The known range of this plant has been greatly extended since 1921. The nine specimens then listed were all from north- west Alaska and ranged in location from Seward Peninsula at Bering Strait to the Dall River north of the Yukon River, at about 150 west longitude. More recent collections have extended its recorded range far to the south and east, as shown below. WEST-cENTRAL ALASKA. Seward Peninsula and adjacent Yu- kon Valley: Seward Peninsula, 1900, A. J. Collier (USN) ; vicinity of Port Clarence: north side and east end of Grantley Harbor, F. A. Walpole 1594 (USN); rocky banks, northwest shore of Imunik Basin, July 30, 1901, Walpole 1624 (CRB, photo; USN, pistillate type); banks of Tuksuk Channel, August 5, 1901, Wal- pole 1742 (CRB, USN staminate type) ; Cape Nome, summer 1900, F. E. Blaisdell (USN); gravelly bluff near road, Hastings Creek, C. W. Thornton 614 (USN); near Nome, Thornton 630 (CRB, USN); tundra, Nome, George N.. Jones 9043B (CRB). Kaltag, Yukon River east of Norton Sound: bank of Yukon River, Rouse 45 (CRB); water’s edge, W. A. & C. B. Setchell 457-459 (CRB, UC). Norru-centrat Axuaska. Valley of Kobuk River, near camp, August 20, 1901, W. C. Mendenhall (USN); Valley of Alaskuk River, 30 miles above mouth, July 21, 1901, Mendenhall (USN); Valley of Alaskuk (Alatna?) River, along Help-me-Jack Creek, near camp, July 26, 1901, Mendenhall (USN); Dall River, 75 miles above mouth, June 25, 1901, Mendenhall (USN, 2 sheets) ; Wiseman, middle fork of Koyukuk River, J. P. Anderson &§ G. W. Gasser 5815 (CRB). East-centrat AxuasKa. Steece Highway, Fairbanks to Circle City: creek near Twelve-Mile Roadhouse, altitude 2450 feet, W. A. & C. B. Setchell 530, 534b (CRB, UC); Faith Creek, some distance below Twelve-Mile Sum- mit, to Cleary Summit, altitude 2600-2700 feet, W. A. & C. B. 232 MADRONO [Vol. 6 Setchell 555, 561 (CRB, UC); Twelve-Mile Roadhouse, Anderson 2433 (CRB). Mt. McKinley National Park: Igloo Camp, altitude 2600 feet, W. A. §& C. B. Setchell 173 (CRB, UC); stream banks near Park Headquarters, dven Nelson 3595, 3604 (CRB); Cant- well, southeast corner of Park, Nelson 4215 (CRB). Richardson Highway: Rapids Roadhouse, altitude 2130 feet, W. A. & C. B. Setchell 110, 120 (CRB, UC); altitude 2700-3000 feet, Paxson, W.A.& C. B. Setchell 87, 90,95 (CRB, UC). Mackenziz, Nortn- west Territory. Alkavik, 68 deg., 13 min. north lat., 135 deg. west long. 4. Dutilly 18054 (CRB, CUA), 18055 (CUA) ; the com- mon willow of alluvial ridges, Mackenzie Delta, Pete’s Creek, east side of Richards Island, between 69 and 70 deg. north lat., J. J. Lynch & C. E. Gilliam 1605 (Herbarium of United States Fish and Wildlife Service). Savix Hinpsiana Benth. var. leucodendroides (Rowlee) comb. nov. Salix macrostachya leucodendroides Rowlee, Bull. Torr. Bot. Club 27: 250, pl. 9, fig. 6 (doubtfully representing this variety). 1900; Abrams, Fl. Los Angeles 102, 1904 (at least in part) ; ibid., Suppl. ed. 102. 1911. S. argophylla Nutt. sensu Rowlee, Bull. Torr. Bot. Club 27: 251, pl. 9, fg. 7. 1900 (Gin part); Albramg ibid. 102. 1904 (probably in part); Jepson, Man. FI. Pl. Calif. 264. 1923 (in part), not of Nuttall. S. exigua virens Rowlee, Bull. Torr. Bot. Club 27: 255. 1900 (in part). S. integrifolia leuco- dendroides Rowlee, Bull. Torr. Bot. Club 27: 250. 1900 (nomen nudum). 8S. longifolia Muhl. sensu Parish, Zoe 4: 847. 1894 (in part, as indicated by localities), not of Muhlenberg. S. longifolia argyrophylla Anders. sensu Jepson, Mem. Univ. Calif. 2: 178. 1910 (in part), not of Andersson. S. sessilifolia Nutt. sensu Britton and Shafer, No. Amer. Trees, 156. 1908 (in part) ; Jepson, Mem. Univ. Calif. 2: 178. 1910 (in part), not of Nuttall. S. sessilifolia leucodendroides (Rowlee) Schneider, Bot. Gaz. 65: 26. 1918, ibid. 67: 819-822. 1919 (synonymy, discussion and citation of speci- mens), Jour. Arn. Arb. 3: 64, 86 (and pages cited for S. sessili- folia). 1922; Ball in Abrams, Illus. Fl. Pac. States 1: 491 (dis- cussed under S. Hindsiana Bentham). 1923; Jepson, Man. FI. Pl. Calif. 264. 19238. The variety leucodendroides differs from the species in having © longer, relatively narrower, more pointed, and always remotely denticulate leaves, usually less densely pilose capsules and flower seales, and less evident styles. Seasonal shoots usually densely white-pubescent, older branchlets less so; blades linear to linear-elliptical, 5-8 cm. long and 4—7 mm. wide or, on vigorous shoots, 9-10.5 cm. long and 6-8 or 9 mm. wide, common sizes, 40.5, 5 x 0.4-0.5, 6 x 0.40.6, 7x 0.5-0.7, 8x0.5-0.7, 9-10 x 0.6—-0.8 em., short-acuminate at apex, tapering at base into a short petiole, always remotely denticulate, especially on the outer half, the teeth sometimes sub- spinulose, densely pilose-pubescent and usually silvery when 1942] BALL: SALIX 233 young, becoming more thinly clad by expansion, and frequently becoming more or less glabrate and greenish with age (and then often referred to S. exigua virens Rowlee); aments coetaneous, leafy-peduncled, solitary or 2—4 together; peduncles 1—5 cm. long in flower, the pistillate up to 8 cm. long in fruit; staminate aments 1 or usually 2-8, or occasionally 4, on one peduncle; pistillate aments 1 or 2 together, 2.5—4 or sometimes 5-6 cm. long; capsules lanceolate, 5—-5.5 or 6 mm. long, closely sessile or occasionally very short-pedicelled, densely to thinly pilose, often becoming glabrate and brownish in age; styles scarcely evident or 0.1—0.3 em. long; stigmas 0.5—0.7 mm. long, divided and reflexed. For convenience, the brief description and discussion given by Rowlee for his new variety are quoted here. “One to three meters high, wood soft: leaves much larger, 10- 12 cm. long, 1 em. wide, densely white tomentose on both sides, largest remotely denticulate: aments larger, cylindrical, 4—5 cm. long, otherwise as in the type.” “S. integrifolia var. leucodendroides is a very striking form from southern California collected by Mr. S. B. Parish, nos. 2134, 2040, and 640. There does not seem to be enough difference to war- rant its separation as a species although intergrading forms are wanting.” The above description and discussion of variety leucodendroides leave much to be desired. First, Rowlee redescribed the common California long-leaf willow, Salix Hindsiana Bentham, under the name S. macrostachya Nuttall, a plant of the Columbia River Val- ley, and cited several California specimens. Then he asserted that, from the description, Bentham’s California plant is the same as S. argophylla Nuttall, another plant from the Columbia Valley. The facts are that S. macrostachya is a synonym of 8S. argophylla and that S. Hindsiana Bentham is a good species, confined to Cali- fornia and adjacent Oregon and much more closely related to S. sessillifolia Nuttall, of Oregon and Washington, than to S. argo- phylla. Finally, the leaf size given by Rowlee is much larger than the average and denticulation is not confined to the larger leaves but is universal. The result was to confuse readers as to the characters and relationships of his variety. As the plant Rowlee held to be S. macrostachya really is S. Hindsiana Bentham, the pres- ent combination merely accomplishes what Rowlee thought he was doing. His reference to a species, S. integrifolia, in the discussion above, is wholly without meaning. It seems probable that Row- lee contemplated renaming Bentham’s plant, which has entire leaves, but later decided it was Nuttall’s 8S. macrostachya. Rowlee obviously did not designate a type, although he cited three collections by Parish (see above). The writer has seen a specimen of number 2040 in the herbarium of the University of California (sheet 55027). On the label, the word “Type” has 234 MADRONO [Vol. 6 been written in Rowlee’s own hand. There are two specimens of number 640 in the United States National Herbarium (sheets 780021, 940755). They are var. leucodendroides but there is no evidence that Rowlee ever saw them. Variety leucodendroides occurs sparingly in at least five coun- ties in the Coast Range and central basin of California north of the Tehachapi, where the species is common. It occurs more abundantly in all of the counties south of this divide, where the species is much less common. Several specimens from Humboldt County, in the northwestern part of the state, have been referred to this variety but they need further study and are not cited here. In the western part of southern California this variety is the dominant representative of section Longifoliae, the long-leaf or sandbar willows. To the eastward it gradually is replaced by varieties of Salix exigua Nuttall. In the northern edge of its southern range it overlaps the range of the species. On the south it extends into Lower California and on the east-into the edge of Arizona. Its altitudinal range is from approximately sea level along the Colorado River and the southern coast to elevations of approximately five thousand feet in the southern mountains. Because specimens of variety leucodendroides are seldom cor- rectly identified, but usually are found under such names as argo- phylla, argyrophylla, exigua, Hindsiana, longifolia, macrostachya, and sessilifolia, it is desirable to cite the numerous specimens which have been referred to it by the writer: CaLIFoRNIA (counties from north to south). Santa Cuara County. Upper east fork, Coyote Creek, W. R. Dudley 4207 (CAS). San Benito County. Creek east of Lookout Mountain, altitude 3300 feet, Hall 9926 (USN); The Pinnacles, Eastwood 6750 (CAS); San Juan, Elmer 4908 (CAS, USN). Monterey County. Nacimiento River, Brewer 544 (USN). Tuareg County. Kern River, Peppermint Valley, altitude 4800 feet, Dudley 779 (SU); Three Rivers, near Britton’s, June 15, 1902, Dudley (SU). Kern County. Bakersfield, Piper 6406 (USN), E. A. McGregor 13 (SU); Santa Fe Railroad, west of Bakersfield, Heller 7591 (SU, 2 sheets; UC). Ventura County. Santa Ynez Mountains: Mati- lija Canyon, 6.5 km. below Matilija Hot Springs, altitude 270-300 meters, Fosberg 7423, 7425 (CRB, 2 sheets each; USN, UC); Ma- tilija Canyon, Ojai Valley, altitude 270 meters, Mrs. H. P. Bracelin 6338-686 (CRB, 3 sheets of each; USN, UC); Shady (?) Canyon near Ojai, altitude 600 feet, May 22, 1866, S. F. Peckham; Sespe, F. W. Hubby 134, 135 (label reads “Santa Barbara Co.) (SU); Sespe Canyon, September, 1914, B. W. Everman (CAS); Piru Creek, 10 miles above Piru, Ralph Hoffman 354 (CRB); Piru Creek, 5 km. above Piru, altitude 270 meters, Santa Barbara Na- tional Forest, Fosberg 7426 (CRB, 4 sheets; USN, 3 sheets) ; east of Piru, altitude 180 meters, Bracelin 629 (CRB, 2 sheets), 630, 631 (CRB, 2 and 8 sheets; USN, UC); Hueneme, April 7, 1902, 1942] BALL: SALIX 235 Burtt-Davy (UC); Oxnard, Patterson Ranch, Burtt-Davy 7630 (UC); delta plain, Santa Clara River, Hoffman 181 (CRB); Ven- tura, along beach, Eastwood 5034, 5035 (CAS). Los ANGELES County. San Gabriel or Sierra Madre Mountains and their south- ern foothills: Arraster, altitude 2750 feet, May 10, 1919, F. W. Peirson (CRB) ; Castaic Creek, below Castaic, Fosberg 7411, 7413 (CRB, USN, UC); Gorman, C. R. §& B. S. Ball 2526 (CRB, 8 sheets; USN, UC); Saugus, Elmer 3650 (USN); Burbank, 1904, J. C. Nevin (SU). San Gabriel Mountains: canyons of Sierra Madre Mountains, May, 1888, Hasse (USN); Little Tujunga Can- yon (near Burbank), P. Parney 233 (CAS); San Gabriel Wash, altitude 700 feet, March 6, 1921, Peirson (CRB); Tujunga Can- yon, altitude 1300 feet, March 30, 1919, Peirson (CRB) ; Tujunga Wash, Stonehurst, San Fernando Valley, Fay A. MacFadden 11047 (CRB); Verdugo Hills, La Tuna Canyon, MacFadden 3069 (SU, UC), 11044 (CRB); west fork of Garapito Creek, altitude 1150 feet, Ewan 4219 (CRB); Puddingstone Canyon, San Jose Hills, Wheeler 1723A, 1723B (CRB). Santa Monica Mountains: be- tween Calabasas and Agoura, Fosberg 5850 (CAS, CRB, 3 sheets; USN, SU, UC, 3 sheets). Los Angeles and vicinity: Elysian Park, George B. Grant 2294 (SU), 1156 (UC); Los Angeles River bot- tom, June, 1883, Hasse (USN), September 9, 1917, F. Grinnell, Jr. (SU) ; El Monte, altitude 300 feet, Johnston 1242 (SU) ; Engle- wood, Abrams 1493 (SU, 2 sheets). Mohave Desert: Lovejoy Dam, Lovejoy Buttes, Peirson 9859 (CRB). San Bernarpino County. Mojave Desert: Cushenberry Canyon, Parish 4931 (SU, on sheet 51351 with S. ezigua); 1.5 miles north of Victorville, altitude 815 meters, Bracelin 597, 598 (CRB, 2 sheets, USN, UC); Helendale (Judson), Mojave River, Bracelin 591 (CRB), 592 (CRB, USN); Hesperia, Mojave River bed, G. I. Mozley 950 (USN). San Bernardino Mountains and foothills: Waterman Canyon, Shaw & Illingsworth 4 (SU) ; mouth of Waterman Canyon, altitude 1500 feet, Parish 11401 (UC); borders of streams, alti- tude 1200 feet, Parish 11763 (UC) ; Keenbrook, Cajon Pass, Parish 4980 (SU); Cajon Pass, Abrams & McGregor 694 (SU). San Gabriel Mountains: Cucamonga Canyon, altitude 38000 feet, Johnston 1241 (SU); Red Hill near Upland, Johnston 1243 (SU). San Antonio Mountains: Prairie Fork of San Gabriel River, alti- tude 5000 feet, Johnston 1685 (SU). San Bernardino and vicin- ity: San Bernardino, P. B. Kennedy 1673 (CAS), Marian L. Camp- bell 45, 46 (CAS); altitude 1000-2500 feet, Parish 4591, 4592 (SU); Santa Ana River, altitude 1000 feet, Parish 4786, 4787 (USN, SU), 5197 (SU), Alfred Rehder 158 (CAS); San Bernar- dino Valley, S. B. §& W. F. Parish 640 (USN, 2 sheets; this number cited by Rowlee), altitude 300 meters, Parish 11134 (UC) ; Colton, May 20, 1882, M. E. Jones (CAS, CRB, UC) ; Chemehuevis Valley, Jepson 5208 (SU). Oraner County. Los Alamitos: July 20, 1908, I. J. Condit (UC, 2 sheets); Bixby Avenue, west of Hansen 236 MADRONO [Vol. 6 Road, C. R. Wolf 3843, 3845 (CRB, USN, UC). Santa Ana River: Santa Ana, Helen D. Geis 553, 554 (SU) ; Santa Ana Canyon, alti- tude 500 feet, J. T. Howell 2440 (CAS, 2 sheets), altitude 120 meters, Wolf 2953, 2954 (CRB, USN, UC). Riversipe Counry. Riverside and vicinity: Santa Ana River near Riverside, May 20, 1888, Parish 2040 (type?) (UC), H. DeForest 3 (CRB); Santa Ana River near Corona, Crawford and Johnston 1244 (SU); Santa Ana River, altitude 500 feet, Peirson 4252 (CRB); Santa Ana River, 4.8 km. north of Arlington, altitude 240 meters, Bracelin 599, 602, 604 (CRB, USN, UC), 605, 606 (CRB, 2 sheets each). San Jacinto, June, 1921, Ethel H. Campbell (CAS); San Jacinto Mountains, east base, along borders of Colorado Desert, Hall 2105 (SU, UC); San Jacinto Valley, June, 1897, George F. Reinhardt (UC); San Jacinto River Canyon, Oak Lodge, altitude 3000 feet, Parish 11702 (UC). Colorado Desert, Thousand-Palm Canyon, DeForest 2 (CRB). San Dirco County. Mountain Spring (International Boundary Commission, United States and Mexico), Edgar A. Mearns 3040 (USN, SU); near Tia Juana River, Tia Juana, August, 1902, A. C. Herre (SU), Abrams 3485 (SU); near Tia Juana, June, 1895, S. G. Stokes (SU); San Diego River, San Diego, Abrams 3419 (SU); Old Town, Bracelin 620-623, 625-628 (CRB, 1 to 8 sheets each; USN, except last 3; UC); flats of San Luis Rey River, west of the Mission, Wiggins 3034 (SU; UC, 2 sheets) ; Jacumba Valley, Abrams 3679 (SU); Laguna Mountains, Eastwood 9253 (CAS) ; Lakeside, Grant 6860 (SU) ; Oneonta, alti- tude 25 (?) feet, H. P. Chandler 5116 (SU); Warner’s Hot Springs, Eastwood 2822 (CAS). Imperial County. Colorado River bottoms, 10 miles from Yuma, Arizona, Roxana S. Ferris 1030 (SU, 2 sheets). Mexico. Baja California, near Tia Juana, M. E. Jones 3730 (CAS). Satix Hinpstana var. Parishiana (Rowlee) comb. nov. Salix Parishiana Rowlee, Bull. Torr. Bot. Club 27: 249, pl. 9, fig. 3. 1900; Abrams, Fl. Los Angeles, suppl. ed., 101. 1911; Schneider, Bot. Gaz. 67: 323-325. 1919, Jour. Arnold Arb. 8: 65, 92, 98. 1921; Ball in Abrams, Illus. Fl. Pac. States 1: 492, fig. 1198. 19238. 8S. exigua var. Parishiana (Rowlee) Jepson, Man. FI. Pl. Calif., 264. 19283, Rowlee drew a fairly adequate description of his new species, Salix Parishiana, when it is considered that the foliage and aments of the type were not yet fully developed. For convenience of discussion and comparison it is reproduced here: ‘““A slender shrub, one to three meters high, bark grayish or brown, young twigs cinereous strigose: leaves linear-lanceolate, minutely and remotely denticulate, 5-7 cm. long by 3 mm. wide, silky canescent when young, glabrous and somewhat coriaceous when mature, veins few but very prominent: stipules none: aments on long leafy peduncles, appearing about April 1, 2-3 cm. 1942] BALL: SALIX 237 long by 1-2 (sic) cm. peduncles often 10 cm. long, the upper leaves of the branch much surpassing the ament: ament densely flowered, scales white densely villous all over, oblong, acute: fila- ments scantly (sic) hairy at the base: capsules densely villous, oblong, closely sessile: style distinct: stigmas linear, three times as long as thick. “A very peculiar form, differing from S. taxifolia by its larger leaves and cylindrical aments and quite distinct from other spe- cies with linear stigmas. “Cartirornia: Matilija Cafion, San Bernardino Co. (F. W. Hobby (sic), nos. 54, 55), Springs Valley, Inyo Co. (F. V. Coville and F. Funston, no. 263).”’ Had Rowlee studied the more mature material, with the con- sequent larger and more evidently denticulate leaves and larger aments, he scarcely would have compared his species with S. taxifolia alone. Nor would he have stated so positively that it was “quite distinct from other species with linear stigmas.” The type, as so frequently is the case, represents an extreme form of the entity. Certain characters assigned by Rowlee, such as the glabrous- ness, linear-lanceolate shape, and veininess of the mature leaves, must have been observed in the collection by Coville and Funston from Inyo County (no. 263), as they are not exhibited by the type specimens. Number 263 probably is a desert form of S. ezxigua, as suggested by Schneider, who in turn considered S. Parishiana prob- ably to be intermediate between S. ezigua as it occurs in southern California and S. sessilifolia var. leucodendroides. Through the courtesy of Dr. A. J. Eames and the late Dr. K. M. Wiegand, the types of Salix Parishiana Rowlee were made available to the writer from the herbarium of Cornell University. Both types, male and female, are mounted on one sheet. The label reads as follows: “S. longifolia, var. argyrophylla And., Pistillate fl., Cliff Glen; staminate fl., Ojai Hot Spgs., Matilija Cafion, Sta. Barbara Co., F. W. Hubby, No. 54, April 3, 1896.” In the upper right corner are pencilled the words: “S. parishiana n. sp. W. W. R.”. A second sheet bears a single more nearly mature pistillate specimen and a label reading: “Salix longifolia var. argyrophylla as to leaf characters; S. sessilifolia var. hindsi- ana as to style and stigma. Matilija Cafion, Kennedy’s, Sta. Bar- bara Co., F. W. Hubby, No. 55, April 19, 1896.” It has the same pencilled annotation as the first label, and both annotations are in Rowlee’s handwriting. The first cited collection by Hubby (no. 54) consists of a male (type) shoot 38 centimeters long with a half dozen aments, and a female (type) shoot 30 centimeters long with some five aments. The second cited collection by Hubby (no. 55) is a single pistillate shoot about 24 centimeters long, with two aments. On each of the herbarium sheets is the inked anno- tation “S. parishiana n. sp., WWR.” in Rowlee’s hand. 238 MADRONO [Vol. 6 From these three specimens, all from Matilija Cafion, and all annotated by Rowlee and cited with his original description, it is possible to give the following emended description. Aspect gray or silvery-gray; seasonal branchlets puberulent to pubescent, those of the first year glabrate to puberulent; leaves subpetiolate, exstipulate, blades linear (not linear-lanceolate), 4—7 or 8 cm. long, 2—3.5 mm. wide, common sizes 5 X 2.5, 6 x 2.5— 3.5, 7X 2-8, 8 x 8.5, acute at base and apex, margins somewhat revolute, remotely and minutely denticulate, the midrib and primary veins slightly raised (not ‘very prominent’) on the gray to silvery puberulent upper surface, the lower surface silvery pubescent (leaves immature and therefore not ‘glabrous and coriaceous when mature’); pistillate peduncle 2-8 cm. long in flower to 4 cm. long in fruit, the staminate 7—10 cm. long in flower, each bearing 8—10 foliage leaves; pistillate aments 2 cm. long in flower, 3 cm. long, 1 cm. wide in fruit (not ‘1-2 cm.’) ; capsule (no. 55, nearly mature) narrowly lanceolate (not ‘oblong’), 4.5—5 mm. long, sessile, pilose, style evident but very short, 0.1—0.2 mm. long, stigmas linear-oblong, 0.5-0.6 mm. long, divided, spreading; flower scales broadly elliptical or elliptical-oblanceolate, 2—2.5 mm. long, thinly pilose-pubescent or subglabrate on the outside, more pilose-pubescent within (not ‘densely villous all over’) ; staminate aments about 2 cm. long and 0.5 cm. wide; stamens two, filaments free or united only at the extreme base, pilose with crinkly hairs on the lower half or two-thirds; ament scales as in the pistillate ament. Variety Parishiana apparently is confined to the southern coastal district of California and occurs chiefly in the mountains from relatively low elevations to five thousand feet above sea level. Present material indicates a range from the Pinnacles in San Benito County to San Diego County at the international boundary. Specimens of the variety are found in herbaria under the names of various species of section Longifoliae, as argophylla, exigua, Hindsiana, longifolia, macrostachya, sessilifolia and their varieties. CatirorNiaA. San Benito County. Stream bank, Pinnacles, J.T. Howell 4620 (CAS); Bear Valley, Pinnacles, Chester Dudley 6 (CAS). Ventura County. (Matilija Canyon is a tributary of the Ventura River, whereas all other streams mentioned are part of the Santa Clara River system.) Matilija Canyon, Cliff Glen (male type), Ojai Hot Springs (female type), April 3, 1896, F. W. Hubby 54; Kennedy’s, April 19, 1896, Hubby 55 (types, Cornell University; photographs, CRB, UC); Matilija Canyon, 6.5 km. below Matilija Hot Springs, Santa Ynez Mountains, altitude 270-— 300 meters, Fosberg 7424 (CRB, 2 sheets; USN, UC) ; Sespe Creek (between Sulphur and Pine mountains), near Ten-Sycamore Flat, altitude 2300-2500 feet, Abrams & McGregor 169 (SU, male and female; leaves 5—7 cm. x 8-4 mm., style 0.2—0.4 mm. long, stigmas 1942] NOTES AND NEWS 239 1 mm. long); Mount Pinos (near center of northern boundary of Ventura County, its eastern flank drained by Lockwood Creek, a tributary of Piru Creek): Lockwood Creek, below Snedden’s, Dudley § Lamb 4632 (SU, bearing cone galls); Goodenough Meadow, Dudley 5 Lamb 4717 (SU, leaves on fruiting branchlet 4—7.5 cm. X 2-3.5 mm., style 0.1-0.2 mm. long; UC, “‘near Lock- wood Valley Schoolhouse, June 26’’) ; Lockwood Creek, June 5, 1930, Hoffman (CAS); Seymour Creek, altitude 5300 feet, Hall 6343 (UC); 3.2 km. east of Piru, altitude 180 m., Bracelin 632 (CRB, 2 sheets; USN, UC). Los Aneetes County. San Fran- cisquito Canyon, Parish 1984 (UC); San Antonio Mountains, Prairie Fork of San Gabriel River, altitude 5000 feet, Johnston 1685 (UC, sterile, leaves shorter and broader than normal) ; near Camp Rock Creek, Pinyon Ridge, San Gabriel Mountains, altitude 4500 feet, Peirson 716 (CRB). Oranese County. Los Alamitos, July 20, 1908, Condit (UC). San Disco County. Tia Juana, Eastwood 2926 (CAS). United States Department of Agriculture, Washington, D. C., October, 1941. NOTES AND NEWS The members of the University of California Expedition to El Salvador, under the capable and energetic leadership of Dr. R. A. Stirton of the Department of Paleontology, returned to the United States on May 25, 1942 after nearly six months of success- ful work in El Salvador. The party was hospitably and graciously received wherever it went. Two men, in particular, were of con- stant and invaluable assistance,—Dr. Mario Lewy of the Depart- ment of Agriculture of El Salvador and Mr. G. A. Swanquist of San Miguel. The personnel was as follows: Mr. John Davis, herpetologist ; Mr. William K. Gealey, geologist; Mr. Nathan Geer, cook and assistant paleontologist; Mr. Milton Hildebrand, mammalogist ; Mr. Joe T. Marshall, ornithologist; Dr. R. A. Stirton, paleontol- ogist; and Mr. John M. Tucker, botanist, representing the Her- barium of the University of California. PROCEEDINGS OF THE CALIFORNIA BOTANICAL SOCIETY February 21,1942. The annual dinner meeting of the Society was held at the Berkeley Women’s City Club on Saturday evening. About fifty members and guests were present. Dr. Alva R. Davis, President, acted as toastmaster and introduced with felic- ity the speaker of the evening, the distinguished mycologist, Dr. A. H. Reginald Buller, Professor Emeritus of Botany, University of Manitoba, and Hitchcock Professor, University of California, | 1942. Dr. Buller discussed the ink fungi—species of the distinc- 240 MADRONO [Vol. 6 tive genus Coprinus (Pers.) Fr.—and their organization, in a lucid and thoroughly interesting manner. The lecture was abun- dantly illustrated by slides from excellent field photographs and from Dr. Buller’s drawings. The arrangements for the dinner were made by Dr. G. Ledyard Stebbins, Jr., Chairman of the Program Committee. The unique mycological table decorations were collected and arranged by Miss Beryl Schreiber, Mrs. Vera Miller, and Mrs. Lincoln Constance. Relatively few members liv- ing at a distance from Berkeley attended the banquet. The exist- ing state of war and the consequent danger and inconvenience of possible “blackouts” are known to have lowered attendance. March 19, 1942. Speaker: Mr. M. W. Talbot, California Forest and Range Experiment Station. Subject: Guayule and other western American rubber plants. The speaker lucidly de- scribed the Government’s guayule (Parthenium argentatum) plan- tation project now being developed in the Salinas Valley. The first objective of the project is to prepare a 700-acre nursery, and to plant all available seed of selected high-yielding guayule strains. When cut at a normally economic age the best strains of guayule yield from 1200 to 2000 pounds of rubber per acre. It is expected that in two years the present program will lead to production of some 56,000 tons of rubber, and a great amount of seed of selected strains. Even before the supply of Hevea plan- tation rubber was largely cut off, guayule rubber was approaching normal economy. Following the talk a lively discussion devel- oped with Mr. Talbot, Dr. Fred E. Foxworthy, retired Malayan forester; Dr. Trumbull of the Goodrich Rubber Company; Dr. D. T. MacDougall of Carnegie Institution of Washington, Dr. A. R. Davis, and others participating. The meeting was attended by about ninety members and guests. April 16, 1942. The “Annual Living Plant and Specimens Meeting” was held under the direction of Dr. G. Ledyard Steb- bins, Jr. Prior to the presentation and inspection of specimens, Dr. A. R. Davis presided over a short business meeting. Many interesting native and exotic plants and plant specimens were exhibited and described by a score or so of the Society's members, including Dr. G. L. Stebbins, Jr., Miss E. E. Morse, Prof. W. W. Mackie, Prof. H. W. Shepherd, Prof. H. E.. McMinn, Dr. H. I: Mason, Mr. L. L. Edmunds, Dr. L. Constance, and others. May 21, 1942. Speaker: Mr. C. R. Quick, Division of Plant Disease Control, United States Department of Agriculture. Sub- ject: Certain methods of forcing seed germination. The speaker presented a classification of the difficulties which may be encoun- tered in the germination of seeds, and discussed methods of obviating the several types of difficulties enumerated. Data representing the successful growth of hard-to-germinate seeds were presented on lantern slides. About 25 members and friends of the Society attended. C.R. Quick, Secretary. 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 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 aes a3 hi \ \ ’ s ‘ yt ; 4 ie t ne 5 Ne ( ‘ aus i y D A vyre Ii? ‘ % * ‘ i My NUMBER 8 MADRONO A WEST AMERICAN JOURNAL OF BOTANY meee a xs ee Se Por Oe OCS ANTAL Ti Ras QS INT UES HLS fp RN, 3 ‘Ww BOR SG NURT Ma. Contents THe GENETIC APPROACH TO PROBLEMS OF RARE AND ENDEMIC SPECIEs, GOUT SUCOOLUS) FT ee Oe a ciel Sale Me Sane te os Bins eke Wie wns 241 Tue Type or CLEMATIS HiIRSUTISSIMA, Ralph O. Hrickson .............. 259 Francis Ramatey, Hdna Lowise Johnson ............ 0.0... eee 260 Review: Thomas H. Kearney and Robert H. Peebles, The Flowering Pionts of Arizona (Lyman Benson) ...... ee oe oe ee 265 NDE TOMNVOLUME NEU. oi ak cele se ee ee Et ORGY rater UN) ann fata cine, ® 267 Published at North Queen Street and McGovern Avenue, Lancaster, Pennsylvania October, 1942 MADRONO A WEST AMERICAN JOURNAL OF BOTANY Board of Editors Hersert L. Mason, University of California, Berkeley, Chairman. LeRoy Asrams, Stanford University, California. Epcar Anperson, Missouri Botanical Garden, St. Louis. Lyman Benson, University of Arizona, Tucson. Hersert F. Corpetann, Sacramento Junior College, Sacramento, California. Ivan M. Jonnston, Arnold Arboretum, Jamaica Plain, Massachusetts. Mixprep E. Maruias, University of California, Berkeley. Bassett Macurire, Utah State Agricultural College, Logan. Marion Ownsey, State College of Washington, Pullman. Secretary, Editorial Board—Etne.t Crum Department of Botany, University of California, Berkeley Business Manager—Witi1am HIesey 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. 1942] STEBBINS: GENETICS AND RARE SPECIES 241 THE GENETIC APPROACH TO PROBLEMS OF RARE AND ENDEMIC SPECIES G. Lepyarp STEBBINS, JR. One of the questions that every field botanist with an inquir- ing mind is bound to ask is: Why are some plant species wide- spread and common, while others are rare and local? The prob- lem of rare species has a twofold fascination; their discovery never fails to provide a thrill, while the analysis of their affinities and distribution often gives valuable clues to the history of floras. It is natural, therefore, that many botanists have given their answer to this question, and that these answers have been as di- verse as are the minds of their proponents. Among these answers there have recently appeared a series which has emphasized the genetic constitution of the species involved. The object of the present article is to review the available evidence upon which these concepts are based, to suggest ways in which new experi- mental evidence for them may be obtained, and to follow out some of their implications when applied to problems of the history of floras and plant evolution. The word “rare” may not always mean the same thing. Some plants are regarded as rare because throughout a large part of their range they are found only as scattered individuals or small groups, separated by miles from their nearest neighbors. Such is the case with many species of orchids, such as Calypso bulbosa (L.) Oakes, Cephalanthera Austinae (Gray) Hel., Cypripedium arietinum R. Br. and Aplectrum hyemale (Muhl.) Torr. In most of these cases, however, there are some regions where the species concerned are abundant. Calypso, for instance, is common enough in the northern Rocky Mountains, as is Cypripedium arietinum in parts of southern Ontario, while the rarity of Aplec- trum is due largely to extermination by man. Another type of rarity is extreme localization. A species may occur in only a few widely separated localities, but may be abundant enough where it is found. This is notably true of Phyllitis Scolopendrium (L.) Newm. var. americana Fernald, the hart’s tongue fern in eastern North America. Many species of serpentine barrens in Cali- fornia, such as Cupressus spp. and Streptanthus spp. are similarly distributed. Still a third type of rareness is extreme endemism. A species may occur only in one or two spots on the entire globe, but in this case it is almost always represented in these spots by hundreds of individuals. These three types are, of course, con- nected by innumerable intermediate cases. In the _ writer’s opinion, the concepts set forth below will apply with modifications to all of them. As a necessary background for this study, let us review briefly the most widely current answers to this question of why certain Manproxo, Vol. 6, pp. 241-272. October 23, 1942. 242 MADRONO [Vol. 6 species are rare. Perhaps the most direct and simple answer is that of Willis (34, 35), who maintains that in general rare and endemic species are beginners, which have not yet had time to spread. The weaknesses and fallacies of this hypothesis have been fully exposed by Fernald (16), Wright (38), and Hubbs (23), so that they need not be dwelt upon here. It will be pointed out below that our present concepts of the genetic structure of species, which have been developed as a result of many painstak- ing experiments, throw into glaring relief the fallacies of Willis’s reasoning. In addition, recent paleobotanical research has added greatly to the number of rare modern species whose fossil ances- tors are known to have been common and widespread, and this is particularly true of the endemics of the California flora (Chaney, 7, Axelrod 2,3). . A second answer was given by Fernald (15, 17, 18), as a result of his keen observations in the field and his careful analysis of the distribution of many rare species in the flora of eastern North America. This is the concept of senescence; that most rare spe- cies were once common, but that their great age and the vicissi- tudes to which they have been subjected have made them ‘“con- servatives,’ and unable to spread. This concept, based as it is upon extensive observations of rare plants as they actually grow in the field, has much to recommend it. Most field botanists will agree with Professor Fernald that conservatism rather than aggressiveness is characteristic of rare plants. In fact, the genetic concept to be reviewed below is based primarily upon this assumption. The weakness, however, of the concept of senes- cence is the implication that conservatism results directly from the age of a species. There are two large objections to this implica- tion. In the first place a number of species, such as Sassafras variifolium (Salisb.) Ktze. Liquidambar styraciflua L. and Ulmus americana L. are known to have close relatives that go far back into the fossil record, and yet the present species are still wide- spread and common, having invaded much of the region that was covered by the Pleistocene ice sheet. The other, and perhaps -more serious objection is that the same species may be rare and conservative in one part of its range and common and aggressive in another. LErigeron compositus Pursh is cited by Fernald (15) as one of the “‘senescent”’ species composing the relict flora of the Gaspé Peninsula. In the Sierra Nevada of California, and pre- sumably also in the Rocky Mountains, this species is far from conservative. The variety of habitats which it occupies is matched by the morphological variability of the species itself. Adenocaulon bicolor Hook. was considered a “senescent” species (Fernald 18) on the basis of the disrupted range and obvious great age of the genus and the rarity of A. bicolor in the Great Lakes region. In California this species grows under redwoods, as Fernald has pointed out, but it is also common under Pseudo- tsuga, Abies, Pinus ponderosa and other conifers. In the Sierra 1942] STEBBINS: GENETICS AND RARE SPECIES 243 Nevada at middle altitudes, however, Adenocaulon is far from con- servative. It is one of the commonest and most aggressive weeds about cabins, being often the first species to occupy disturbed ground, if sufficiently shaded. From the hypothesis of senes- cence one would be forced to conclude that Erigeron compositus, Adenocaulon bicolor, and similar species are old in the east and young in the west. This conclusion seems illogical in the ex- treme. And in one genus, Antennaria, there is direct evidence that the conservative, “senescent” species of the Gulf of St. Lawrence area are actually younger than their common, widespread western relatives. With one exception these Gaspé and Newfoundland antennarias are exclusively apomictic; staminate plants are un- known in them. They therefore are “dead ends” from an evolu- tionary point of view, and must have originated from sexually | reproducing species (cf. Stebbins, 29). Their only close sexual relatives, and therefore their presumable ancestors (A. umbrinella Rydb., A. microphylla Rydb., A. reflexa Nels., A. media Greene, A. monocephala T. & G., ete.) all occur in western North America, and are for the most part widespread, common, and aggressive _ enough to have colonized extensively areas vacated by the Pleisto- cene glaciers. Therefore, the conservatism of the relict Anten- naria species cannot be due to age alone, since their ancestors have still retained “youthful” characteristics. The third answer to this problem of rare species is the genetic concept which is to be reviewed in the present paper. It is based upon the realization, as a result of the experiments of Turesson and others (cf. Turesson, 31; Hiesey, 22; Clausen, Keck and Hiesey, 8), that most widespread and common plant species con- sist of a large number of genetically different biotypes, many of which differ widely in their ecological preferences. This is, of course, the basis of the ecotype concept, which conceives of these widespread species as consisting of several clusters of similar bio- types, each cluster, or ecotype differing from other ecotypes in its ecological preferences. On the basis of this concept, the range of ecological tolerance of a species, in the sense of Good (20), embraces the tolerance ranges of all of its component ecotypes and biotypes. Naturally, therefore, a species with many eco- types and biotypes will be widespread and common. And con- versely, a species which is poor in biotypes, and has only one ecotype, will be rare, unless its individual biotypes have a wide range of ecological tolerance, or unless the particular conditions to which they are adapted are widespread. A rare species, there- fore, may be conceived of in genetic terms as one poor in biotypes, and with its biotypes so specialized that they can grow and com- pete with other species in only a limited area. Aggressiveness, or the ability of a species to colonize new areas, and to crowd out other species, is the result of the possession of a great store of genetic variability either evident or concealed. This consists of genetic heterozygosity, of biotypes preadapted to new conditions 244 MADRONO [Vol. 6 which the species might encounter, or of a rapid mutation rate, by which new biotypes may be produced. A species is conservative, on the other hand, if it contains few biotypes, most of which are homozygous or nearly so, and has a low mutation rate. This concept was foreshadowed by Darwin’s classic statement that “wide ranging, much diffused, and common species, vary most.’ It was hinted at some time ago by Turesson (31), but so far as this writer is aware was first clearly stated by Anderson (1, p. 496). Hultén (24) made it the cornerstone of his brilliant analysis of the history of the Arctic flora, while Camp (6) used it to explain the relative constancy and limited distribution of some species of the interesting genus Befaria. Cain (5) pointed out the advantages of this concept over that of senescence, while Raup (28) recognized it as an important factor in the distribution of species of boreal America. Fassett (14) made the determina- tion of genetic constancy in certain areas a major objective of his interesting and valuable study of variation in Rubus parviflorus. The main difference between this genetic concept and that of senescence is that it aims to interpret the rarity of species prima- rily as a result of their present constitution, without implying any- thing about the past history or future fate of the species con- cerned. Many, and perhaps most rare species were once more common and aggressive, but not all. The phenomenon of insular species, many of which have always been rare, will be discussed below. The rare conservative species which were once common have been characterized by Turesson (31) and Hultén (24) as having been ‘“‘depauperated with regard to their biotype contents” (Turesson 31, p. 97). Since the word depauperate is generally applied to plants of small size, its use in the present sense seems inadvisable. The word depleted expresses the situation more precisely and has no other connotation. To those who accept this genetic hypothesis, therefore, the writer suggests that the term “depleted” be used for those rare, conservative species which appear to have been formerly more common and aggressive; i.e., the “‘senescent”’ species of other authors. Griggs (21) has recently sought to explain the rarity of plant species on the basis of competition. He states that (p. 592) “a species is rare because it cannot compete successfully with the common plants,” and that “most rare species find their habitats in the early stages of the ecological succession.” These statements are supported by a wealth of evidence derived from a study of rare plants in eastern North America. They lead to a conclusion similar to that implied by the term senescence, namely that those rare species which have ranges at present disrupted, but formerly continuous, “are therefore slowly dying out.’ On the basis of this hypothesis Griggs admittedly has difficulty in explaining the fact that many of the plants which are rare in eastern North America are common in the west. And if one examines the plants which are rare in western America, particularly those of 1942] STEBBINS: GENETICS AND RARE SPECIES 245 California, one finds that Griggs’ hypothesis does not apply to a large number of them. The most famous rare species in Cali- fornia is the big tree, Sequoiadendron giganteum (Lindl.) Buchholz. Others, almost equally famous to botanists, are Cupressus macro- carpa, Pinus Torreyana Parry and P. radiata Don, Picea Breweriana Wats., Abies venusta (Dougl.) Koch, Quercus Sadleriana R. Br., Crossosoma californicum Nutt. and Lyonothamnus floribundus Gray. None of these species can be said to “find their habitats in the early stages of ecological succession.” They are sub-climax, climax, or post-climax types. Furthermore, such observations as have been made indicate that in restricted areas and under certain conditions these species can compete very well with their common associates. Mr. Woodbridge Metcalf of the Division of Forestry, University of California (unpubl. bulletins and oral comm.), has found that seedlings of the big-tree may under certain conditions become established in great numbers. Once established, they grow very rapidly and in one forest, started through natural re- seeding in the early eighties, ‘“‘none of the associated species have been able to keep pace with the sequoias in height, though there are some excellent specimens of sugar pine, Pinus Lambertiana, and white fir, Abies concolor, in situations where they have not been too much crowded by the big-trees.”’ Apparently the limits to the spread of this most famous of rare plants, are the specialized conditions necessary for the successful establishment of seedlings. These are chiefly a disturbed mineral soil, and a sufficiently early onset of the fall rains during the early years of growth. In these respects the seedlings of the common species of Sierran trees are much less particular. Another rare Californian, Pinus radiata, the Monterey pine, is a very good competitor in the regions where it grows naturally. Last spring the writer led a class through a clearing in a grove at the northernmost of its three natural locali- ties, Ano Nuevo Point. Although this clearing had gone over completely to grassland (the predominant plant formation for miles along the coast both north and south of the four mile stretch of pine forest) it was filled with vigorously growing pine seed- lings, which will soon crowd out the grass, and restore the area to its natural cover of pines. Furthermore, there were abundant seedlings of P. radiata throughout the stand, and in some places beyond its edges, so that one could not possibly draw the infer- ence that the species is dying out. Mr. H. A. Jensen, of the California Forest Experiment Station has informed the writer that the southernmost grove of P. radiata, at Cambria, was once extensively lumbered, and has since restored itself. Hence neither the statement that rare species occupy chiefly pioneer habitats nor that they are slowly dying out applies to the most famous of Californian rarities. Griggs’ hypothesis, however, still is of great value in inter- preting many of the rare plants of eastern North America. Furthermore his emphasis upon ability to compete as a major 246 MADRONO [Vol. 6 factor in the distribution of both rare and common plants is fully justified and is an important part of the concept of genetic homo- geneity as here presented. This seems evident from the writer’s preliminary observations of one of the most interesting endemics of the San Francisco Bay region, Dirca occidentalis Gray. This species is restricted to an area about ninety miles long and twelve miles broad, being most abundant in the Oakland and Berkeley hills. Its nearest relative, which it resembles rather closely, is the wide-spread eastern American D. palustris L. (fig. 1). In con- trast to the swamp habitat of the eastern species, D. occidentalis occurs principally upon well-drained hill slopes, where its chief competitors are other shrubs, such as Tozicodendron diversilobum e f 3) D. palustris BBD. occidentalis Fic. 1. Ranges of Dirca palustris and D. occidentalis. (T. & G.) Greene, Baccharis pilularis DC., Rhamnus californica Esch., and Osmaronia cerasiformis (T. & G.) Greene. In Wildcat Canyon, just east of Berkeley and still within the summer fog belt, Dirca is rather common, and in a few places forms ‘almost pure stands. Here the writer has observed several clearings in sheltered north and east facing slopes, where Dirca seedlings were more abundant than those of any other shrubs, and were compet- ing on equal terms with Towicodendron, and doing better than Baccharis or Rhamnus. On such slopes young Dirca seedlings can be found everywhere under the other shrubs, so that there seems no more reason to suppose that it is dying out than that Rhamnus, Osmaronia, or Symphoricarpus albus are disappearing from this 1942] STEBBINS: GENETICS AND RARE SPECIES 24:7 area. On the sunnier west and south facing slopes, however, adult Dirca shrubs are sometimes found, but no seedlings have been observed. Towxicodendron and Baccharis, on the other hand, are equally vigorous and self-perpetuating in both sites. If one travels two miles east from Wildcat Canyon, crossing a ridge 1500-2000 feet high, one reaches the inner edge of the fog belt, where the summer weather is considerably drier and hotter. Here Dirca is rather local, and occurs only in shade. The only extensive stand seen by the writer was in dense shade under a grove of live oaks (Quercus agrifolia Nee), a habitat which it never occupies in Wildcat Canyon. Here it was accompanied, as usual, ty S. canus g HS. antennariifolius ey ope Se a eT m8) 8! Colse: pe) sw 1 OP ch ite ee ec Car) ee 0 8 FS eae . ce Cec ae We om ed a eeeey fede) viewet Saye cea s i8 Pee eh te) fe) jel ie) 20% 8 2 ae ce OR E Vme gare | io: Meh usr 2 a Terk iay 2° Pearce, we ce 6 . ° '. o os ih Der Due tae Ce ako Beeoe Ree ere ML . a ° . o 8 oe ° 9 o: et eel. ° . Chl Bad 5 0 ° a ° ° rae ge Oo - Oo ° ° . .? ° ‘ Piianc: Fic. 2. Ranges (generalized) of Senecio canus (sens. lat.) and S. anten- narufolius. by poison oak (Tozicodendron), but the latter species was also abundant on the open, sunny slopes away from the oaks. Going eastward from Orinda, one would have to travel 1500 miles before he would see Dirca again. The logical inference from these ob- servations is that D. occidentalis contains only a few biotypes, with a limited range of tolerance. These are successful only in sheltered spots, and may require the rather heavy type of soil characteristic of the Berkeley Hills. In the fog belt, they are only moderately tolerant of shade, while in the warmer regions east of the fog the seedlings can compete successfully only in deep shade, where they are met and surpassed by the more vigorous 248 MADRONO [ Vol. 6 competition of such undershrubs as Rhus, Osmaronia, Symphori- carpus albus (L.) Blake and Corylus rostrata Ait. var. californica A. DC. These more common competitors, on the other hand, appear to possess biotypes which can withstand a variety of sun and shade conditions in both Wildcat Canyon and Orinda. The above argument may be summed up by a definition of the concept of genetic homogeneity, as follows. In continental areas, most rare or narrowly endemic species are genetically homogene- | ous, and may therefore be termed homogenic. They consist of relatively few biotypes which are themselves relatively homo- zygous. In contrast, the common and widespread species are usually heterogeneous in their genetic makeup. They include many biotypes, often grouped into more or less distinct ecotypes, and a large proportion of their individuals are genetically hetero- zygous, possessing a store of genetic variability beyond that which is evident from the appearance of their phenotypes. This concept of genetic homogeneity, however plausible it may seem, is nevertheless only a working hypothesis. It should be put to the acid test of experimentation. Two lines of attack occur to the writer. In the first place, the genetic variability of typical rare species should be tested by growing under constant, controlled conditions progenies from all parts of their ranges. This variability should then be compared with that found in a series of progenies of their most common competitors, taken from various parts of the range of the rare species, as well as from beyond that range. Finally, in the case of species that are rare in one part of their range and common in another, or of those rare species that have close relatives elsewhere it should be pos- sible to inject new variability, and therefore new aggressiveness into them by means of hybridization. In other words, if the con- cept of genetic homogeneity as the cause of “conservatism’”’ is valid, wide intervarietal or interspecific crossing should replenish the biotype supply, and make for increased aggressiveness. Some of the new hybrid derivatives should then be easier to estab- lish in new sites than the original rare species. While recognizing that this hypothesis needs experimental confirmation, we can nevertheless follow out some of its implica- tions. In the first place, it is entirely incompatible with the hypothesis of “Age and Area.” The concept of genetic homo- geneity is based partly on the assumption that in terms of the geological time-scale the migration of young species into new territory is rapid or at least fast enough to keep up pretty nearly with the prevailing rates of climatic change. The distribution of a species will be a reflection of its store of genetic variability only if each ecotype can occupy all of the contiguous territory to which it is suited in a relatively short period of time. This assumption, diametrically opposed to Age and Area, was emphasized by Glea- son (19) and explains the facts of distribution as the present writer has observed them. Many species whose seeds have no 1942] STEBBINS: GENETICS AND RARE SPECIES 249 obvious means of rapid dispersal are widespread in new territory. The lupines of California are good examples. On the other hand, many species with apparently excellent means for dispersal are highly restricted. To cite just one instance, there are in Cali- fornia a number of native species of thistle (Cirsium), all with presumably equally efficient methods of seed dispersal. Some of these, Cirsium fontinale (Greene) Jepson, C. campylon Sharsmith, C. Andrewsti (Gray) Jepson, and C. Vaseyi (Gray) Jepson are more or less rare and local. One cannot ascribe this localization to the recent origin of the forms in question and the consequent lack of time for their distribution, since various European thistles, such as Cirsium lanceolatum (L.) Scop., C. arvense Scop., and Silybum marianum Gaertn., with apparently no better methods of seed dispersal than the native species mentioned, have become common in various parts of California within the past hundred years orless. These rare California thistles have failed to spread because they are not adapted to any of the areas adjoining their present ranges. Furthermore, the genetic concept of intraspecific variability offers an entirely different, and in the writer’s opinion more satisfactory, explanation of the facts upon which Willis has based his hypothesis. Willis’s two main lines of evidence are first that endemic species are in general rarer even in the places where they occur than are widespread species in the same area, and second that the ranges of related species often overlap in “chain- mail” fashion, so that at the limits of their ranges they may inter- mingle with each other. The first point, which is borne out by most rare species, is entirely compatible with the concept of the genetic uniformity of rare species. These species are adapted to only a few ecological niches, and these niches are not only re- stricted geographically, but are in general of small extent even in the regions where they do occur. To use a simile: a physician belongs to a widespread and common profession. Not only is there room for physicians in every town in the world, but in addi- tion a city can absorb a large number. A botanist, on the other hand, belong to a profession which is rare and local. There are only a few cities, those which possess a large institution of learn- ing, in which a professional botanist can survive at all, and in these botanists are much rarer than physicians, because there are many fewer places which they can occupy. Similarly a plant species with a narrow range of tolerance will tend to be not only localized geographically but also rare where it does occur. The “chain-mail” pattern of distribution can be explained equally well upon the genetic concept, without resorting to “Age and Area.” Willis argues that if two related species are found together in the same forest one cannot assume that they have different ecological preferences. This may be in part true, al- though one cannot help remarking that every forest or meadow has inequalities, however slight, of topography, exposure, soil, 250 MADRONO [Vol. 6 moisture, etc., which would permit individuals with unques- tionably different ecological preferences to grow near each other. But even if we grant that two groups of individuals belonging to different species have the same genetically conditioned ecological requirements, we need not conclude from this fact that the two species as wholes have the same range of tolerance. The biotypes which are ecologically equivalent may represent opposite ex- tremes of the ranges of genetic variability of the two species. The normal or average biotypes of the two species may be very different from each other. For instance, Pinus ponderosa (sens. lat.) and P. contorta var. Murrayana overlap in the Sierra Nevada in typical “‘chain-mail” fashion, so that forests exist where the two species grow side by side, although in general P. ponderosa grows at lower altitudes and in drier situations than P. contorta var. Murrayana. This overlapping may mean that the hardiest, most moisture tolerant individual biotypes of P. ponderosa are nearly or quite equivalent to the least hardy, most drouth resistant ones of P. contorta var. Murrayana, but it certainly does not mean that the two species as wholes are ecologically equivalent. In fact, the only reasonable conclusion which one can draw from the ranges of these two species, which occur separately over enormous stretches of territory, at very different altitudes and latitudes from each other, is that they have very different ranges of tolerance. And there is no case of “‘chain-mail’ distribution known to the writer to which the same explanation cannot be applied. When “Age and Area” has been eliminated, there remain two possible conditions of the past history of a rare species. One is that the species was once more common, widespread, and richer in biotypes than now, so that its present rarity is due to depletion of the store of genetic variability. The other is that the species never was common, but diverged from a small group of indi- viduals of a widespread ancestral species, following the establish- ment of these individuals upon a small insular area. There are thus two types of homogenic rare species, depleted species and insular species. The evidence from both paleontology and present distribu- tion indicates that depleted species are frequent, and constitute a large proportion if not a majority of rare species (Fernald, 16, Axelrod, 2). The process of depletion has two stages. First, the widespread, common species becomes reduced in geographic distribution and in numbers through climatic or geological changes which eliminate many of its original habitats. During this process many biotypes and ecotypes are automatically de- stroyed, both through the complete elimination of the species from many areas and through more rigid selection in the few regions where it can survive. After this reduction in numbers, the species may still preserve a considerable amount of individual genetic variability, as well as a store of potential variability in ae 1942] STEBBINS: GENETICS AND RARE SPECIES 251 the form of recessive genes for which the individuals are heterozygous. Its continued existence as a series of small, com- pletely isolated populations will, however, automatically lead to the further depletion of each population. As Wright (37) and Dobzhansky (12, p. 834) have pointed out, such small popula- tions become more uniform genetically on account of inbreeding. Recessive genes tend to express themselves phenotypically, and thus become eliminated through adverse selection. Furthermore, the process known as random fixation takes place, so that purely by the vagaries of chance each population becomes uniform for a series of non-adaptive characteristics which in the larger popula- tion varied from individual to individual. This, of course, leads to the divergence of the isolated populations. It explains the fact that depleted species are usually sharply defined, that is morphologically very distinct from their nearest relatives, as well as being relatively uniform. In addition to the depleted species, there is also a large body of rare species which have always been so because they have never had an opportunity to spread. Since such a condition is most characteristic of islands, rare species of this type can be termed insular species (Kinsey, 25). If through some accident a small group of individuals of a continental species becomes es- tablished upon an island, they will carry with them only a small part of the genetic variability of the original species. Further- more, inbreeding and random fixation will tend further to make this insular population more uniform and more different from its continental ancestor as the years of its isolation progress. Thus the genetic structure of a restricted insular species becomes homogenic as does that of a depleted one. There is, however, one way in which insular populations can maintain a certain degree of variability. If the insular areas are near enough to the continental ones or to other islands so that the migration of individuals to the island can occur repeatedly, the insular population can periodically be enriched with a new infusion of genetic variability. It becomes the semi-isolated population which, according to Wright (37), has the best po- tentialities for evolutionary progress. If the insular area or areas are small, the species will remain rare, but it will have an un- expected amount of variability. Thus a rare species confined to several small, insular areas partly isolated from each other is an exception to the hypothesis stated above of genetic uniformity for rare species. It has a potential aggressiveness, but cannot spread because it has no place to go. Insular species are most easily recognized when they occur on actual islands, but they also exist within continental floras. Any species which occurs in a small area of favorable territory sur- rounded by extensive areas which neither it nor any of its close relatives could possibly occupy is as isolated as if it were on an island (cf. Kinsey, 25). This is true of the species of isolated 252 MADRONO [ Vol. 6 mountain tops which contain an alpine flora but are surrounded by great stretches of temperate or tropical lowland; of those found in oases in a desert, whether the oases are associated with streams, springs, or isolated mountain ranges, and of those in many types of habitats which are radically different from their surroundings. It is on one of these terrestrial “islands” that there occurs the example best known genetically of a plant species which, though rare, has an unexpected amount of variability due to its ex- istence in a series of semi-isolated colonies. This is Oenothera organensis Munz (O. macrosiphon Wooton & Standley) endemic to the Organ Mountains of New Mexico, which occurs in a series of small colonies along the only living streams found in this arid range of mountains completely surrounded by desert. It is a meso- phyte living in the only mesophytic habitats available to it. Emer- son (13) found that O. organensis has an unexpectedly high number of genes for self-incompatibility. Wright (36) on the basis of his mathematical deductions, could explain this situation only by assuming that the total number of about five hundred indi- viduals found in the species was divided into a series of small, semi-isolated colonies, an assumption fully warranted by its distri- bution. Oenothera organensis may have a good deal of potential aggressiveness, which might result in a spreading of the species if an increasingly moist climate should open up new habitats to it. In discussing the flora and fauna of actual islands, the dis- tinction is often made between continental islands, which were formerly connected with some large land mass, and oceanic islands, which have never been so connected (Baur, 4). The flora and fauna of the former are said to be harmonious, since they are derived entirely from one continental area, while those of the latter are termed disharmonious, being derived from two or more different continental areas, and by several different mi- grations from each area at widely separated intervals. Terres- trial insular areas may be similarly classified as to their origin. Many, such as most of the alpine regions in mountain ranges of the north temperate zone, were once connected with extensive continental areas of similar ecological conditions; others were never so connected. The latter nearly always provide striking cases of endemism. One such area in the eastern United States is the famous series of shale barrens in the Appalachian Mountains, extending from southern Pennsylvania to southwestern Virginia and eastern Ten- nessee. These barrens occur wherever rocks of certain geologi- cal formations outcrop on steep slopes. They are mildly arid as well as poor in mineral matter, so that they support a flora more xerophytic than that in the surrounding hills (Core, 11). Al though each slope is obviously a pioneer habitat, destined to dis- appear as soil accumulates on it, new barrens are constantly being created by weathering and stream erosion, so that the shale barren habitat has probably existed continuously ever since the 1942] STEBBINS: GENETICS AND RARE SPECIES 253 uplift of the Appalachians began early in the Tertiary period, and will continue to exist as long as these mountains stand. They thus represent, like the seashore, a “pioneer” habitat of permanent duration, at least so far as present-day species are concerned. Although the climate of the Appalachian region may at times have been drier than it is now (Gleason, 19, Core, 10), there is no reason to believe that it was ever arid or even semi-arid, so that the possibility that the shale barrens were ever joined to the large semi-arid areas in the central and western United States by a continuous stretch of territory similar to the present barrens is rather remote. Hence they represent semi-xerophytic islands in a region dominated by a mesophytic forest. The flora of these shale barrens, as analyzed by Wherry (32, 33), contains species of rather diverse affinities. Some like Senecio antennarifolius Brit- ton, are closely related to xerophytic western species. Others, like Oenothera argillicola Mackenzie, show certain characteristics in common with western species, but no close relationship. Cleland (9) has pointed out that O. argillicola resembles the western O. Hookeri alliance in its large flowers, self-incompati- bility, and in forming pairs of chromosomes rather than rings at meiosis. In other morphological characteristics, however, it is more like some of the eastern species, and the arrangement of its chromosome segments is somewhat different from that of any other species. Still other shale-barren species, like Pseudotaenidia montana Mackenzie, are of very obscure affinities. Finally there is a series of shale barren species which have obviously evolved from mesophytes of the surrounding flora. Some of them are specifically different from their mesophytic relatives, others are apparently only ecotypes, while still others are of doubtful status. In view of this diversity of affinities, the hypothesis of Wherry (32, 33) that all of the shale barren plants originated somewhere to the northwest, seems unlikely. The plant associ- ation of these barrens appears rather to have been gradually built up over a long period of time through the addition at widely separated intervals of plants derived from very different sources. In the west, such permanently isolated areas are more com- mon. The mountain ranges of the southern Great Basin are excellent examples. They were uplifted during the latter part of the Tertiary period (Louderback, 26), and it is very unlikely that the forest and alpine areas of their higher slopes were ever continuous with those of other mountains. The best known of them floristically is the Charleston Range of southern Nevada. The enthusiastic and thorough explorations of Mr. Ira W. Clokey have uncovered a large number of endemic species of diverse affinities, and the high montane flora as a whole differs from all others in the world. Another series of examples on a much smaller scale are the serpentine barrens of central California. Species of certain genera, like Streptanthus, will grow in this region only on serpen- 254 MADRONO [ Vol. 6 tine, so that these barrens for them are and probably have always been islands. Some of these Streptanthus species are endemic to only one or two barrens (Morrison, 27). These have probably always been rare, and may have been derived rather recently from the more widespread species by the establishment and di- vergence of an insular population as described above. So far as the writer is aware, no study of the geographic affinities of the serpentine barren endemics has been attempted; on the basis of the present discussion, these affinities should be diverse. Not all species now endemic to islands are insular in the sense that their present population has been derived from a few indi- viduals of an existing continental species. Relict, depleted spe- cies have been preserved on many islands, as well as on terrestrial insular areas (Baur, 4). This is particularly true of islands of continental derivation, and may be due to the lower intensity of competition as compared with continental areas. The Channel Islands off the coast of southern California, for instance, have preserved the last remnants of such species as Lyonothamnus flori- bundus Gray, Prunus Lyoni (Eastw.) Sarg. and Quercus tomentella Engelm., which are known through fossil evidence to have been formerly more widespread, and very likely consisted of several ecotypes (Axelrod, 2, 3). Even on oceanic islands or those with remote continental connections there are often found species which are apparently either the last relics or the immediate deriva- tives of ancient genera now extinct elsewhere. This is well illus- trated by some of the arboreal Compositae of the Pacific Basin. Two of the archipelagoes west of South America, namely Juan Fernandez and the islands of San Felix and San Ambrosio contain endemic genera of the tribe Cichorieae; Dendroseris on Juan Fernandez and Thamnoseris on San Felix and San Ambrosio, which are related neither to each other nor to any other genus found in the Southern Hemisphere. TYhamnoseris appears to be nearest to Stephanomeria of western North America, particularly S. Blairu Munz & Johnston, an anomalous endemic of San Clemente, one of the Channel Islands. Dendroseris is of more obscure affinities but is also most nearly related to North American genera. The modern species of both Thamnoseris and Dendroseris therefore, must have had more widespread ancestors, and are to be con- sidered as relict, depleted genera. Another case is Hesperoman- nia, endemic to Hawaii, where it is very rare. Its nearest rela- tives are Augusta (Stifftia) of Brazil and Nouelia of southwestern China. All three are apparently the last remnants of a group which must have been widespread in the Northern Hemisphere in Cretaceous or early Tertiary time (Stebbins 30). fFuitchia, en- demic to two islands of Polynesia, is an extraordinary genus which combines the characteristics of the tribes Mutiseae, Heliantheae, and Cichorieae, but has no close relatives in any of them. It is obviously a relict genus, perhaps a survivor of an ancient stock of 1942] STEBBINS: GENETICS AND RARE SPECIES 255 Compositae which existed before the present tribes became differentiated from each other. Since both insular and depleted species may occur on insular areas, terrestrial as well as actual islands, the differentiation be- tween the two types in such areas is a difficult problem. No set rule will hold for all species, but two criteria can be considered as valuable. First, if the endemic is closely related to a widespread species which occurs on an adjacent continental area, it is probably a strictly insular species, while if it is closely related to no other living form, or has its relatives in some remote corner of the globe, it is more likely a depleted species or a derivative of one. Second, if the endemic is morphologically a highly specialized type in relation to its continental relatives, it is probably an insular descendant of these; while if it is less specialized it may be their depleted ancestor. Finally we must consider the future of these homogenic rare species. The most obvious fact is that they are more at the mercy of climatic changes than are the common, variable species. Just as their genetic rigidity prevents them from occupying new habi- tats, so it must also reduce their power to adapt themselves to climatic changes. Hence if the climate remains the same they may persist as rare species indefinitely. If it becomes more un- favorable to them, they are likely to disappear. This is the fate awaiting many Californian trees and shrubs, notably Pinus Torrey- ana, P. radiata, Abies venusta and Cupressus macrocarpa, if the pro- gressive desiccation of our climate continues. On the other hand, if conditions become more favorable for the spread of their few remaining biotypes, they may become more common. Then, if two or more isolated colonies of a depleted species are enabled to spread until they meet, a partial replenishment of the store of genetic variability may take place. During their isolation the disjunct colonies must have acquired some different genetic characteristics, both by mutation and by random fixation (Wright, 37, Dobzhansky, 12). Thus when two such colonies reunite, hybridization between genetically different individuals is made possible. By this means, a large number of new genetic combina- tions may arise, replenishing the store of genetic variability, and opening up new possibilities for the spread of the species. Examples of species that have apparently been replenished in this fashion are provided by relatively widespread and common members of mono- or ditypic genera which occur in the same region and appear to be somewhat related to each other, but are very sharply set off morphologically and have no close interrela- tionships. Such a group is found among the Compositae, tribe Cichorieae of the Sonoran desert of western North America. Here there are two monotypic genera, Anisocoma and Atrichoseris, and three ditypic ones, Rafinesquia, Calycoseris, and Glyptopleura. These are all related to each other and to the larger genera Stephanomeria and Malacothriz. Together with Pinaropappus of 256 MADRONO [Vol. 6 Texas and Mexico as well as the insular Thamnoseris mentioned above, they form a natural group, not closely related to any other Cichorieae. The small size and remarkable distinctness of the genera and most of the species of this group is in sharp contrast to the situation in the Cichorieae of the Old World, where most of the genera are relatively large and are so closely interrelated that generic boundaries are very difficult to define. Supposing, however, one were to select a dozen species from each of the large Old World genera Lactuca and Crepis, and one or two each of Hypochaeris, Leontodon, Sonchus, Launea, Izxeris, and Youngia, choosing the more xerophytic members of each genus and should then deposit these species upon a semi-arid or desert area com- pletely devoid of Cichorieae, and they all became established in this area, the resultant pattern of variation would closely simulate that now found in the western American Cichorieae mentioned above. This suggests the hypothesis that exactly such a process of selection has taken place in the history of the latter. Their history may have been somewhat as follows. Once the entire group consisted of a few fairly large closely interrelated genera, or of one genus divided into several sections. Then the group became much reduced in numbers, due to the reduction in extent of the habitats which it occupied, and was broken up into many partly or completely isolated populations. The larger of these retained their store of genetic variability throughout the period of reduction. The smaller ones, being subject to intense selection and random fixation diverged sharply from the ancestral stock, and at the same time became much depleted genetically. Then with the restoration of conditions favorable to the group, all of the remaining species were able to spread again. Those less completely isolated then gave rise to the more closely inter- related species of the larger genera Stephanomeria and Malacothriz, while the descendants of the strongly isolated and depleted small populations became the distinctive mono- and ditypic genera. This history may be greatly oversimplified ; perhaps several cycles of depletion and replenishment were necessary to produce the pattern of variation found in this group. Nevertheless, the oc- currence of such cycles in the evolutionary history of this and other groups is a very likely possibility, and may have been of considerable importance in the differentiation of species and genera throughout the plant kingdom. The writer is much indebted to Dr. Daniel Axelrod, Dr. Lincoln Constance, Dr. Carl Epling, and Dr. Herbert L. Mason for helpful advice and criticism. SUMMARY The concept of age and area and that of senescence of species in the stricter sense is not considered adequate to explain the occurrence of rare and endemic species. Instead the writer re- gards as most important the concept of genetic homogeneity. 1942] STEBBINS: GENETICS AND RARE SPECIES 257 This assumes that most common and widespread species are genetically diverse, while rare and endemic ones contain rela- tively little genetic variability, that is relatively few biotypes. They are therefore termed homogenic. This homogeneity re- duces the number of ecological niches in which the rare species can compete successfully with other species, but if the climate is a stable one, does not necessarily cause their extinction. From the historical point of view there are two types of homogenic species. Depleted species are those which formerly were wide- spread and genetically diverse, but have lost many or most of their biotypes. A species may become depleted in only one part of its range, remaining common and variable in another. Insular species are those which have developed on an island or an isolated ecological habitat on a continent. They have originated from a few individuals or a single individual of the ancestral species, and have never possessed great genetic variability. The distinction between depleted and insular species is often hard to recognize, because depleted species often find their last refuges in insular areas. The future of rare, homogenic species depends upon the future of the ecological niches to which they are adapted. If the environment remains stable, they can persist indefinitely as rare species. If changes occur which obliterate their restricted habitats, they will become extinct. If, however, environmental changes result in an increase of the particular ecological condi- tions to which the species is adapted, it can spread. Then if this spreading permits the coming together of two isolated colonies of a homogenic species or of two such species which are still capable of interchanging genes, the populations thus united will both ac- quire new genetic variability. By this means a homogenic species may become diverse and widespread. If it is a depleted species, part or all of its original diversity may be restored. Division of Genetics, University of California, Berkeley, February, 1942. LITERATURE CITED 1. Anperson, E. The species problem in Iris. Ann. Mo. Bot. Gard. 23: 457- 509. 1936. AxeEtrop, D. A Miocene flora from the western border of the Mohave Desert. Publ. Carn. Inst. Wash. no. 516. 129 pp. 1939. . The concept of ecospecies in Tertiary paleobotany. Proc. Nat. Acad. Sci. 27: 545-551. 1941. Baur, G. On the origin of the Galapagos Islands. Am. Nat. 25: 217-229; 307-319. 1891. Cain, S. A. Some observations on the concept of species senescence. Ecol- ogy 21: 213-215. 1940. 6. Camp, W. H. Studies in the Ericales. A discussion of the genus Befaria in North America. Bull. Torrey Bot. Club 68: 100-111. 1941. 7. Cuaney, R. W. The succession and distribution of Cenozoic floras around the northern Pacific Basin. Essays in Geobot. in honor of W. A. Setchell: 55-85. 1936. 8. CLaAuseN, J.. D. D. Keck, ann W. M. Hiesry. Experimental studies on the nature of species. I. The effect of environment on western North American plants. Carnegie Inst. Wash. Publ. no. 520. 452 pp. 1940. 258 MADRONO [Vol. 6 CLELAND, R. E. Species relationships in Onagra. Proc. Amer. Phil. Soc. 77: 477-542. 1937. . Corr, E. L. Plant migrations and vegetational history of the southern Appalachian region. Lilloa 3: 5-29. 1938. The shale barren flora of West Virginia. Proc. West Vir- ginia Acad. Sci. 14: 27-36. 1940. . DoszHansky, TH. Genetics and the origin of species. Revised ed. 446 pp. Columbia University Press. 1941. . Emerson, S. A preliminary survey of the Menorhens organensis popula- tion. Genetics 24: 524-537. 1939. . Fassett, N.C. Mass collections: Rubus odoratus and R. parviflorus. Ann. Mo. Bot. Gard. 28: 299-368. 1941. . Fernatp, M. L. The persistence of plants in unglaciated areas of boreal America. Mem. Am. Acad. Sci. 15: 239-342. 1925. The antiquity and dispersal of vascular plants. Quart. Rev. Biol. 1: 212-245. 1926. ——_———.. Some relationships of the flora of the northern hemisphere. Proc. Int. Congr. Pl. Sci. Ithaca 2: 1487-1507. 1929. Critical plants of the upper Great Lakes, region of Ontario and Michigan. Rhodora 37: 197-222. 1935. . Greason, H. A. The vegetational history of the Middle West. Ann. Assoc. Am. Geogr. 12: 39-85. 1923. . Goon, R. D’O. A theory of plant geography. New Phyt. 30: 149-171. 1931. . Grices, R. F. The ecology of rare plants. Bull. Torrey Bot. Club 67: 565-594. 1940. . Hiesey, W. M. Environmental influence and transplant experiments. Bot. Rev. 6: 181-203. 1940. . Huszs, Cart L. The course of evolution (review). Am. Nat. 76: 96— 101. 1942. . Hurren, E. Outline of the history of arctic and boreal biota during the Quaternary period. 168 pp. 1937. Kinsey, A.C. An evolutionary analysis of insular and continental species. Proc. Nat. Acad. Sci. 23: 5-11. 1937. . LovupersBack, G. F. Basin range structure in the Great Basin. Univ. Calif. Publ. Geol. Sci. 14(10) : 329-376. 1923. . Morrison, Jonn L. Studies in the genus Streptanthus Nutt. I. Two new species in the section Euclisia Nutt. Madrofio 4: 204-208. 1938. . Ravur, H. M. Botanical problems in boreal America. Bot. Rev. 7: 147- 248. 1941. . STEBBINS, G. L., JR. Apomixis in the angiosperms. Bot. Rev. 7: 507— 542. 1941. . Additional evidence for a holarctic dispersal of flowering plants in the Mesozoic era. Proc. 6th Pacific Sci. Congr. pp. 649-660. 1941. . Turesson, G. Contributions to the gen-ecology of glacial relics. Hereditas 9: 81-101. 1927. Wuerry, E. T. Plants of the Appalachian shale-barrens. Jour. Wash. Acad. Sci. 20: 43-52. 1930. Four shale-barren plants in Pennsylvania. Proc. Penn. Acad. Sci. 7: 161-164. 1933. . Wiis, J.C. Age and Area. 259 pp. Cambridge Univ. Press. 1922. . The course of evolution. 207 pp. Cambridge University Press. 1940. . Wricut, S. The distribution of self-sterility alleles in populations. Genetics 24: 538-552. 1939. The statistical consequences of Mendelian heredity in rela- tion to speciation. The New Systematics, J. Huxley, editor, pp. 161- 183. 1940. ————__-——.. The “age and era” concept extended. Ecology 22: 345- 347, 1941. 1942] ERICKSON: CLEMATIS HIRSUTISSIMA 259 THE TYPE OF CLEMATIS HIRSUTISSIMA Ratpeu O. Erickson Clematis hirsutissima was described by Pursh on the basis of material collected on the Lewis and Clark Expedition of 1804-— 1806. Pursh’s description is adequate and unmistakably refers to a Clematis, but in his comment on the species he points out that “it very much resembles [ Anemone Pulsatilla] in several respects,” and expresses the opinion that “all the division of Anemones with caudated seeds . . . belong to this genus [ Clematis], or at least to one separate from Anemone.” Later Pursh mistakenly applied his name C. hirsutissima to specimens of an Anemone collected by Nuttall (Jour. Acad. Nat. Sci. Phila. 5: 159. 1825). Pursh’s contemporaries were not willing to place the “Anemones with caudated seeds” in the genus Clematis, but rather assumed that his C. hirsutissima was an Anemone, and this was the traditional view until Meehan in 1896 discovered duplicates of most of the Lewis and Clark plants in Philadelphia. In the meantime, C. Douglasiu Hooker had been published with a long description and a plate, and this name was generally accepted for the western Clematis. Meehan (Proc. Acad. Nat. Sci. Phila. 50: 12-49. 1898) lists the specimen of C. hirsutissima under C. Douglasii with the nota- tion, “[ No label; a single flower, but well identifiable].”” Further on he lists another specimen with the comment, “‘[ Composite? Poor, sterile, and not placed; leaves opposite, much divided into narrow segments, very pubescent]. One of the most common _ plants of the plains of Columbia. May 27, 1806.” During a recent visit. to the herbarium of the Academy of Natural Sciences of Philadelphia, the writer found the “single flower’ of C. hirsutissima mounted on a sheet with the following note by Meehan, “Clematis Douglasii H. no ticket with the speci- men probably Kooskooskee. It has been collected on the Clear- water by Spalding.” Attached to this sheet with a paper clip was a second sheet bearing the specimen which Meehan had labelled, “Compositae ?’’, and a label in Pursh’s hand, “One of the most common plants of the plains of Columbia. May 27, 1806.” It was obvious at once that both were C. hirsutissima, and careful examination of the broken ends of the peduncle showed that they could be fitted together exactly. Credit for associating these two fragments apparently belongs to Rydberg and to Piper, since the “Compositae ?”’ sheet bore the annotations, “Clematis Douglasii? P. A. R. 1905,” and (apparently later) ‘certainly C. Douglasii C. V. P.” They have now been remounted on a single sheet. It is now possible not only to settle once and for all the status of Pursh’s name C. hirsutissima, but to establish the type locality more accurately. On May 27, 1806, Lewis and Clark were at “Camp Chopunnish,’ which was “in Shoshone Co., across the river from, and nearly opposite, . . . present Kamai or Kamiah, 260 MADRONO | [Vol. 6 in Nez Percés Co., Ida.’’ (Coues, Proc. Acad. Nat. Sci. Phila. 50: 293 & 306. 1898). From a study of the Lewis and Clark jour- nal for that day, it is possible to say only that the plant could not have been collected at any great distance from the camp. Appended is a portion of the pertinent synonymy: CLEMATIS HiIRSUTISSIMA Pursh, Fl. Am. Sept. 2: 385. 1814. Clematis Douglasii Hook. Fl. Bor. Am. 1:1, pl. 1. 1829. ANEMONE Notrartiiana DC. Syst. 1: 193. 1818. A. patens var. hirsutissima Hitchcock, Trans. Acad. Sci. St. Louis 5: 482. 1891. Pulsatilla hirsutissima Britt. Ann. N. Y. Acad. Sci. 6: 217. 1891. Anemone hirsutissima MacMillan, Metasp. Minn. Valley, 239. 1892. Missouri Botanical Garden, St. Louis, Missouri, April 6, 1942. FRANCIS RAMALEY Francis Ramaley,' for forty years head of the Department of Biology of the University of Colorado and professor emeritus since 1989, died June 10, 1942. He was bern in St. Paul, Minne- sota, November 16, 1870. The University of Minnesota granted him his bachelor’s and master’s degrees in 1895 and 1896. He served as instructor in botany there for three years and then came to the University of Colorado in 1898 as assistant professor of biology. The following year, after receiving the degree of doctor of philosophy from Minnesota, he became professor and head of the Department of Biology at Colorado; from this time until his retirement, the untiring devotion and wise guidance which he gave the department as well as his insistence upon high standards were factors largely responsible for its growth and high repu- tation. In 1904 Professor Ramaley made a trip around the world, spending several months in study at botanical gardens at Buiten- zorg, Java, and Peradenyia, Ceylon; he also visited the gardens at Tokyo, Japan. This year of travel and study stimulated his natural interest in economic botany and resulted in valuable col- lections for the University Herbarium and Museum. In addition to his heavy teaching load and administrative duties, Professor Ramaley served the University in many other ways. He was acting president in 1902, acting dean of the Col- lege of Pharmacy from 1917 to 1919, and acting dean of the Graduate School in 1929 and again from 1932 to 1934. Because of his sound judgment and clear insight, he was a valuable mem- ber of many important University committees. In line with his policy of encouraging high standards of scholarship, he aided in the organization of chapters of Phi Beta Kappa and Sigma Xi, while the University was still young. From the time of the establishment of the “University of Colorado Studies” in 1902, he was the editor, a position which he held until his death. 1 See frontispiece. 1942] FRANCIS RAMALEY 261 His unfailing interest in research which continued even after the beginning of his final illness, resulted in the publication of about ninety scientific and educational papers and several books. Characterized by accuracy and clarity, his writings are interest- ing and intelligible to the layman as well as to the scientist. His publications give evidence of his broad interests which included not only his special field of botany, but human heredity and hygiene as well. While his earlier botanical papers indicated an interest in seedling anatomy, his most outstanding work was in the field of plant ecology. As early as 1899 he introduced a course in ecology at the University of Colorado. In 1940 when he was honored by being selected to give the annual research lecture at the University, it was natural and fitting that he should select as his subject “The Growth of a Science” in which he traced the history of ecology from the time of Aristotle to the present. This address which was printed in the “University of Colorado Studies” is of interest to all ecologists. The University of Colorado Mountain Laboratory at Tolland, Colorado, was established by Professor Ramaley in 1909; for the following ten years while he served as director of this successful mountain laboratory, he published many ecological papers deal- ing with zonation, succession, and the distribution of plant com- munities in the montane and sub-alpine zones of this section of the Colorado Rockies. He first became interested in sand-hill vegetation while he was on the summer faculty of the University of California in 1917. After he returned home he began to study similar areas in Colorado. As a result of extensive field studies pursued over a number of years, a paper entitled “Sand-hill Vege- tation of Northeastern Colorado” was published in ‘Ecological Monographs” in 1939. His last work, “Vegetation of the San Luis Valley in Southern Colorado,” which was published in March of this year, is an extensive and important contribution based on numerous field trips taken to this interesting region at different seasons during the past fourteen years. His interest in physio- logical ecology was indicated by experimental work involving over one hundred species of plants which were supplied with sup- plemental artificial light. Two published bibliographies of day length and artificial illumination as affecting growth of seed plants and three other papers constitute his contributions to this field. “Plants Useful to Man” (with W. W. Robbins) and “‘Colo- rado Plant Life’ are perhaps his best-known books. In the latter, which was written for the people of the state and pub- lished as part of the celebration of the Semicentennial of the Uni- versity in 1927, he presents a study of the native trees, grasses and flowering herbs. Sections of this book indicate to some extent his deep appreciation of the beauties of the world of nature which, as he tells us in the preface, he learned to know and enjoy when as a child and youth he accompanied his father through woodland and over prairie in his native Minnesota. 262 MADRONO [ Vol. 6 The ninety-eight titles listed in the bibliography do not include numerous book reviews which appeared from time to time in “Ecology,” “Science,’ “Torreya,” “Botanical Gazette,’ and “American Journal of Botany’’; articles in the “Biology News- Letter of the University of Colorado”; nor many abstracts which were published in the “Journal of the Colorado-Wyoming Acad- emy of Science.’’ During the past five years he learned to read Italian and since 1939 he has prepared approximately one hun- dred abstracts of articles from Italian botanical journals for publication in “Biological Abstracts.” A rare faculty for presenting material in a clear, forceful manner as well as the ability to summarize and emphasize impor- tant principles combined to make Professor Ramaley an outstand- ing teacher and lecturer. His interest in teaching is indicated by some ten publications dealing with the teaching of science or with some other phase of education. He believed that under- graduates should be given a liberal training without too much stress on their major field. As a member of the Council on Honors at the University and as its chairman at the time of his retirement, he urged students not to neglect broad fundamentals. He believed “‘that the specialist with a wide background of scien- tific knowledge will achieve the most.’’ His own broad knowl- edge of the biological sciences and related fields enhanced the value of his instruction. While he expected his students to main- tain high standards of accomplishment, his dealings with them were characterized by a kindly spirit of helpfulness, genuine interest in their progress and an unusual sense of fairness. Graduate students particularly prized his constructive criticism, his inspiring standards of thoroughness and accuracy, his unfail- ing patience and the generous giving of his time to their prob- lems. They, as well as his colleagues, valued him as a counselor and a friend and admired his progressive and tolerant spirit, his unassuming modesty, and his thoughtful consideration of others. Professor Ramaley was an active member of many scientific societies including the Botanical Society of America, American Society of Naturalists, Limnological Society, Society for Experi- mental Biology and Medicine, Colorado-Wyoming Academy of Science, and American Association for the Advancement of Sci- ence, serving as president of the Southwestern Division in 1930. He was a charter member of the Ecological Society of America, a group which elected him vice-president in 1931, and nine years later, president and botanical editor of ““Ecology.”’ In Professor Ramaley were combined the qualities of an emi- nent scientist with those of an outstanding teacher and executive; above all, he was a tolerant and understanding friend. He is sur- vived by his wife, Ethel Jackson Ramaley, and by four sons— Edward J., David, John D., and Francis——Epna Louise JoHNson, University of Colorado. 1942] 1898. 1899. 1900. 1901. 1902. 1903. 1904. 1905. 1906. 1907. 1908. FRANCIS RAMALEY 263 BIBLIOGRAPHY Modern botany in secondary schools. Colo. School Jour. 13: 216-7. Notes on plant ecology. Colo. School Jour. 14: 60-1. Plant societies. Colo. School Jour. 14: 9-10. Comparative anatomy of hypocotyl and epicotyl in woody plants. Min- nesota Bot. Studies 2: 87-136. Seedlings of certain woody plants. Minn. Bot. Studies 2: 70-86. Teaching of human physiology. Colo. School Jour. 15: 56-7. Seed and seedlings of the western larkspur (Delphinium occidentale Wats.) Minn. Bot. Studies 2: 417-21. Remarks on the distribution of plants in Colorado east of the divide. Postelsia, Yearbook of the Minnesota Seaside Station, pp. 21-53. Mesa vegetation. Science 15: 455. Sex in seed plants. Science 15: 996. Trichome structures of Erodium cicutarium. Bot. Gaz. 34: 140-2. Cotyledons and leaves of certain Papilionaceae. Univ. Colo. Studies 1: 239-43. Observations on Egregia menziesii. Minn. Bot. Studies 3: 1-9. Pubescence of species of Astragalus. Torreya 3: 38—40. Anatomy of cotyledons. Bot. Gaz. 37: 388-9. Some thought on college entrance requirements. Education 24: 277-80. Botanical Garden at Buitenzorg, Java. Pop. Sci. Monthly 67: 579-89. Botanists’ trip to Java. Plant World 8: 139-50. Later day histology. Univ. Colo. Medical Bull. 2: 25-7. Scholarly mind. Amer. Education 9: 203-4. Study of certain foliaceous cotyledons. Univ. Colo. Studies 2: 255-64. Teaching of botany and zoology. Univ. of Colo. Investigations of Dept. of Psych. and Educ. 2: 19-20. Mental independence. Univ. of Colo. Investigations of Dept. of Psych. and Educ. 3: 58-9. Plants of the Florissant region in Colorado. Univ. Colo. Studies 3: 177— 85. Seed and seedling of the mountain globe flower. Univ. Colo. Studies 3: 93-5. Tokyo Botanical Garden. Plant World 10: 251-8. Account of (botany) collections made on scientific expedition to north- eastern Colorado. Univ. Colo. Studies 4: 161-5. Plant zones in the Rocky Mountains of Colorado. Science 26: 642-3. Silva of Colorado. Part I. Trees of the pine family in Colorado. Univ. Colo. Studies 4: 109-22. Part II. Poplars, aspens, and cottonwoods. Univ. Colo. Studies 4: 187-97. Part III. Woody plants of Boulder County. Univ. Colo. Studies 5: 47-63. Botanical Gardens of Ceylon. Pop. Sci. Monthly 73: 193-206. Botanical opportunity in Colorado. Univ. Colo. Studies 6: 5-10. Botany of northeastern Larimer County, Colorado. Univ. Colo. Studies 5: 119-31. Color variation in some Colorado flowers. Plant World 11: 17-8. New Colorado species of Crataegus. Bot. Gaz. 46: 381-4. Studies in mesa and foothill vegetation. Part II. Climatology of the mesa near Boulder. Univ. Colo. Studies 6: 19-31. Two imperfectly known species of Crataegus (with G. S. Doods). Bull. Torr. Bot. Club 35: 581-3. Ecological notes from north-central Colorado (with W. W. Robbins). Univ. Colo. Studies 5: 111-7. Plant zones of the mountain lakes in northern Colorado (with W. W. Rob- bins). Science 27: 208. Rock ridge vegetation of northern Colorado (with W. W. Robbins). Science 27: 208-9. 264 1909. 1910. 1911. 1912. 1913. 1914. 1915. 1916. 1917. 1918. 1919. 1920. 1922. 1923. MADRONO [Vol. 6 College courses in preparation for life. Colo. School Jour. 24: 420-2. Educational significance of Minot’s Theory of Age and Growth. Educa- tional Review 38: 282-7. European plants growing without cultivation in Colorado. Ann. du Jardin Botanique de Buitenzorg, Series 2, Suppl. 3, pp. 493-504. Geology and natural history of Colorado. Colo. School Jour. 24: 351-5. University of Colorado Mountain Laboratory. Univ. Colo. Studies 7: 91-5. Wild flowers and trees of Colorado. 78 p. Boulder. Greenman. Silva of Colorado: Part IV. Forest formations and forest trees. Univ. Colo. Studies 6: 249-81. Studies in lake and streamside vegetation. I. Red Rock Lake near Ward, Colorado (with W.W. Robbins). Univ. Colo. Studies 6: 133- 68. Summer laboratory for mountain botany (with W. W. Robbins). Plant World 12: 105-10. Remarks on some northern Colorado plant communities, with special reference to Boulder Park (Tolland, Colorado). Univ. Colo. Studies 7: 223-36. Ecological cross-section of Boulder Park (Tolland, Colorado) (with Louis Mitchell). Univ. Colo. Studies 8: 277-87. Field observations on the so-called “Anemone” (with Marie Gill). Univ. Colo. Studies 8: 289-93. Tuberculosis as an economic and sociologic factor. Univ. Colo. Studies 8: 181-98. Grass-flora of Tolland, Colorado, and vicinity (with Esther Elder). Univ. Colo. Studies 9: 121-41. Mendelian proportions and the increase of recessives. Amer. Nat. 46: 344-51. What is biology and what is a biological survey? Science 35: 60-1. Prevention and control of disease (with C. E. Giffin). 386 p. Boulder. Inheritance of left-handedness. Amer. Nat. 47: 730-38. Insanity, its nature, causes and prevention. Univ. Colo. Bull. Vol. 13, No. 11. The amount of bare ground in some mountain grasslands. Bot. Gaz. 57: 526-528. The relative importance of different species in a mountain grassland. Bot. Gaz. 60: 154-157. Dry grassland of a high mountain park in northern Colorado. Plant World 19: 249-270. Quadrat studies in a mountain grassland. Bot. Gaz. 62: 70-74. Vascular plants of the Tolland region in Colorado. Univ. Colo. Studies 12: 27-51. Notes on dune vegetation at San Francisco, California. Plant World 21: 191-201. Xerophytic grasslands at different altitudes in Colorado. Bull. Torr. Bot. Club 46: 37-52. The role of sedges in some Colorado plant communities. Amer. Jour. Bot. 6: 120-130. Some mountain plant communities of sandy soil. Plant World 22: 313- 328. Vegetation of undrained depressions on the Sacramento plains. Bot. Gaz. 68: 380-387. Sub-alpine lake-shore vegetation in north-central Colorado. Amer. Jour. Bot. 7: 57-74. College Zoology Outlines. Pamphlet. Revised editions, 1924, 1935, 1936. Laboratory Manual of College Botany. Pamphlet. Revised editions, 1924, 1927, 1931. Outlines of Economic Botany. Pamphlet. Revised editions, 1926, 1930. Check-list of the plants of University Camp area in Boulder County, Colorado. Pamphlet of 11 pages. 1942] REVIEW 265 1925. Survey of the Plant Kingdom. Pamphlet. Revised editions, 1929, 1936. 1926. Colorado (with W. W. Robbins). Naturalist’s Guide to the Americas. Baltimore. 1927. Colorado Plant Life. 284 pages. University of Colorado. 1929. Botany of San Luis Valley in Colorado. Univ. Colo. Studies 17: 27-44. 1930. Specialization in science. Science 72: 325-326. 1931. Growth of plants under continuous light. Science 73: 566-7. Some caryophyllaceous plants influenced in growth and structure by artificial illumination supplemental to daylight. Bot. Gaz. 92: 311- 320. Vegetation of chaparral-covered foothills southwest of Denver, Colo- rado. Univ. Colo. Studies 18: 231-237. Autumn vegetation of the foothills near Boulder, Colorado (with Leon Kelso). Univ. Colo. Studies 18: 239-255. 1933. American Botany, 1886-1932, as shown in the Botanical Gazette. Science 78: 365. Plants Useful to Man (with W. W. Robbins). 428 pages. Second edi- tion, 1937. Philadelphia. A working bibliography of day-length and artificial illumination as af- fecting growth of seed plants. Univ. Colo. Studies 20: 257-263. 1934. Influence of supplemental light on blooming. Bot. Gaz. 96: 165-174. 1936. Stem and leaf anatomy as influenced by supplemental light. Univ. Colo. Studies 23: 245-250. 1937. The honors system at the University of Colorado. School and Society 45: 480-482. The recrudescence of a confusing terminology. Science 86: 36. A working bibliography of day-length and artificial illumination as af- fecting growth of seed plants, supplement. Univ. Colo. Studies 24: 121-126. Plant Science Manual. Pamphlet of 58 pages. 1939. Sand-hill vegetation of northeastern Colorado. Ecolog. Monographs 9: 1-51. Present-day botany in Italy. Science 90: 81. 1940. Growth of a science. Univ. Colo. Studies 26: 3-14. Control of prickly pear in Australia. Science 92: 528-529. 1942. Vegetation of the San Luis Valley in southern Colorado. Univ. Colo. Studies, series D, 1: 231-277. REVIEW The Flowering Plants and Ferns of Arizona. By Tuomas H. KEARNEY AND Rosert H. Peesues. United States Government Printing Office, Washington, D.C. 1942. $2.00. The flora of Arizona was studied with great interest and effec- tiveness by Asa Gray, John Torrey, Sereno Watson, and George Engelmann, and the half-century- and century-old papers and reports of this group of great systematic botanists have been the most useful works for general identification of plants from all but certain segments of the State. “The Flowering Plants and Ferns of Arizona’”’ fills a demand of long standing for an up-to-date, comprehensive study of the flora of Arizona as a unit. The book is based upon a sound piece of research, and it is particularly valuable for inclusion and evalu- ation of the numerous papers on special groups published prior to the time the book went to press in 1940 and for its references to these papers. This manual should serve as the foundation and 266 MADRONO [Vol. 6 the stimulus for further study of a flora of unusual interest in- vestigated so far only by travellers and a handful of resident botanists and by no means thoroughly known. The long experi- ence of the authors in Arizona makes their work carry unusual authority, and the combination of this field experience with care- fully considered organization of taxonomic units is a happy one. Treatment of particular families or genera by twenty-two special- ists adds much to the value of the book, although, as is inevitable in such cases, it introduces some variation in the weight accorded taxonomic categories, such as genera, species, and varieties. A section of the introduction entitled “The Vegetation of Arizona” by Forrest Shreve summarizes the results of long and intensive study of the Arizona flora by one of its most critical and thorough students. Valid criticisms of this book are few and minor. Division of the index into two parts, one for popular and one for scientific names may have some advantages, but there is a tendency to “land” in the wrong index, and a longer index including both of these would be scarcely more difficult to use for either lay or technical names. To one familiar with the excellent photog- raphy of the group at the U.S. Field Station at Sacaton and par- ticularly with the photographs taken by Mr. Peebles, there is disappointment in some of the illustrations, although others are excellent. It is probable that the weak contrasts in some of the half-tones are to be attributed to poor reproduction of the originals. Lack of descriptions of species is unfortunate but not to be criticized, since production of the first flora of a state is a task so great that it might have been impossible to accomplish if descrip- tions had been a part of it. A strong compensating factor is the thoroughness, fullness, and reliability of the keys, which are in excellent contrast to those of most of the descriptionless floras of the past. Concise statements of carefully selected characters of the genera contribute to the value of the work. The interpretation of species cannot be classified as either “liberal” or strongly “‘conservative”’ (in the botanical vernacu- lar), although it tends somewhat toward the latter. The un- usually unimportant rank assigned to varieties in the makeup of the book will be disappointing to some but probably pleasing to others.—Lyman Benson, Department of Botany, Agricultural Experiment Station, University of Arizona, Tucson. 1942] INDEX 267 INDEX TO VOLUME VI For classified articles see: Notices of Publication; Reviews. names are printed in bold-face type. Abies and Picea, A comparison of the embryogeny of, 156 Abies Pinsapo, 162, pl. 165; venusta, 159, 163, pl. 161 Abrams, L., New Limnanthes from Oregon, 27 Agropyron Saundersii, 87; Scribneri, 87 Alaska and Yukon, Arnica in, 153 Alaska cedar in California, 90 Alpine flora of San Francisco Moun- tain, Arizona, 65 Ammophila arenaria, pl. 169; Link, Anatomy and ecology of, 167 Amphioxus Tidestromii, 212 Anemone globosa var. lithophila, 133; hirsutissima, 265; lithophila, 133; Nuttalliana, 265; patens var. hir- sutissima, 265 Antennaria luzuloides var. oblanceo- lata, 136; oblanceolata, 136 Aplopappus carthamoides var. rigidus, 136; hirtus var. lanulosus, 136, var. sonchifolius, 136; Howellii, 136; racemosus var. brachycepha- lus, 136, var. congestus, 136, var. duriusculus, 136, var. glomeratus, 136, var. halophilus, 136; uni- florus var. Howellii, 136 Applegate, E. I., and Peck, M. E., New Frasera from Oregon, 12 Aquilegia flavescens, 133; formosa var. flavescens, 133 Archer, W. A., Field work of the Bureau of Plant Industry in Ne- vada, 94 Arenaria aberrans, 24; aculeata var. uintahensis, 133; capillaris, 24, subsp. formosa, 24; compacta, 24; Kastwoodiae var. adenophora, 24; Fendleri, 23, subsp. brevi- folia, 23, subsp. brevifolia var. brevicaulis, 23, subsp. genuina, 23, var. Porteri, 24; formosa, 24; uintahensis, 133 Arizona, A new species of Astragalus from, 18; San Francisco Moun- tain, Alpine flora of, 65 Arnica alpina subsp. angustifolia, 153, var. angustifolia, 153, subsp. attenuata, 153, subsp. tomen- tosa, 153; amplexicaulis, 154; amplexifolia, 154, subsp. genuina, 154, 155, var. borealis, 154, subsp. prima, 154, 155; angustifolia, New scientific 153; attenuata, 153; borealis, 154; brevifolia, 153; Chamissonis, 154, subsp. genuina, 154, var. in- terior, 154, var. typica, 154; cordifolia, 154, subsp. genuina, 154, var. pumila, 154; frigida, 153; illiamnae, 153; in Alaska and Yukon, 153; Lessingii subsp. genuina, 155, subsp. Norbergii, 155; louiseana subsp. frigida, 153, var. brevifolia, 153, var. genu- ina, 153, var. illiamnae, 153, var. Mendenhallii, 153, var. pilosa, 154; Mendenhallii, 153; nutans, 153; pulchella, 153; pumila, 154; Sancti-Laurenti, 153; tomentosa, 153 Artemisia campestris var. pacifica, 136, var. pycnocephala, 136, var. spithamaea, 136; pacifica, 136; spithamaea, 136 Aspidium Lemmonii, 224; mohrioides, 227 Astragalus acutirostris, 222; aequalis, 212, 215, 216, fig. 219; albens, 221; amphioxys, 212, 213, 214; arrectus var. remotus, 212, 217; artipes, 212, 216; Beathii, 18, pl. 19; bernardinus, 213, 220; Beck- withii var. purpureus, 212, 217; calycosus, 213, 221; coccineus, 212, 215; conjunctus var. Shel- doni, 135; didymocarpus var. di- spermus, 222; dispermus, 213, 222; Douglasii, 216, var. Parishii, 216; Forwoodii var. wallowensis, 135; Hoodianus, 134; Fremontii, 218; funereus, 214; hemigyrus, 213, 220, 221, fig. 219; Hookeri- anus var. siskiyouensis, 134; humi- stratus var. sonorae, 212, 215; kernensis, 218, subsp. charleston- ensis, 213, 218, fig. 219; lagopi- nus, 134; Layneae, 221; Laurentii, 134; lentiginosus, 218, var. cornu- tus, 135, var. Fremontii, 213, 218, var. platyphyllidium, 135, var. sierrae, 218; mancus, 213, 222; Marcusjonesii, 214; melanocalyx, 214; Minthorniae, 213, 220; mo- havensis, 212, 217; A new species of, from Arizona, 18; Newberryi, 212, 214, var. funereus, 212, 214; Notes on the flora of the Charles- ton Mountains, Clark County, 268 MADRONO Nevada. IV., 211; Nuttallianus var. acutirostris, 213, 222, var. imperfectus, 213, 222, var. tricho- carpus, 213, 222; Pattersoni var. praelongus, 217; platytropis, 213, 217; praelongus, 212, 217; Preusii, 212, 216; Purshii var. coccineus, 215, var. funereus, 214; reventus var. Hoodianus, 134; Sonorae, 215; subglaber, 134; Tidestromii, 214 Atriplex angustifolia var. obtusa, 133; patula var. obtusa, 133 Ball, C. R., Far western novelties in Salix, 227 Batiophaca sonorae, 215 Beetle, A. A., Certain North and South American distributions in Scirpus, 45 Benson, L., Review: Flowering plants and ferns of Arizona, 266 Bicuculla occidentalis, 133 Boykinia Jamesii var. heucheriformis, 173 Brachyphragma mohavensis, 217 Bradshaw, K. E., Field characters distinguishing Pinus ponderosa and Pinus Jeffreyi, 15 Brickellia oblongifolia subsp. lini- folia, 88, var. linifolia, 88, subsp. typica, 88, var. typica, 88 Brodiaea Douglasii var. Howellii, 133; grandiflora var. Howellii, 133; Howellii, 133 Buchholz, J. T., Comparison of the embryogeny of Picea and Abies, 156; Podocarpus gracilior in cul- tivation, 119 Calamaria Nuttallii, 8; Suksdorfii, 8 Calandrinia Howellii, 133 California, Alaska cedar in, 90 California Botanical Society, Pro- ceedings of, 64, 96, 175, 208, 239 California, Mount Hamilton Range, New species of Lotus from, 56; Problem of life zones on Mount Shasta, 49; Ribes petiolare Dougl. in, 30; Undescribed species of Ceanothus from, 171 Canotia holacantha, 129 Cantua parviflora, 200 Cardamine unijuga, 87 Carex Hepburnii, 87; occidentalis, 87 Carter, A., Review: Weeds of Cali- fornia, 174 Caryophyllaceae, Great Basin Plants— III., 22 Caulanthus hastatus, 134 [ Vol. 6 Cave, M. S., Development of the fe- male gametophyte in Erythro- nium helenae and Erythronium tuolumnense, 177 Ceanothus, An undescribed species of, from California, 171; gloriosus var. exaltatus, 171; Masonii, 171; purpureus, 173; ramulosus, 173; rigidus, 171, var. grandifolius, 172 Cephalanthera Austinae, 205; ore- gana, 205 Cerastium Beeringianum, 22, var. grandiflorum, 23; pulchellum, 23 Cercocarpus ledifolius var. hypo- leucus, 134 Chamaecyparis nootkatensis, 90 Chaparral, 199 Charleston Mountains, Clark County, Nevada, Notes on the flora of, IV., Astragalus, 211 Chloraea Austinae, 205 Chrysothamnus nauseosus subsp. glareosus, 88, subsp. leiospermus, 89; Parryi subsp. attenuatus, 89, subsp. Howardii, 89 Cicuta ampla, 150; arguta, 150; cali- fornica, 149; cinicola, 150; crassi- folia, 150; dakotica, 150, var. pseudomaculata, 150, var. pseudo- virosa, 150; Douglasii, 149, var. occidentalis, 150; fimbriata, 150; frondosa, 150; grandifolia, 150; Bolanderi, 149; bulbifera, 151; Curtissii, 149; mackenzieana, 150; maculata, 149, 150, var. Cur- tissii, 149; occidentalis, 150, f. ari- zonensis, 150, f. californica, 150, f. frondosa, 150, f. oregonensi- idahoensis, 150, f. wyomingensis, 150; purpurata, 150; Sonnei, 150; subfalcata, 150; Synopsis of the American species of, 145; vagans, 150; valida, 150; Victorinii, 150; virosa var. californica, 150, var. maculata, 150 Cicutaria bulbifera, 151 Clark County, Nevada, Notes on the flora of the Charleston Mountains, IV. Astragalus, 211 Clematis hirsutissima, 259; Type of, 259 Clokey, I. W., Notes on the flora of the Charleston Mountains, Clark County, Nevada, IV. Astragalus, 211 Cnidium Monnieri, 95 Constance, L., Reviews: Flora of Whatcom County, State of Wash- ington, 59; Manual of the higher plants of Oregon, 92 1942] Constance, L., Three alien plants new to Oregon, 95 Constance, L., and Mathias, M. E., Synopsis of the American species of Cicuta, 145 Cooke, W. B., Problem of life zones on Mount Shasta, California, 49 Copeland, E. B., Review: The evolu- tion of land plants, 142 Copeland, H. F., Further studies on Monotropoideae, 97 Core, E. L., New species of Parony- chia from Mexico, 21 Cotyledon glanduliferum, 134; oregon- ensis, 134 Crepis atribarba var. originalis, 137; Bakeri var. Cusickii, 137; Cu- sickii 137; modocensis var. sub- acaulis, 137; occidentalis var. pumila, 137, var. subacaulis, 137; pumila, 137; runcinata_ subsp. imbricata, 136, var. imbricata, 136 Cronemiller, F. P., Chaparral, 199 Cronquist, A., Noteworthy plants from Idaho, 87 Crum, E., Reviews: Forests and trees of the Western National Parks, 92; Ornamental shrubs and woody vines of the Pacific Coast, 174; Plant hunters in the Andes, 174 Cystium cornutum, 135; Fremontii, 220; platyphyllidium, 135; platy- trope, 217 Dasystephana monticola, 151; obtusi- loba, 151 Delphinium glaucum, 85; scopulorum glaucum, 85; splendens, 84 Descurainia pinnata var. filipes, 133, var. halictorum, 133, var. Nel- sonii, 134, var. paradisa, 134; Richardsoni var. viscosa, 133 Detling, L. E., Albert Raddin Sweet- ser, 20 Dicentra Cucullaria var. occidentalis, 133 Direa occidentalis, distribution map, 246; palustris, distribution map, 246 Doyel, B. E., and Goss, M. L., Some details of the reproductive struc- tures of Sarcodes, 1 Draba caroliniana f. stellifera, 133; reptans var. stellifera, 133 Eburophyton Austinae, 205 Eburophyton Heller: Valid genus of the Orchidaceae, 205 Endemic species, Genetic approach to problems of rare and, 241 INDEX 269 Epilobium cinerascens, 135; glandu- losum var. cinerascens, 135; Ham- mondi, 135; paniculatum var. Hammondi, 135 Erickson, L. C., Study of Isoetes in San Diego County, California, 7 Erickson, R. O., Type of Clematis hirsutissima, 259 Eriogonum ochroleucum, 133; ovali- folium var. ochroleucum, 133 Erythronium helenae and Erythro- nium tuolumnense, Development of the female gametophyte in, 177 Eucalyptus, Some chemical properties of, in relation to their evolution- ary Status, 181 Euphorbia Esula, 88 Frasera fastigiata, 12; New, from Oregon, 12; speciosa, 12; ump- quaensis, 12 Genetic approach to problems of rare and endemic species, 241 Gentiana barbellata, 151; calycosa, 151, subsp. asepala, 151, subsp. typica, 151; Notes on, Great Basin Plants, VI, 151; Parryi, 153 Geum gracilipes, 134 Gilia Clokeyi, pl. 201, 202; incon- spicua, 200; laciniata, 202; parvi- flora, 200; valdiviensis, 202 Goss, M. L., and Doyel, B. E., Some details of the reproductive struc- tures of Sarcodes, 1 Great Basin plants, III. Caryophyl- laceae, 22; VI. Notes on Genti- ana, 151 Grover, F., Mary Fisk Spencer, 82 Habenaria elegans var. multiflora, 133 Haines, L., Variation in Yucca Whip- plei, 33 Hamosa acutirostris, 222; austrina, 222; bernardina, 220; calycosa, 221; imperfecta, 222; manca, 222; Minthorniae, 220 Hemitomes congestum, 109, pls. 107, 111; pumilum, 109; spicatum, 110; subterraneum, 109 Hesperastragalus dispermus, 222 Holacantha Emoryi, 131; Stewartii, 131 Holacanthoid plants of North Amer- ica, 128; distribution of, 130 Homalobus Laurentii, 134; subglaber, 134 Homapappus glomeratus, 136 Horkelia capitata, 134; congesta subsp. nemorosa, 134, var. nemo- 270 MADRONO rosa, 134; fusca var. capitata, 134, var. filicoides, 134, var. parviflora, 134, var. pseudocapitata, 134; parviflora, 134; pseudocapitata, 134 Hypopithys Hypopithys, 100 Hypopitys Monotropa, 99, 103, pl. 105; uniflora, 100 100, pl. Idaho, Noteworthy plants from, 87 Ipomopsis inconspicua, 200 Isoetes Howellii, 8, 10; melanopoda var. californica, 8, 10; nuda, 8, 10; Nuttallii, 8, var. Orcuttii, 9; opaca, 8; Orcuttii, 8, 9; Study of, in San Diego County, California, 7; Suksdorfii, 8; Underwoodii, 8, 10 Johnson, E. L., Francis Ramaley, 260 Jones, G. N., New species of vascular plants from the Northwest Coast, 84 Jonesiella praelonga, 217 Keck, D. D., Review: Desert wild flowers, 175 Keraskomion bulbiferum, 151 Knopf, C. S., Significance of certain plant names, 209 Koeberlinia spinosa, 129, var. tenui- spina, 129, var. verniflora, 129 Lapsana apogonoides, 96 Larix occidentalis, 90 Lewisia Cotyledon var. Howellii, 133 Life Zones, Problem of, on Mount Shasta, California, 49 Limnanthes alba, 27; Bellingeriana, 27; Douglasii, 27; floccosa, 27; from Oregon, New, 27; gracilis, 27; Howelliana, 27; pumila, 27; rosea, 27 Linanthus Nuttallii var. floribundus, 204; Wigginsii, 203, 204, pl. 201 Little, E. L., Jr., Alpine flora of San Francisco Mountain, Arizona, 65 Lotus, New species of, from Mount Hamilton Range, California, 56; denticulatus, 58; humistratus, 58; rubriflorus, 56, 58, fig. 57, sub- pinnatus, 58 Love, R. M., and Stebbins, G. L., Jr., Undescribed species of Stipa from California, 137 Lychnis Drummondii, 26, var. nuda, 26; striata, 26 Macbride, J. F., Review: Sinopsis de la flora del Cuzco, 143 [ Vol. 6 Maguire, B., Arnica in Alaska and Yukon, 153; Great Basin plants, III, Caryophyllaceae, 22; Great Basin plants, VI, Notes on Genti- ana, 151; Range extensions in species of Western North Amer- ica, 173 Mason, H. L., Alaska Cedar in Cali- fornia, 90; Notes on Polemonia- ceae, 200; Taxonomic status of Microsteris Greene, 122 Mason, H. L., Reviews: Experimental studies on the nature of species, 1. Effect of varied environments of Western North American plants, 60; Ferns and fern allies of Wisconsin, 90; Flora of Ari- zona and New Mexico, 144; Standardized Plant Names, 206 ~Mathias, M., Review: Flora of Indi- ana, 29 Mathias, M. E., and Constance, L., Synopsis of the American species of Cicuta, 145 McMinn, H. E., An undescribed spe- cies of Ceanothus from Califor- nia, 171 McNair, J. B., Some chemical proper- ties of Eucalyptus in relation to their evolutionary status, 181 Menziesia ferruginea var. glabella, 135; glabella, 135 Messerschmidia sibirica, 95 Mexico, Grassland and related vege- tation in northern, 190; New spe- cies of Paronychia from, 21 Microsteris Greene, Taxonomic status of, 122 Mielke, J. L., Ribes petiolare Dougl. in California, 30 Monotropoideae, Further studies on, 97 Monotropa australis, 104; coccinea, 104; Hypopithys, 100; uniflora, 100, 104, pl. 107 Morrison, J. L., Review: Monograph of the genus Calochortus, 58 Mount Hamilton Range, California, New species of Lotus from, 56 Mount Shasta, California, Problem of life zones on, 49 Muller, C. H., Holocanthoid plants of North America, 128 Nevada, Notes on the flora of the Charleston Mountains, Clark County, IV. Astragalus, 211 Newberrya congesta, 109; longiloba, 109, 110; pumila, 109; spicata, 109; subterranea, 109 1942] Notes and News, 30, 63, 94, 144, 173, 239 Notices of Publication, 31, 32 Oligosporus pycnocephalus, 136 Oregon, New Frasera from, 12; New Limnanthes from, 27; New plants from, 13; Three alien plants new to, 95 Paronychia albomarginata, 21, 22; New species of, from Mexico, 21; Wilkinsonii, 22 Parrya cheiranthoides var. lanugi- nosa, 134; Menziesii var. lanugi- nosa, 134 Peck, M. E., New plants from Ore- gon, 13 Peck, M. E., and Applegate, E. LI. New Frasera from Oregon, 12 Penstemon Whitedii, 88 Phace artemisiarum, 217; artipes, 216; platytropis, 217; Preusii, 216; sis- kiyouensis, 134 Phaceomene artemisiarum, 217 Phlox canescens, 136; condensata var. Hendersonii, 135; diffusa subsp. subcarinata, 135, var. subcari- nata, 135, subsp. longistylis, 135, var. longistylis, 135, var. scle- ranthifolia, 136; divaricata var. occidentalis, 135; Douglasii var. Hendersonii, 135, var. rigida, 135; Drummondii var. tenuis, pl. 125; gracilis, pl. 125; Hoodii var. ca- nescens, 136; humilis, 135; lanceo- lata, 135; linearifolia var. lon- gipes, 135; longifolia subsp. calva, 135, var. calva, 135, var. com- pacta, 135, var. humilis, 135, var. longipes, 135; rigida, 135; scle- ranthifolia, 136; speciosa var. lanceolata, 135, var. occidentalis, 135; Stansburyi var. compacta, 135 Picea and Abies, Comparison of the embryogeny of, 156 Picea Omorika, 158; Smithiana, 156, pl. 157 Pinus Jeffreyi and Pinus ponderosa, Field characters distinguishing, 15 Piperia multiflora, 133 Podocarpus elongatus, 119; falcatus, 120; gracilior, fig. 121, in culti- vation, 119 Polemoniaceae, Notes on, 200 Polystichum Lemmoni, 223; mohri- oides, 227 Porter, C. L., New species of Astraga- lus from Arizona, 18 Potentilla dichroa, 134; Douglasii var. INDEX 271 filicoides, 134; gracilipes, gracilis var. dichroa, 134 Proceedings of the California Botani- cal Society, 64, 96, 175, 208, 239 Pterospora andromedea, 97, pl. 103 Pulsatilla hirsutissima, 265 Purer, E. A., Anatomy and ecology of Ammophila arenaria Link, 167 Pyrrocoma brachycephala, 136; con- gesta, 136; duriuscula, 136; halo- phila, 136; lanulosa, 136; rigida, 136; sonchifolia, 136 260; 134; Ramaley, Francis, frontispiece Reviews: Bailey, Forests and forest trees of the western national parks, 92; Benson, Cacti of Ari- zona, 28; Campbell, Evolution of land plants, 142; Clausen, Keck, Hiesey, Experimental studies on the nature of species. I. Effect of varied environments on west- ern North American plants, 60; Deam, Flora of Indiana, 29; Good- speed, Plant hunters in the An- des, 174; Graham and McMinn, Ornamental shrubs and woody vines of the Pacific Coast, 174; Herrera, Sinopsis de la flora del Cuzco, 143; Jaeger, Desert wild flowers, 175; Kearney and Peeb- les, Flowering plants and ferns of Arizona, 266; Kelsey and Day- ton, Standardized plant names, 206; Muenscher, Flora of What- com County, State of Washing- ton, 59; Ownbey, Monograph of the genus Calochortus, 58; Peck, Manual of the higher plants of Oregon, 92; Robbins, Bellue, Ball, Weeds of California, 174; Tide- strom and Kittell, Flora of Ari- zona and New Mexico, 144; Tryon, Fassett, Dunlop, Diemer, Ferns and fern allies of Wiscon- sin, 90 Ribes petiolare Dougl. in California, 30 Rosa californica var. ultramontana, 134; pisocarpa var. transmon- tana, 134, var. ultramontana, 134 portrait, St. John, H., Type locality of Poly- stichum Lemmoni Underwood, 223 Salix anglorum, 228; argophylla, 232, 233; balsamifera, 230; cordifolia, 228; exigua, 234, 237, var. Parish- iana, 236, subsp. virens, 232, 233; Far western novelties in, 227; Farrae subsp. Walpoleii, 230; Geyeriana, 84, subsp. meleina, 84, 272 MADRONO var. meleina, 84; glauca, 228; glaucophylla, 228; Hindsiana, 232, 233, var. leucodendroides, 232, 234, var. Parishiana, 236, 238; integrifolia subsp. leucoden- droides, 232, var. leucodendroides, 233; laevigata, 228; lasiandra, 228; longifolia, 232, subsp. ar- gyrophylla, 232; lutea, 228; macrostachya, 233, subsp. leuco- dendroides, 232; meleina, 84; ovalifolia, 228; Parishiana, 236, 237; phylicoides, 229; pulchra, 228, var. Looffiae, 228, var. Pal- meri, 229; pyrifolia, 230, 231; re- ticulata, 228; Richardsonii, 228; sessilifolia, 232, 233, subsp. leuco- dendroides, 232, var. leucoden- droides, 237; taxifolia, 237; Wal- poleii, 230, 231 San Diego County, California, Study of Isoetes in, 7 San Francisco Mountain, Arizona, fig. 66, pl. 69; Alpine flora of, 65 Sarcodes, Some details of the repro- ductive structures of, 1; san- guinea, 1, pls. 3, 5 Saxifraga eriophora, 173; oppositi- folia, 87 Scirpus, Certain North and South American distributions in, 45 Sedum glanduliferum. 134; nesio- ticum, 86; oregonense, 134; steno- petalum, 86 Senecio antennariifolius, distribution map, 247; canus, distribution map, 247; triangularis var. trigo- nophyllus, 136; trigonophyllus, 136 Serapias Austinae, 205 Sharsmith, H. K., New species of Lotus from the Mount Hamilton Range, California, 56 Shreve, F., Grassland and_ related vegetation in Northern Mexico, 190 Sidaleea maxima, 13; spicata, 14, var. tonsa, 14 Silene filisecta, 133; oregana var. fili- secta, 133; pectinata var. sub- nuda, 26; Petersonii, 24 Sisymbrium incisum var. filipes, 133 Sium Douglasii, 149 Sophia halictorum, 133; Nelsonii, 134; paradisa, 134; viscosa, 133 [ Vol. 6 Sophora Leachiana, 13 Spencer, Mary Fisk, 82, pl. facing 82 Stebbins, G. L., Jr., Genetic approach to problems of rare and endemic species, 241 Stebbins, G. L., Jr., and Love, R. M., Undescribed species of Stipa from California, 137 Stipa, Undescribed species of, from California, 137; cernua, 137, 141, fig. 138, pl. 139; comata, 141; pulchra, 137, 141, fig. 138, pl. 139 Streptanthus hastatus, 134 Sweetser, Albert Raddin, 20, pl. facing 20 Thamnosma montana, 129 Tium remotum, 217; Sheldoni, 135 Tofieldia intermedia, 133; occidentalis var. intermedia, 133 Tucker, J. M., Eriodictyon capitatum, 173 Viola atriplicifolia, 135, linguaefolia, 135; Nuttallii var. major, 135; praemorsa var. linguaefolia, 135, var. major, 135; purpurea var. atriplicifolia, 135 Wiggins, I. L., Review: Cacti of Ari- zona, 28 Williams, L. O., Eburophyton Heller, a valid genus of the Orchidaceae, 205 Xylophacos amphioxys, 213; cocci- neus, 215; funereus, 214; lago- pinus, 134; melanocalyx, 214; Newberryi, 214; Tidestromii, 214 Yucca graminifolia, 44; Whipplei, dis- tribution map, 34, fig. 37, subsp. caespitosa, 37, 40, 42, 43, distri- bution map, 34, fig. 35, var. caespitosa, 34, subsp. intermedia, 40, 42, 43, distribution map, 34, fig. 35, fig. 37, fig. 40, subsp. Parishii, 42, 44, figs. 37, 40, var. Parishii, 44, subsp. percursa, 43, distribution map, 34, figs. 35, 37, 40, subsp. typica, 42, 44, distribu- tion map, 34, figs. 37, 40; Vari- ation in, 33 Yukon and Alaska, Arnica in, 153 ERRATA Page 233, line 33: for sessillifolia read sessilifolia. Page 129, line 26: for montanum read montana. 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 i. - < re = ss ne soe $4 po <2 i i ee rie — Sere es OMA 3 9088 01111 7991